1
7th ASEM-Workshop
Advanced Electron Microscopy
April 20th – 21st, 2017
Venue: Technische Universität Wien TUtheSky, Getrteidemarkt 9, 1060 Wien
We cordially thank our sponsors:
2
Scope of the workshop
This workshop is a meeting for electron microscopists and all those interested in electron microscopy
and its applications in modern science and technology. Students and scientists from all fields of
microscopy in life sciences, materials science and physical science are welcome. The goal of this
workshop is to join student and expert scientists. It provides an Austrian-wide discussion forum for
the work done in academia and industrial based research. As a platform for oral presentations by
young scientists, this year it is especially oriented in view of the two big European conferences in
Lausanne and Rovinj, the MC2017 and the MCM2017, respectively.
The conference language is English.
Local Organizers
Prof. Johannes Bernardi
University Service Centre for Electron Microscopy (USTEM), Technische Universität Wien
Wiedner Hauptstraße 8-10, 1040 Vienna
(T) +43 (0) 1-58801-45210
(F) +43 (0) 1-58801-9-45210
Prof. Michael Stöger-Pollach
University Service Centre for Electron Microscopy (USTEM), Technische Universität Wien
Wiedner Hauptstraße 8-10, 1040 Vienna
(T) +43 (0) 1-58801-45204
(F) +43 (0) 1-58801-9-45204
3
TUtheSky
Getreidemarkt 9, 1060 Wien
Lageplan
Mit dem Aufzug in das 11. Stockwerk fahren!
4
Programme
Thursday, 20th April 2017
13:00 – 13:25 Registration
13:25 – 13:30 Opening
Session 1 – Life Science
13:30 Dietmar Pum (invited)
S-layer proteins
14:00 FEI – Firmenvortrag: Ben Lich
Cryo EM workflows for Single Particle Analysis and Tomography of hydrated, intact cells
14:15 Philipp Steiner
Stress induced fusion of mitochondria visualized by electron tomography in plants
14:30 Sabrina Oberwegser
Tsunamia transpacifica – TEM investigations in a newly-discovered red algal genus
colonizing Japanese tsunami debris
14:45 Margret Eckhard
How to preserve a moss for element analysis? Different ways of preparation for electron
microscopy
15:00 Daniel Serwas
How Cells Build Their Antenna: Centrioles Initiate Cilia Assembly, But Are Dispensable for
Cilia Maturation and Maintenance
15:15 David Kleindienst
Compartment-specific association of GABAB receptors and their effector ion channels in
cerebellar Purkinje cells
15:30 Gatan – Firmenvortrag: Andreas Kastenmüller
Recent technology improvements for Electron Microscopy
15:45 – 16:15 Coffee break
Session 2 – Materials Science
16:15 Fritz-Grasenik Preisvortrag: Markus Herbst Characterization of the Vasa vasorum in the human great saphenous vein by SEM and 3D-morphometry of vascular corrosion casts
16:45 Tia Truglas
The effects of double annealing on medium manganese steel
17:00 Tomasz Wojcik
Phase characterization in Ni-base superalloy Rene 65
17:15 Christian Ebner
Viscoelastic stress relaxation of TiAl thin film under tension measured by selected area
electron diffraction
5
17:30 JEOL – Firmenvortrag: Georg Raggl
The JIB-4700F - JEOLs new high performance FIB multi-beam system
17:45 Ulrich Haselmann
HRTEM study of Ca doped Bismuth Ferrite
18:00 Bernhard Bayer
Introducing overlapping grain boundaries in chemical vapor deposited hexagonal boron
nitride monolayer films
18:15 Stefan Pfeiffer
Combined analytical TEM and magnetic investigation of the effects of neutron irradiation
on Nb3Sn superconductors
19:00 – 21:30 Workshop dinner
Friday, 21st April 2017
Session 3 – Life Science
08:30 Zaoli Zhang (invited)
Advanced characterization of materials using atomic resolution TEM
09:00 Mariella Sele
High Resolution Visualisation of Iron Deposits in the Human Brain in Health and Disease
09:15 Christoph Dibiasi
Impact of fibrinogen concentration on blood clot formation
09:30 Leica – Firmenvortrag: Robert Ranner
Optimized sample preparation by using a correct workflow
09:45 Virginie Hubert
Using electron microscopy as a method to monitor autophagy
10:00 Stefan Schulz
Serum derived exosomes as a putative diagnostic tool for ANCA associated vasculitis
10:15 Carolina Borges-Merjane
Flash and Freeze: combining high-pressure freezing and optogenetics to evaluate
synaptic transmission
10:30 Jacek Plewka
SEM on agarose-based chromatographic beads – how to recalculate a reference SAXS
scattering signal from an image
10:45 – 11:15 Coffee break
6
Session 4 – Materials Science
11:15 Michal Horak
Babinet principle for plasmonic antennas: complementarity and differences
11:30 Franz-Philipp Schmidt
Hybrid plasmonics: From plasmon-plasmon to plasmon-exciton coupling
11:45 Wolfgang Wallisch
Influences of the CMR effect on dielectric properties
12:00 Semir Tulic
Reaction of Ni and C thin films studied by TEM and SEM
12:15 ZEISS – Firmenvortrag: Wolfgang Schwinger
Analytical FIB-SEM Tomography without Compromises
12:30 Robert Sriemitzer
Dealing with light refraction in 3D mapping in combined Raman/SEM
12:45 Philipp Siedlacek
SEM Characterization of functionalized Carbon Nanotubes
13:00 Robert Winkler
FEBID Based Direct-Write of 3D Plasmonic Gold Structures
13:15 Jürgen Sattelkow
Direct-Write Fabrication of Electric and Thermal High-Resolution Nanoprobes on Self-
Sensing AFM Cantilever
13:30 Snacks and Farewell
Poster:
Martin Meischel
Nanoscale studies of mechanical properties of rat bones around biodegradable implants
Stefan Löffler
Convergent-Beam EMCD: Efficient Magnetic Measurements on the Nanoscale
Thomas Schachinger
Vortex Filter EMCD: Towards an Alternative EMCD Approach
Christoffer Müller
Flash-annealed CuZr based bulk metallic glass studied by electron microscopy methods
Harald Fitzek
Understanding surface enhanced Raman spectroscopy using accurate simulations of
electric nearfields
Cornelia Trummer
Preparation of Transmission Electron Microscopy Samples by Mechanical Techniques in
Combination with Low-Voltage Ion Milling
7
Daniel Knez
In situ studies of high-purity mono- and bimetallic nanostructures in experiment and
simulation
Manfred Nachtnebel
Polymer fracture – What can the 3D reconstruction of the crack region tell us about the
microscopic fracture mechanisms
Martina Dienstleder
Challenges in sample preparation for HRSTEM analysis
Angelina Orthacker
Investigation of the non-equilibrium formation of stoichiometric precipitates in multi-
component aluminium alloys
Stefan Geyer
High resolution episcopic microscopy (HREM): a tool for 3D imaging of organic materials
Thomas Götsch
The Electronic Phase Diagram of YSZ
Stefan Noisternig
Lamellae in FeAl deformed under hydrostatic pressure
Katharina Keuenhof
Preparation Methods of Biological Samples: a Comparison of Chemical Fixation and
High-Pressure Freezing (HPF)
Eveline Fisselthaler
Quantitative Analysis of Internal Interfaces: Structural and quantitative analysis via High
resolution STEM
Walid Hetaba
In-situ electron microscopy for heterogeneous catalysis
8
9
S-layer proteins
Dietmar Pum(1), Uwe B. Sleytr (1)
(1) University of Natural Resources and Life Science, Vienna, Department of Nanobiotechnology, Institute of Biophysics, Muthgasse 11, 1190 Vienna, Austria
Crystalline bacterial cell surface layer (S-layer) proteins are one of the most abundant biopolymers on
earth and form the outermost cell envelope component in a broad range of archaea and bacteria
(Fig.1) [1, 2]. These S-layer protein lattices represent the simplest biological membranes developed
during evolution. S-layer lattices are highly porous protein mesh works with unit cell sizes in the range
of 3 to 30 nm and thicknesses of ∼10 nm. One of the key features of S-layer proteins is their intrinsic
capability to form self-assembled mono- or double layers in suspension, at solid supports, the air-water
interface, planar lipid films, liposomes, nanocapsules, and nanoparticles.
Basic research on S-layer proteins enabled us to use the unique self-assembly properties of native and,
in particular, genetically functionalized S-layer fusion protein lattices as matrices for the binding of
molecules and the synthesis of nano materials. In addition, most recently S-layer proteins were used
as scaffolds for making hybrid organic-inorganic nanostructures.
This contribution summarizes the state-of-the art in the reassembly of S-layer proteins, their non-
classical pathway of matrix assembly, and application as templates in the controlled deposition of
inorganic materials, such as biogenic silica.
Figure 1. (a) TEM micrograph of a freeze-etched and metal shadowed preparation of a bacterial cell of Lysinibacillus sphaericus with an S-layer (SbpA) as the outermost cell envelope component. The S-layer exhibits square (p4) lattice symmetry. The numerous lattice faults are a consequence of the bending of the S-layer lattice at the rounded cell poles. In addition, the rope-like structures are the flagella of the bacterial cell. Bar, 200 nm. (b) Atomic force microscopical image of a monolayer of SbpA S-layer proteins reassembled on a silicon surface. Image data were complemented with TEM tomography data. Unit cell size is 13.1 x 13.1 nm.
[1] Sleytr, U.B., Schuster, B., Egelseer, E.M., & Pum, D. (2014) FEMS Microbiol Rev, 38, 823-864.
[2] Pum, D., Sleytr, U.B. (2014) Nanotechnology, 25, 312001.
We kindly acknowledge financial support by the Air Force Office of Scientific Research (AFOSR)
[FA9550-15-1-0459].
10
Cryo EM workflows for Single Particle Analysis and Tomography of hydrated,
intact cells
W. Voorhout, M. Storms, G. van Duinen, J. Lengyel, M. Vos and B. Lich
FEI Company, Achtseweg Noord 5, 5651 GG Eindhoven, The Netherlands
A new frontier exists in unraveling interactive biological and biochemical processes and pathways at
the macromolecular level. Of critical importance is the three-dimensional visualization of
macromolecular structures and molecular machines in their native functional state. Three techniques
play a major role, NMR, XRD and Cryo-TEM.
Nuclear magnetic resonance (NMR) has the capability to study specific protein domains or fragments
and their functional role in protein folding and dynamics and in ligand binding whereas X-Ray
crystallography (XRD) allows visualizing high-resolution but more static 3D structures of apo and
liganded proteins, mainly in a monomeric or dimeric state after crystallization. To unravel more
physiologically relevant situations however, it is essential to visualize multimeric complexes in their
tertiary and quaternary state and their interaction with other complexes. Cryo-TEM applications like
single particle analysis one can visualize multimeric complexes. In this so-called translational
methodology, cryo-TEM thus provides complementary information to NMR and XRD that can be crucial
for a detailed structural analysis for a better understanding of the mechanism of the physiologically
relevant complex.
Latest developments in the cryo-TEM workflow have brought the 3 major structural biology
technologies closer together. Now, finally, a continuum has been reached on all important aspects
with regards to resolution and macromolecular scales which allows for the full deployment of the
combination of these technologies.
We will discuss the future of structural biology based on the latest developments of the FEI workflow
and its components.
11
Stress induced fusion of mitochondria visualized by electron tomography in
plants
P. Steiner (1), M. Luckner (2), G. Wanner (2) and U. Lütz-Meindl (1)
(1) University of Salzburg, Cell Biology Department, Hellbrunnerstraße 34, A-5020 Salzburg, Austria (2) Ludwig-Maximilians-University Munich, Faculty of Biology, Ultrastructural Research, Großhadernerstr. 2-4,
D-82152 Planegg-Martinsried, Germany
Physiological and molecular reactions of plant and animal cells to stress are well known from numerous
investigations. However, there is a tremendous lack in information on sub-structural alterations of
organelles that accompany stress induced processes. In higher plants and algae, structural stress
hallmarks have been reported in organelles like dictyosomes and chloroplasts [1, 2]. Advanced electron
microscopic techniques, such as electron tomography, have already provided evidence for structural
changes of mitochondria as consequence of stress or disease, both in plant and animal cells. [3, 4]
Alzheimer disease for example causes sub-structural alterations of the inner and outer membrane of
mitochondria in transgene mouse brain [4]. In the present study we investigate structural effects of
mitochondria in the unicellular freshwater alga Micrasterias denticulata after exposure to KCl by
means of FIB/SEM and TEM tomography. Whereas mitochondria in untreated control Micrasterias
cells are single, spherical or slightly elongated organelles, 3-D reconstructions of KCl exposed cells show
that they form 3-dimensional aggregates during stress. The membranous connections by which the
aggregates are formed are established by elongation of the outer mitochondrial membrane. In this
way mitochondria do not only fuse with each other but also with degenerated dictyosomes. As the
mitochondrial respiration potential of KCl stressed cells was almost the same as in controls [5] and as
the sub-structural alterations were reversible, we assume that mitochondrial aggregation is important
for maintaining essential cellular functions such as respiration during stress. We obtained similar
effects on mitochondria in the aquatic higher plant Lemna sp. after KCl exposure.
Figure 1: Protuberance of the outer
mitochondrial membrane induced by
150 mM KCl in Micrasterias denticulata.
(a) TEM micrograph, (b) reconstruction
from TEM tomography series. Protrusion
in orange.
[1] Lütz-Meindl, U., Luckner, M., Andosch, A. & Wanner, G. (2015) Journal of Microscopy, 263, 129-141. [2] Santos, C. L., Campos, A., Azevedo, H. & Caldeira, G. (2001) Journal of Experimental Botany, 52, 351–360. [3] Vartapetian, B. B., Andreeva, I. N., Generozova, I. P., Polyakova, L. I., Dolgikh, Y. I., Stepanova, A. Y. (2003) Annals of Botany, 91, 155-172. [4] Choi, K. J., Kim, M. J., Je, A. R., Jun, S., Lee, C., Lee, E., Jo, M., Huh, Y. H. & Kweon, H. (2014) Journal of Biosciences, 39, 97–105. [5] Affenzeller, M. J., Darehshouri, A., Andosch, A., Lütz, C. & Lütz-Meindl, U. (2009) Journal of Experimental Botany, 5, 854-855.
1 µm
12
Tsunamia transpacifica – TEM investigations in a newly-discovered red algal
genus colonizing Japanese tsunami debris
Sabrina Obwegeser (1), John West (2), Ursula Lütz-Meindl (3), Andreas Holzinger (1)
(1) University of Innsbruck, Department of Botany, Functional Plant Biology, 6020 Innsbruck, Austria
(2) School of Biosciences 2, University of Melbourne, Parkville, VIC 3010, Australia
(3) University of Salzburg, Cell Biology Department, 5020 Salzburg, Austria
On March 11, 2011 a tremendous earthquake off the pacific coast of Tohoku shook the northeastern
shore of Japan, unleashing 40m high tsunami waves leading to the worldwide known nuclear
meltdown in the Fukushima nuclear power plant. Large amounts of debris covered the land surface
and 5 million tons were washed into the ocean [1]. Debris is still (2017) carried by the North Pacific
Current and travels a distance of about 7200 km to the west coast of North America and can be found
in Oregon and Washington since 2013. In 2015 small plastic debris carrying dense pink crusts of algae
were recovered, where a sample was then isolated. The culture was analyzed by three-gene phylogeny
and revealed a new genus and species of the red algal class Stylonematophyceae, Tsunamia
transpacifica, referring to its origin [1]. The cells have a massive wall and thick extracellular matrix of
complex polysaccharides, a single central nucleus and a purple to pink multi-lobed parietal plastid
lacking a pyrenoid [1]. The first attempts to prepare the new genus for transmission-electron
microscopy by standard chemical fixation protocols were not successful. Preservation of the cells did
not allow distinguishing cellular components, except for the massive cell walls and extracellular matrix,
which may potentially lead to a deficient infiltration of the specimens during fixation and/or
embedding processes. This is surprising, as in other marine and freshwater red algae standard
protocols were successful [2, 3]. Therefore, we used a high pressure freezing and freeze substitution
protocol, using 2% OsO4 and 0.05% uranyl acetate for postfixation during the substitution as earlier
described [4]. This yields a high quality fixation of the ultrastructure of T. transpacifica and cellular
structures not described before in the class of Stylonematophyceae. Various vacuoles as well as
unknown electron dense bodies surrounding the nucleus were found. While such electron dense
bodies are frequently observed in brown algae, where they are described as physodes, e.g. in the arctic
Saccharina latissima [5], they are uncommon for red algae. Physodes are described as phorotannin-
containing bodies and due to their spectral properties have well characterized protective functions
against UV radiation. Insights in the ultrastructure of T. transpacifica contribute to a detailed
morphological knowledge and might contribute to an understanding of their adaptations to extreme
living conditions when colonizing floating plastic debris. Electron energy loss spectra (EELS) provide
insights into the chemical composition of the electron dense bodies in T. transpacifica and indicate
elevated levels of phosphorus and cobalt.
[1] J. A. West, G. I. Hansen, T. Hanyuda, G. C. Zuccarelllo (2016) Algae, 31, 1-13 [2] A. Holzinger, U. Karsten, C. Lütz, C. Wiencke (2004) Plant Biol. 6, 568-577
[3] S. Aigner, A. Holzinger, U. Karsten, I. Kranner (2017) Eur. J. Phycol., DOI: 10.1080/
09670262.2016.1274430 [4] N. Aichinger, U. Lütz-Meindl (2005), J Microsc. 219, 86-94 [5] A. Holzinger, L. Di Piazza, C. Lütz, M. Y. Roleda (2011) Phycol. Res., 59, 221-235
We kindly acknowledge financial support by the Austrian Science Fund (FWF):[I 1951-B16]
13
How to preserve a moss for element analysis? Different ways of preparation
for electron microscopy
Margret Eckhard(1), Siegfried Reipert(1), Ingeborg Lang(1) (1) University of Vienna, Core Facility Cell Imaging and Ultrastructure Research (CIUS), Althanstraße 14 1090
Vienna
In my master study, I look for the best preservation method for electron energy loss spectroscopy (EELS) in the moss Physcomitrella patens. The final aim is the quantification and localization of zinc that is stored in the plant cells. I am using the moss Physcomitrella patens because it is a well-known model organism; it is easy to cultivate in the laboratory and it is very tolerant to heavy metals [1], thus it is the perfect study object for my research. However, the moss cell wall is a tight barrier and therefore challenging for electron microscopic preparations.
To achieve this goal, I am combining various methods from light and electron microscopy. At the
electron microscopy level, I compare preparation protocols for chemical fixation and cryofixation using
the Leica HPM 100 and the Leica AFS 2 with a new agitation module for accelerated freeze substitution
[2].
By now, really good fixation results were established by using high-pressure freezing combined with
rapid freeze substitution. The probes are presently used for the element analysis and zinc detection in
the transmission electron microscope. To adapt the chemical fixation protocol, previous light
microscopic observations showed very divers reactions to the provided fixation buffers by the
gametophore and the protonemata of Physcomitrella patens, so that it is necessary to adjust the buffer
depending on the part of the moss that should be preserved.
[1]Sassmann, S., et al. (2015) Environmental and experimental botany, 118, 12-20
[2] Goldammer, H., et al. (2016) Protist, 167/4, 369- 376
C
CW
Figure 1 Physcomitrella patens. A: overview of gametophyte; B: gametophore leaf, cryofixed and freeze substituted. Gametophore (G), protonemata (P), vacuole (V), chloroplast (CL), starch (S), cytoplasm (C), cell wall (CW).
A B
P
G
S
V
CL
CW
C
2µm 200µm
14
How Cells Build Their Antenna: Centrioles Initiate Cilia Assembly, But Are Dispensable for Cilia Maturation and Maintenance
Daniel Serwas(1), Alexander Dammermann (1)
(1) Max F. Perutz Laboratories, University of Vienna, Vienna, Austria
In order to fulfill their versatile functions, cells need to be able to receive and transmit external
signals. Most human cells possess an evolutionarily conserved antenna-like structure on their
surface, the primary cilium, which participates in these processes.
Cilia form from centriole-derived basal bodies that serve as a platform for the assembly of multiple
structures including transition fibers, transition zone and axoneme. While it is clear that cilia
assembly is absolutely dependent on the presence of centrioles, it is not known whether centrioles
only trigger ciliogenesis or actively participate in downstream events. We used the nematode worm
C. elegans as an experimental model to address this question, since centriolar structures do not
persist at the base of mature cilia, but rather degenerate during ciliogenesis. Ciliary structures which
form after centriole loss clearly cannot be directly dependent on centrioles.
Using a combination of light microscopy and electron tomography, we generated the first timeline of
ciliogenesis. We found that the centriolar structural components SAS-6 and SAS-4 are lost during late
embryogenesis, leaving splayed microtubules as a remnant of the centriole wall at the ciliary base.
The transition zone and axoneme are not completely formed at this time, indicating that cilia
maturation does not depend on the presence of intact centrioles. The hydrolethalus syndrome
protein HYLS-1 is the only known centriolar protein that continues to localize to the base of mature C.
elegans cilia. Loss of HYLS-1 severely impairs docking and entry of intraflagellar transport particles
and thus cilia assembly. Surprisingly, targeted degradation of HYLS-1 after initiation of ciliogenesis
does not appreciably affect ciliary structures. Taken together, our findings show that centrioles serve
as a structural template to initiate cilia formation but are dispensable for their maturation and
maintenance.
15
Compartment-specific association of GABAB receptors and their effector ion
channels in cerebellar Purkinje cells
David Kleindienst(1), Rafael Luján(2), Carolina Aguado(2), Francisco Ciruela(3,4), Javier Cózar(2), Luis de la
Ossa(2), Kevin Wickman(5), Masahiko Watanabe(6), Yugo Fukazawa(7), Ryuichi Shigemoto(1)
Metabotropic GABA (GABAB) receptors mediate slow inhibition through their interaction with Gi/o
protein and downstream effector molecules such as G protein-coupled inwardly-rectifying potassium
channels (GIRK) or voltage-gated calcium channels (Cav). This interaction is reported to be
compartment specific. Thus, in presynaptic terminals activation of GABAB receptors reduces
neurotransmitter release by inhibition of Cav [1], whereas in postsynaptic elements, it activates GIRK
channels, causing a hyperpolarizing outflow of K+ [2]. Here, we investigated the spatial relationship of
GABAB receptors with two of their effector molecules, Cav2.1 and GIRK2, in dendritic shafts and spines
of cerebellar Purkinje cells and in presynaptic active zones of parallel fibres in mouse cerebellum. To
this end, we used SDS-digested freeze-fracture replica labelling immunoelectron microscopy, a
sensitive method enabling quantitative high-resolution detection of membrane proteins in brain tissue
[3]. We conducted double labelling of GABAB1 with Cav2.1 or GIRK2 using immunogold particles of
distinct sizes. To assess whether these molecules are co-localized, we developed the Gold Particle
Detection and Quantification (GPDQ) software. GPDQ first semi-automatically detects gold particles in
a delineated area of the image and then carries out two types of simulation, random and fitted. Using
random simulation in GPDQ, we found that GABAB1, GIRK2 and Cav2.1 are all significantly clustered in
all compartments. Fitted simulation (Fig. 1) takes the original distribution of one kind of particles into
account and ensures that distribution of distances between the simulated particles of the molecule of
interest are not significantly different from that of distances between the corresponding real particles.
Comparison of Nearest Neighbour Distances (NNDs) from real or simulated GIRK2/Cav2.1 to real
GABAB1 particles can then reveal whether a significant association of each of these molecules to GABAB1
exists. We found significant associations of GABAB1 and GIRK2 in dendritic spines, but significant
dissociation in dendritic shafts. On the other hand, GABAB1 and Cav2.1 were selectively associated in
dendritic shafts. These results indicate compartment- and molecule-specific regulation of co-clustering
of GABAB1 and its effector molecules, which may support compartment-specific GABAB1 functions.
Figure 1: Example of a fitted simulation of GIRK2 in a dendritic spine. Scale bar: 100nm.
[1] Kaupmann, Klemens, et al. 1998, Proceedings of the National Academy of Sciences 95, no. 25, 14991–96. [2] Takahashi, Tomoyuki et al. 1998, Journal of Neuroscience 18, no. 9, 3138–46. [3] Masugi-Tokita, Miwako and Ryuichi Shigemoto. 2007, Current Opinion in Neurobiology 17, no. 3, 387–93.
16
Characterization of the Vasa vasorum in the human great saphenous vein
by SEM and 3D-morphometry of vascular corrosion casts
Markus Herbst (1), Thomas Hölzenbein (2), Bernd Minnich (3)
(1) University of Salzburg, Physics Didactic Unit , Hellbrunnerstraße 34 , 5020 Salzburg (2) University Clinics of Vascular and Endovascular Surgery (PMU), Müllner Hauptstraße 48, 5020 Salzburg
(3) University of Salzburg, Vascular & Exercise Biology Unit, Hellbrunnerstraße 34 , 5020 Salzburg
"Vasa vasorum" (VV) derives from Latin and literally means "vessels of the vessels". Hence, the VV are
a network of small arterioles, venules and capillaries which supply the outer two layers of the wall
tissue of large blood vessels with oxygen and nutrients. The largest blood vessels in the body (e.g. the
human great saphenous vein, the aorta, etc.) depend on this supporting network to maintain their
health and function. Thus, the Vasa vasorum are an important part of the blood circulatory system.
In this study VV were studied in explanted segments of the human great saphenous vein (Vena saphena
magna, HGSV), taken during harvesting for coronary bypass grafts or extirpation of varicose vein
segments at the University Clinics for Vascular and Endovascular Surgery (PMU Salzburg), using
vascular corrosion casting (VCC), scanning electron microscopy (SEM, FEI/Philips XL-30 ESEM) and 3D-
morphometry (M3).
The aim of this study was the examination of the three-dimensional arrangement of the Vasa vasorum
in healthy and pathological (varicose) conditions. Moreover, it was intended to identify the most vital
segments of the HGSV in order to improve the results of bypass surgeries.
A meticulous analysis of the whole delicate microvascular system of the VV of the HGSV and its three-
dimensional arrangement (Fig. 1) is presented. It is one of the first studies yielding detailed quantitative
data on the geometry of the HGSV’s Vasa vasorum. Hence, a detailed insight into the optimality
principles (minimal lumen volume, minimal pumping power, minimal lumen surface and minimal
endothelial shear force) underlying the design of this microvascular network is given.
Arterial feeders originating from nearby arteries were
found to approach the HGSV every 15mm,
subsequentially forming a rich capillary network within
the adventitia and the outer two thirds of the media in
normal HGSV. In HGSV with intimal hyperplasia capillary
meshes of the VV were found to extend into the inner
layers of the media.
Measurements of spatial branching-off angles in
bifurcations and consecutive optimality calculations
showed that in both, the medial and distal part of the
HGSV, data are homogenously distributed close to the
theoretical optimum of vessel diameters.
Figure 1: The Vasa vasorum run predominantly parallel
to the longitudinal axis (LA) of the HGSV. Vessels having a longitudinal arrangement are defined as
orders 1 & 3. Orders 2 & 4 indicate vasa with a circular arrangement. Arterial vasa (A) are coloured in
red, venous vasa (V) in blue & capillaries (c) in orange. Arrows indicate the direction of blood flow.
[1] Herbst, M., Hölzenbein, T. & Minnich, B. (2014) Microscopy and Microanalysis, 20, 1120–1133.
17
The effects of double annealing on medium manganese steel
Tia Truglas(1), Christian Commenda(2), Martin Arndt(2), Daniel Krizan(2), Heiko Groiss (1, 3)
(1) CEST Competence Center for Electrochemical Surface Technology, Viktor Kaplan Str. 2, 2700 Wiener Neustadt
(2) voestalpine Stahl GmbH, voestalpine-Straße 3, 4031 Linz
(3) Zentrum für Oberflächen- und Nanoanalytik, Johannes Kepler Universität Linz, Altenberger Str. 69, 4040 Linz
The mechanical properties of batch or continuously annealed medium manganese steel grades with
an ultrafine-grained α + γ microstructure make them a promising candidate for the third generation of
advanced high strength steels. In the present study the effects of advanced continuous annealing
processes were investigated for a 0.1C6Mn2.2Al medium manganese steel with various electron
microscopy techniques, whereby a full microstructure analysis with EBSD and TEM was done after
single and double annealing. Compared to the former heat treatment the latter one contained an
additional annealing step using full austenitization and subsequent quenching prior to the final
intercritical annealing. EBSD phase maps revealed a fully α + γ microstructure with fine grain diameters
of around 400 nm for both steel types, whereas the simple annealed steel still contained larger ferrite
grains and a higher amount of low-angle grain boundaries (<15°). The double annealed steel contained
tendentially more retained austenite, which was also confirmed by XRD measurements.
Because of the ultrafine grain structure the limits of the EBSD methods were reached, thus grain sizes
were also measured using the line intercept method on STEM bright field micrographs and the phases
of individual grains were determined by selected area diffraction. Generally, the TEM showed a higher
dislocation density in the simple annealed steel and a more recrystallized structure in the double
annealed one. Extended EDX investigation showed a similar partitioning of manganese to the different
phases for both annealing types, whereas the partitioning of aluminium to ferrite was more
pronounced in the double annealed steel. The presence of manganese carbides in the simple annealed
steel in contrast to the precipitate-free microstructure of the double annealed one, constituted
another significant difference between the results of the two annealing processes. Their crystal
structure and chemical composition were determined by high resolution TEM, selective area
diffraction and EDX (see figure 1).
Figure 1: High resolution TEM micrograph of a manganese carbide (left). The Fourier transformation
of the image section (middle) together with the simulated diffraction pattern (right) of Mn5C2
orientated in the [-1,3,0]-direction allow the determination of the carbide type.
18
Phase characterization in Ni-base superalloy Rene 65
Tomasz Wojcik(1), Markus Rath(1), Ernst Kozeschnik(1)
(1) Institute for Materials Science and Technology, TU Wien, Getreidemarkt 9, 1060 Wien, Austria
The Ni-base superalloy Rene 65 is a newly introduced cast and wrought alloy, used for turbine disk
applications in aero-engines. The fine-grained and γ’-strengthened alloy was developed to increase
service temperatures up to over 700°C and therefore, enhance the efficiency of aero-space turbines.
Additional, this cast and wrought alloy can be manufactured at lower costs than the conventional
powder metallurgy alloys.
In this work, the secondary phases occurring in the as-received material as well as in different heat
treatment conditions are characterized by transmission electron microscopy (TEM). In the as-received
material condition, a tri-modal size distribution of γ’ precipitates is found ranging from a few
nanometers up to approx. 5 µm. In addition, borides are found preferable at grain boundaries with a
size of approx. 1 µm. The boride phases could be identified by means of energy-dispersive X-ray
spectroscopy (EDX), electron energy loss spectrometry (EELS) with selected area electron diffraction
(SAED) as tetragonal M3B2 and M5B3, respectively, with Cr and Mo as the main metallic constituents.
For different cooling rates, a change in the morphology and size distribution of the γ’ precipitates is
found. These results are compared with thermo-kinetic precipitation simulations using the MatCalc
software package.
19
Viscoelastic stress relaxation of TiAl thin film under tension measured by
selected area electron diffraction
Christian Ebner(1), Rohit Sarkar(2), Jagannathan Rajagopalan(2), Christian Rentenberger(1)
(1) University of Vienna, Physics of Nanostructured Materials, Boltzmanngasse 5, 1090 Vienna, Austria
(2) Arizona State University, Department of Materials Science and Engineering, School for Engineering of Matter Transport and Energy, Tempe 85287, USA
Metallic glasses are a new class of materials with very distinct properties, making them promising
materials for structural applications [1]. Understanding the mechanisms of deformation and modelling
these on an atomic level is a challenge, which has to be overcome to fully take advantage of the
materials properties.
Here, we present our study of the time-dependent viscoelastic strain response of an amorphous TiAl
thin film to changes of the external stress. In-situ tensile tests are performed using a Phillips CM200
TEM operating at 200kV. Selected area electron diffraction (SAED) is used as a method to extract the
local atomic-level elastic strain [2]. Elliptic distortions of the radial intensity maxima positions of the
SAED patterns are introduced by tensile straining (cf. Fig. 1(a)). By precisely measuring these
distortions, the 2-dimensional strain tensor is calculated with respect to a reference pattern. This
allows to quantify the principal strain e1 (parallel) and e2 (perpendicular to the loading direction) as a
function of the external stress σ. The specimen is loaded and unloaded stepwise to different levels of
external stress, denoted as states 0-4. After each stress change a time series of SAED patterns with 2s
resolution is acquired. Patterns are recorded for times up to 1h. The changes in principal strain Δe1
with respect to the first pattern of the time series are shown in Fig. 1(b). Fitting of these changes with
a relaxation time model function gives a good fit only if two different relaxation times are used. In
addition, a dependence of the relaxation times τ on the stress step Δσ is observed.
Figure 1: (a) SAED pattern of TiAl: By applying a uniaxial tensile stress to the specimen, elliptic
distortions of the diffraction pattern arise as illustrated by the sketch. The peak shifts are used to track
the local atomic level elastic strain. (b) Changes of the strain over time: The time dependent strain
response of the tensile specimen is recorded for two loading and two unloading steps. The difference
in strain over a period up to 1h is extracted from the SAD patterns and fitted by a model function
consisting of two distinct relaxation times, to obtain the best fit.
[1] Greer, A.L. (2009) Materials Today, 12, 1-2, 14-22. [2] Ebner, C., Sarkar, R., Rajagopalan, J. & Rentenberger, C. (2016) Ultramicroscopy, 165, 51-58. C. E. and C. R. acknowledge financial support by the Austrian Science Fund FWF: [I1309]. R. S. and J. R. acknowledge funding from the National Science Foundation (NSF) grants CMMI 1400505 and DMR 1454109.
20
HRTEM study of Ca doped Bismuth Ferrite
Ulrich Haselmann (1), Yurii P. Ivanov (1) and Zaoli Zhang (1), (2)
(1) Erich Schmid Institute of Material Science, Jahnstraße 12, 8700 Leoben
(2) Montanuniversität Leoben, Franz-Josef-Straße 18, 8700 Leoben
Bismuth Iron Oxide (BiFeO3) has been attracting lots of scientific attention in the past years, especially
for being one of the few single phase multiferroic materials with magnetoelectric coupling at room
temperature, of whom it shows a very high antiferromagnetic Néel temperature (TN ≈ 370°C) and
ferroelectric Curie temperature (TC ≈ 830 °C). In particular, the electrical control of the
antiferromagnetic domains was successfully demonstrated. [1]
In bulk form BiFeO3 shows the rhombohedral space group R3c, but when deposited as thin film it can
take other forms due to epitaxial strain induced during the growth, as for example super-tetragonal
and distorted rhombohedral forms with LaAlO3 as substrate [2].
Here we present a structural characterization of BiFeO3 doped by Ca (Ca0.1Bi0.9FeO3) via transmission
electron microscopy using a JEOL 2100F equipped with CS-Corrector and operated at 200 keV. The
Ca0.1Bi0.9FeO3 was grown on a Strontium Titanium Oxide (SrTiO3) substrate (001) with a buffer layer of
Strontium Rubidium Oxide (SrRuO3) (001) used also as electrode for electrical characterization. For
TEM study cross-sections have been prepared by mechanical grinding and polishing and subsequent
ion milling. Some preliminary data will be shown.
[1] T. Zhao et al., “Electrical control of antiferromagnetic domains in multiferroic BiFeO3 films at room
temperature.,” Nat. Mater., vol. 5, no. 10, pp. 823–9, 2006.
[2] R. Huang et al., “Atomic-scale visualization of polarization pinning and relaxation at coherent
BiFeO3/LaAlO3 interfaces,” Adv. Funct. Mater., vol. 24, no. 6, pp. 793–799, 2014.
We kindly acknowledge the financial support by the Austrian Science Fund (FWF): No. P29148-N36.
Our gratitude also goes to the group of Prof. Yinghao Chu from the National Chiao Tung University in Taiwan for
providing the samples.
21
Introducing overlapping grain boundaries in chemical vapor deposited
hexagonal boron nitride monolayer films
Bernhard C. Bayer (1), Sabina Caneva (2), Timothy J. Pennycook (1), Jani Kotakoski (1),
Clemens Mangler (1), Stephan Hofmann (2), Jannik C. Meyer (1)
(1) University of Vienna, Faculty of Physics, A-1090 Vienna, Austria
(2) University of Cambridge, Department of Engineering, CB3 0FA, Cambridge, UK
Hexagonal boron nitride (h-BN) is a two-dimensional (2D) insulator with a wide application profile,
including its use as an ultimately thin dielectric in electronics, a tunnel barrier in spintronics, an
encapsulation- and barrier-layer in electronics and metallurgy and a suspended separation membrane
in nanofluidics. A key technological challenge is the scalable manufacture of h-BN, in particular as a
continuous film of controlled layer number and high crystalline quality. Catalytic chemical vapor
deposition (CVD) has emerged as a promising technique to achieve growth of such h-BN films, including
exclusive monolayer growth [1-3]. CVD h-BN films are typically poly-crystalline and control of their
microstructure, in particular grain boundary (GB) structure, is important for many applications. Current
literature reports the nature of as-grown GBs in h-BN as atomically stitched, composed of defect lines
within a h-BN monolayer. Preferential pinhole formation is reported at such atomically stitched GBs,
and due to their structure they are intrinsically prone to electrical breakdown, chemical attack or
mechanical failure, all of which may render monolayer h-BN films ineffective in their envisaged
applications.
Here we show using complementary (scanning) transmission electron microscopy ((S)TEM) techniques
that GBs in monolayer h-BN films grown by scalable catalytic CVD can not only be atomically stitched
but can also be overlapping in nature. We show that in these overlapping GBs two h-BN monolayer
grains merge via the self-sealing formation of a turbostratic bilayer region of limited width and thereby
without formation of a defect line within the monolayer. We characterize this overlapping GB structure
in detail, identify catalytic CVD conditions that result in such GB structure and propose possible
underlying catalytic growth mechanisms. Our data suggests that overlapping GBs are comparatively
resilient against detrimental pinhole formation, as evolving defects in one layer are sealed by the
second layer. Thus overlapping GBs may be technologically advantageous for the many h-BN
applications for which continuous pinhole-free h-BN monolayers are key.
[1] Chem. Mater., 26, 6380 (2014).
[2] Nano Lett., 15, 1867 (2015).
[3] Nano Lett., 16, 1250 (2016).
We kindly acknowledge financial support from the European Union’s Horizon 2020 research and innovation
program under the Marie Skłodowska-Curie Grant Agreement 656214-2DInterFOX (B.C.B.) and from the Austrian
Science Fund (FWF, P25721-N20).
22
Combined analytical TEM and magnetic investigation of the effects of neutron irradiation on Nb3Sn superconductors
Pfeiffer Stephan(1), Bernardi Johannes(1), Stöger-Pollach Michael(1), Baumgartner Thomas(2),
Eisterer Michael(2), Ballarino Amalia(3)
USTEM, TU Wien, Wiedner Hauptstraße 8-10, 1040 Vienna, Austria
(2) Atominstitut, TU Wien, Stadionallee 2, 1020 Vienna, Austria
(3) CERN, 1211 Geneva 23, Switzerland
An increase of the high field critical currents in commercial Nb3Sn wires by about 50 % is required for
the design of FCC-hh (Future Circular Collider study for hadron/hadron collisions) superconducting
magnets. A previous study touched already this ambitious goal by producing additional pinning
centers. They were created by inducing crystal defects in the superconducting material by means of
fast neutron irradiation. In the present study, the underlying mechanisms are investigated through
combined microstructural and magnetic analysis. This knowledge will be important for industrial
manufacturing of the required high-performance superconducting cables.
The nuclear research reactor of TU Wien was used to irradiate Nb3Sn wires and already prepared
TEM (transmission electron microscopy) specimens. Micro- and nanostructural examinations of grain
geometry, grain boundary morphology, compositional gradients, local texture and defect structure
were performed in the TEM before and after irradiation by employing high-resolution TEM, EDX
(energy-dispersive X-ray spectroscopy), EELS (electron energy loss spectroscopy) and selected area
diffraction.
The results thereof are correlated with measurements of the superconducting properties, in
particular scanning Hall probe experiments and SQUID magnetometry to determine the global critical
current as well as the local critical current density within the superconducting subelements.
This study contributes to a better understanding of the influence of irradiation damage and the
resulting microstructure on local superconducting properties and ultimately on the macroscopic
performance of the superconductor.
Figure 1: High resolution image of neutron impact site (left) and remanent field Hall scan (right).
23
Advanced characterization of materials using atomic resolution TEM
Zaoli Zhang
Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, Leoben, 8700, Austria
In this presentation, some recent results acquired using our aberration corrected TEM will be shown.
The first part will focus on bulk nanostructured materials prepared by severe plastic deformation.
Here, the evolution of the structural and chemical composition in the nanostructured materials with
temperature was tracked in real-time via simultaneous in-situ imaging and spectroscopy analysis. It
demonstrates that the nanostructured materials are not only subjected to a structural change but also
to an obvious chemical composition fluctuation upon annealing. Real-time imaging and composition
determination allow further analyzing the dynamic behavior in nanocrystalline materials in details, i.e.
deducing the instantaneous diffusion coefficients and excess vacancy concentration generated by
deformation.
The second example will be on the CrN/AlN multilayers. CrN/AlN multilayers exhibit a peak in hardness
of ~40 GPa under certain bilayer period (Λ). These improvements in mechanical properties in
comparison with their monolithic counterparts have a close relationship with the existence of a
metastable face-centered cubic (fcc) AlN phase which can be epitaxially stabilized in thin films. Here,
interplanar spacing oscillations in cubic CrN/AlN multilayers were experimentally observed by using
spherical aberration-corrected high-resolution transmission electron microscopy (HRTEM), and were
corroborated by first principles calculations. Electron spectroscopy and microscopy were employed to
analyze the strain distribution in the multilayers and obtain generalized relationships between the
electronic structure on the one hand, and (non-)stoichiometry or strains in the multilayers on the other
hand. The present study provides atomic-scale insights in the mechanisms of extraordinary strength
pertaining to the CrN/AlN multilayers.
I would like to thanks to Reinhard Pippan, Jinming Guo, Julian Rosalie and Xunlong Gu (at the Erich
Schmid institute), Matthias Bartosik and Paul H.Mayrhofer (TU Wien), David Holec, Rostislav Daniel
and Christian Mitterer (Montanuniversität Leoben) for discussions and depositing the film materials.
24
High Resolution Visualisation of Iron Deposits in the Human Brain in Health
and Disease
Mariella Sele(1), Christoph Birkl (2), Stefan Ropele (2), Johannes Haybäck (3), Walter Gössler (4) and Gerd
Leitinger (1)
(1) Medical University of Graz, Institute of Cell Biology, Histology and Embryology, Harrachgasse 21, 8010 Graz (2) Medical University of Graz, Division of General Neurology, Auenbruggerplatz 22, 8036 Graz
(3) Medical University of Graz, Institute of Pathology, Auenbruggerplatz 25, 8036 Graz (4) University of Graz, Institute of Chemistry, Universitätsplatz 1, 8010 Graz
During aging from birth until the fourth decade of life, iron accumulates in various areas of the brain to different degrees [1]. Until now little is known how and why some brain areas contain significantly more iron than others. Moreover, iron accumulations are associated with many inflammatory and neurodegenerative diseases like Alzheimer’s disease (AD) or multiple sclerosis [2, 3]. Dysregulation of iron homeostasis or its release from damaged tissue can induce the production of radical oxygen species, cause oxidative stress and consequently apoptosis. It is known that Iron in the brain is mostly stored in glial cells or neurons [4] but how it gets there or why is it so unequally distributed in the brain is not known. Therefor we aim to elucidate the distribution, size and composition properties of the iron - containing ferritin in the human brain. With our comprehensive approach we combine findings from quantitative magnetic resonance imaging, mass spectrometry, analytical electron microscopy and immunochemical tests. In the analytical EM part we use energy filtered transmission electron microscopy (EFTEM) and Energy-dispersive X-ray spectroscopy (EDX). This interdisciplinary course of action will enable us to investigate which cell types and which subcellular compartments act as iron stores in the human brain. We thus aim to elucidate the mechanism that accumulates iron in the brain. Preliminary results confirm that the cellular and subcellular distribution of ferritin iron differs between areas with high and low iron content in human brain samples. The basal ganglia which consist of the globus pallidus (205 ± 32 ppm Iron) and the putamen (153 ± 29 ppm Iron) is the brain area in which the most iron was found [5]. In these areas were are able to show clusters of iron-loaded particles within oligodendrocytes. Our aim is to further characterise these Iron/Ferritin clusters. We propose that iron deregulation must be detectable in Alzheimer’s disease patient’s samples when studying the iron distribution of the basal ganglia. A better understanding of the iron distribution and the iron metabolism in the human brain could open new possibilities in treatment of neurodegenerative diseases.
[1] Hallgren, B. & Sourander P. (1958) Journal of Neurochemistry, 3, 41-51. [2] Smith, M., Harris P., Pauly, S., Sayre, L., & Perry G. (1997) Proc Natl Acad Sci U S A., 94, 9866-9868. [3] Khalil, M., Teunissen, C. & Langkammer, C. (2011) Mult Scler Int, 2011, 6 [4] Merugo, R., Asano, Y., Odagiri, S., Li, C. & Shoumura, K. (2008) Arch Histol Cytol, 71, 205-222 [5] Langkammer C., et al. (2010) Radiology, 257, 2, 455-462
We kindly acknowledge financial support by the Austrian Science Fund (FWF):[P-29370B27]
25
Impact of fibrinogen concentration on blood clot formation
Christoph Dibiasi1, Leon Ploszczanski2, Helga Lichtenegger2, Ursula Windberger1
(1) Department of Biomedical Research, Medical University of Vienna, Vienna, Austria
(2) Institute of Physics and Material Science, University of Natural Resources and Life Sciences Vienna
Blood is a fluid organ composed of cells embedded in blood plasma. During coagulation blood changes
it’s state to a solid by interlinking fibrin monomers to form a complex polymer attached to the cellular
components. This process can be quantified by measuring the viscoelastic properties of the clot, i.e.
shear storage modulus G’ by rheometry.
The aim of this study was to measure the impact of fibrinogen concentration (the precursor protein of
fibrin) on clot forming kinetics and final clot composition. For each measurement, 0.58mL blood of
human volunteers (n = 8) was sheared in the plate-plate measurement geometry of the rheometer
Physica MCR 301 (Anton Paar, Graz, Austria) until establishment of a G’ plateau, at which point the
blood clot was removed and fixated in formaldehyde. After drying and coating with Au the specimens
have been examined at high vacuum conditions and 20kV in a FEI 250 FEG ESEM.
We had defined two sample groups: One with normal fibrinogen concentration (mean 232.20 ± 50.75
mg/dL) and one with fibrinogen added to a final concentration of 1010.00 ± 196.83 mg/dL. SEM
pictures of clots from both groups are shown in fig 1 and 2. The fibrin network shows a greater density
with more fibers and less voids in the network. This corresponds to a higher G’ (223.50 ± 59.68 Pa in
the baseline group vs. 327.88 ± 58.86 Pa).
Fig 1: Clot with normal fibrinogen concentration
Fig 2: Clot with high fibrinogen concentration
26
Using electron microscopy as a method to monitor autophagy
Hubert Virginie1, Langer Brigitte1, Rees Andrew1, and Kain Renate1.
1Institute of Clinical Pathology, Medical University of Vienna
Autophagy is an evolutionary process used to eliminate cytoplasmic material through its accumulation
into a sealed structure the autophagosome that will then fuse with a lysosom. In the last few years,
major improvements have been made in the methods applied to monitor this process; among them is
transmission electron microscopy (TEM), which presents a much higher resolution than, for instance,
indirect immunofluorescence detection of antigens and subcellular structures. Following embedding
in epoxy resins, the autophagic compartments can easily be identified based on morphologic features;
autophagosomes are large structures often localized close to the endoplasmic reticulum, surrounded
by a double membrane and containing cytoplasmic material while lysosomes present a spherical shape
filled with electron dense material. Moreover, quantifying methods allow to gain information into the
nature of of proteins and (degraded) organelles accumulating within the cells and the subcellular
compartments. They thus allow to monitor the process of organelle trafficking and fusion in detail and
to investigate pathological processes, like blockage of fusion. Using TEM to study autophagy in
fibroblast cell lines sufficient and deficient for the lysosomal associated membrane protein-2 (LAMP-
2), we successfully demonstrated a new role of LAMP-2 in the fusion of the autophagosome with the
lysosomes. Moreover, we could also identify the presence of intact lysosome-like vesicles in the
autophagosomes of LAMP-2 deficient cells, a phenomenon previously unreported and identifiable only
by electron microscopy. This example successfully demonstrated the necessity of TEM to study
autophagy and the need to develop new methods by combining them with other techniques such as
immunofluorescence.
27
Serum derived exosomes as a putative diagnostic tool for ANCA associated
vasculitis
Stefan M. Schulz(1), Dario A. Leone(1), Helga Schachner(1), Andrew J. Rees(1), Renate Kain(1)
(1) Clinical Institute of Pathology, Medical University of Vienna, Vienna, Austria
Exosomes are extracellular vesicles present in most bodily fluids that recently came into the focus of
research for their unique properties and their wide range of potential applications in diagnosis and
treatment.
Exosomes are enclosed by a double membrane, secreted by most cell types and contain proteins,
DNA and RNA. They originate from various cell types and their distinct molecular signatures could
potentially constitute a novel diagnostic tool in autoimmune disease and cancer. While exosome
isolation from cell culture supernatants is well established, isolation from serum or plasma remains
challenging.
The aim of this project is to isolate individual populations of exosomes from human serum, to
identify their different cellular origin and determine whether their molecular composition is modified
in ANCA associated vasculitis (AAV), an autoimmune disease characterised by inflammation of small
blood vessels. We established reliable and robust protocols to isolate exosomes from serum or
plasma of AAV patients and healthy controls that are used to identify proteins expressed on the
surface of exosomes that could serve as disease specific biomarkers.
The purity of exosomes isolated from serum by ultracentrifugation and/or commercially available
reagents was assessed and their size and morphology validated using transmission electron
microscopy (TEM). Localization of membrane proteins was confirmed using immunogold labelling.
Sample protein content was evaluated using 1D-SDS-PAGE and Coomassie staining. Exosome
markers, e.g. the tetraspanins CD9 and CD63, were used to confirm the presence of exosomes in
Western Blot assays and real-time as well as QPCR were used to analyse their RNA/DNA content.
Our early results show, in accordance with previous publications [1], that ultracentrifugation resulted
in exosome populations with a larger diameter and achieved far lower protein yields than
commercial kits that are based on solubility that produced highly concentrated samples with a wide
particle size range.
[1] Helwa I, Cai J, Drewry MD, Zimmerman A, Dinkins MB, Khaled ML, et al. (2017) PLoS ONE 12(1): e0170628.
doi:10.1371/journal.pone.0170628
28
Flash and Freeze: combining high-pressure freezing and optogenetics to
evaluate synaptic transmission
Carolina Borges-Merjane, Olena Kim, Peter Jonas
Institute of Science and Technology (IST) Austria, Am Campus 1, 3400 Klosterneuburg
The visualization of the ultrastructure of synapses by electron microscopy with high pressure
freezing (HPF) has enabled a better understanding of their morphological properties and subcellular
structures. However, synaptic transmission is a dynamic process, and HPF alone only captures static
images. The recently developed “Flash and Freeze” technique1,2,3 combines optogenetics with HPF
allowing for the visualization of action potential evoked membrane dynamic changes at synapses
during neurotransmission. With this powerful technique, a brief pulse of light activates the genetically
expressed light-activated channel channelrhodopsin in targeted cells, leading to action potential
initiation and inducing vesicle fusion to the membrane. After light stimulation the tissue is frozen by
HPF with a pre-set timed delay after onset of stimulus, thus allowing for capture of events at different
time points after synaptic transmission onset, from exo- to endocytosis.
We are using the Leica ICE with Light Stimulation system4 for “Flash and Freeze”, to assess
mechanisms underlying synaptic transmission at the mossy fiber-to-CA3 pyramidal cell synapse - in
mouse hippocampus during basal transmission and after short-term plasticity. We use acute brain
slices from 21 to 28 day-old mice and organotypic slice culture, prepared from 5 to 7 day-old mice,
maximum two weeks old. We are using transgenic mice, specifically with cre-recombinaseER in dentate
gyrus granule cells under the prox1 gene promoter, and are currently testing crosses with reporter
lines expressing the light-activated channel channelrhodopsin for specific expression. With this
method and approach, we hope to have a better understanding of presynaptic changes that occur at
the mossy fiber terminals, contacting CA3 pyramidal neurons in hippocampus. Figure 1: Left: Confocal z-stack (0.5 μm steps) of a horizontal section of hippocampus from an Ai27 het/Prox1-
creER hem mouse injected with tamoxifen. Neurons labeled with anti-NeuN antibody (cyan) and dentate gyrus
(DG) granule cells labeled in red showing specificity of expression of channelrhodopsin. Scale 200 μm. Middle:
Voltage-clamp recording of a granule cell from an acute brain slice of a transgenic mouse showing response to 5
ms blue light stimulation. Overlaid traces showing multiple trials. Right: Transmission electron microscope image
of a DAB stained, fixed hippocampal slice after pre-embedding with anti-RFP antibody (in this mouse TdTomato
is fused with channerhodopsin). MFs: mossy fibers; MFB: mossy fiber bouton. Scale 1 μm.
[1] Watanabe, S., Liu, Q., Davis, M.W., Hollopeter, G., Thomas, N., Jorgensen, N.B. & Jorgensen E.M. (2013) Elife.
Sep 3;2:e00723. doi: 10.7554/eLife.00723.
[2] Watanabe, S., Rost, B.R., Camacho-Pérez, M., Davis, M.W., Söhl-Kielczynski, B., Rosenmund, C., & Jorgensen,
E.M. (2013) Nature, 504(7479):242-7. doi: 10.1038/nature12809.
[3] Watanabe, S., Davis, M.W., & Jorgensen, E.M. (2014) Nanoscale Imaging of Synapses, Chapter 3, 43-57
[4] Leica Microsystems, Vienna, Austria. http://www.leica-microsystems.com/products/sample-preparation-for-electron-microscopy/cryo-preparation-systems/details/product/leica-em-ice We thank our funding sources: C.B.M MSCA H2020 708497; PJ (OK) FWF W1205-B09; PJ ERC 692692
29
SEM on agarose-based chromatographic beads – how to recalculate a reference SAXS scattering signal from an image
Jacek Plewka1,2, Leon Ploszczanski1, Heinz Rossbacher1, Rupert Tscheliessnig2, Alois Jungbauer2, Harald Rennhofer1, Helga Lichtenegger1
(1) Institute of Physics and Material Science, University of Natural Resources and Life Sciences Vienna (2) Austrian Centre of Industrial Biotechnology, ACIB GmbH
With the total sales of $75 billion, monoclonal antibodies are the most lucrative product on bio-
pharmaceutical market accounting for over 50% of worldwide market. Agarose-based chromatography
media, used for protein-A affinity chromatography - method for antibody capturing, are then of the
upmost importance in biopharmaceutical industry. Being the most expensive steps in antibody
purification process its full understanding, including the mechanical properties on nanometer scale, is
essential to ensure the performance.
Here, we would like to demonstrate a method for inner structure visualization of agarose-based
chromatographic beads using Scanning Electron Microscopy (SEM) approach and subsequent image
processing to reconstruct Small Angle X-Ray Scattering (SAXS) images recorded on the same material.
Although, those two methods provide quite different approaches (microscopy gives local details on
surface, whereas SAXS provides global parameters), they are often employed together for deeper
understanding of analyzed materials on the nanoscale. However, traditionally SEM is only used for
morphological examination of specimen, whereas we propose to use it for further processing to get a
reference signal for SAXS method as well.
In the Figure below a short summary of used methods is shown. A dehydrated resin is embedded in LR
white resin and sliced using microtome to ensure smooth surface of the specimen and then coated
with thin layer of gold prior SEM image capturing. Captured images of sufficiently good resolution in
nanometer scale are then further processed to obtain a reference SAXS signal using 2D image fast
Fourier transform (FFT) and subsequent 1D radial averaging of the image in reciprocal space.
Alternatively, one can also binarize the image and randomly probe the surface of a bead with
significant number of points to calculate the Pair Distribution Function out of them (measure of
distances and their probabilities), which using the Debye formula can be recalculated to yield the SAXS
reference signal as well. Such a reference signal can be used to check the background subtraction
quality for SAXS method or to extrapolate the SAXS signal to the very low-q regime (corresponding to
a size range larger than 100 nm), where due to experimental restrictions no SAXS information is
available.
30
Babinet principle for plasmonic antennas: complementarity and differences
Michal Horák (1), Vlastimil Křápek (1,2), Martin Hrtoň (1,2), Michael Stöger-Pollach (3),
Tomáš Šamořil (1,2), Filip Ligmajer (1,2), Tomáš Šikola (1,2)
(1) Central European Institute of Technology, Brno University of Technology, Purkyňova 123, 612 00 Brno, Czech Republic
(2) Institute of Physical Engineering, Brno University of Technology, Technická 2, 616 69 Brno, Czech Republic
(3) University Service Center for Transmission Electron Microscopy, Vienna University of Technology, Wiedner Hauptstrasse 8-10, 1040 Vienna, Austria
Babinet principle relates the optical response of apertures in thin films and their complementary
analogues – solid barriers or particles. Originating in the wave theory of light and analysis of diffraction,
it has recently entered the field of plasmonics. According to Babinet principle, localized surface
plasmons in complementary particles and apertures have identical resonance energies and their near
field are closely linked: The electric field distribution of a specific in-plane polarization for an aperture
corresponds to the magnetic field distribution of perpendicular polarization for a particle [1]. On the
other hand, additional differences can be related to different fabrication processes and experimental
techniques involved in the characterization of real structures.
To assess the theoretically predicted Babinet complementarity, we have studied a set of gold disc-
shaped plasmonic antennas with various diameters, both particles and apertures. Plasmonic antennas
were fabricated by focused-ion-beam lithography of thin gold layer on silicon nitride membrane.
Localized plasmon resonances were characterized by cathodoluminescence and electron energy loss
spectroscopy (EELS). Babinet complementarity was confirmed for main plasmon properties such as
resonance energies, but differences were found, for example, for the excitation efficiency (Fig. 1).
Figure 1: Left: Cathodoluminescence spectra of gold dics-shaped plasmonic particles and apertures of
various diameters. The excitation electron beam was focused on the edge of the structure to maximize
the excitation efficiency. Note generally stronger response of the apertures. Right: Dispersion relation
of localized plasmon resonances. Peak energy of cathodoluminescence is shown as a function of
reciprocal value of the antenna diameter (resembling the wave number). Experimental values show
no difference between particles and apertures and closely follow the values obtained from numerical
simulations.
[1] Hentschel, M., Weiss, T., Bagheri, S., & Giessen, H. (2013) Nano Letters, 13, 4428–4433.
We kindly acknowledge financial support by Czech Science Foundation, project No. 17-25799S.
31
Hybrid plasmonics: From plasmon-plasmon to plasmon-exciton coupling
Franz-Philipp Schmidt(1,2), Harald Ditlbacher(1), Andreas Hohenau(1), Ulrich Hohenester(1), Ferdinand
Hofer(2), and Joachim R. Krenn(1)
(1) Institute of Physics, University of Graz (2) Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology
The coupling of plasmonic nanoparticles can lead to extremely localized fields and is thus a central
topic in nanooptics research and application. In our work, we show that the spectral response of a
single rectangular plasmonic nanoparticle can be interpreted due to coupling of edge excitations,
leading to bonding and antibonding modes (Fig.1a,b) [1]. On one hand, we rely on high-resolution
experimental data from electron energy-loss spectroscopy, applied to a single lithographically
prepared silver cuboid. On the other hand, we use numerical simulations by the boundary element
method, finding excellent agreement with the experiment.
Going one step further we couple metallic with semiconducting nanostructures (Fig.1c) in terms of
plasmon-exciton coupling [2]. The importance of high energy resolution to differentiate subtle energy
shifts and splittings is demonstrated using a monochromated system in combination with advanced
data post processing routines (Fig.1d) [3].
Figure 1:
Plasmon-plasmon coupling: (a) EEL spectra extracted from three different regions of a silver
nanocuboid as indicated in the inset. (b) Simulated charge distribution of the dipolar plasmon edge
mode along a 300 nm long silver edge, which splits up into a bonding and antibonding mode due to
coupling of the opposite edge plasmons.
Plasmon-exciton coupling: (c) “Monochromated” HAADF images of a silver dimer and CdSe/ZnS
quantum dots and (d) corresponding EEL spectra in the gap region before (blue) and after (red) data
post processing.
[1] Schmidt, F.-P., Ditlbacher, H., Hohenau, A., Hohenester, U., Hofer, F., Krenn, J. R. (2016) Nano Letters, 16, 5152–5155. [2] Wei, J., Jiang, N., Xu, J., Bai, X., Liu, J. (2015) Nano Letters, 15, 5926–5931. [3] Schmidt, F.-P., Hofer, F., Krenn, J. R. (2017) Micron, 93, 43–51. This research was supported by the Austrian Science Fund FWF (P21800-N20, SFB F49), NAWI Graz and the Graz Center for Electron Microscopy.
32
Influences of the CMR effect on dielectric properties
Wolfgang Wallisch (1), Michael Stöger-Pollach (1), Edvinas Navickas (2)
(1) Technische Universität Wien, University Service Centre for TEM, Wiedner Hauptstrasse 8-10, 1040 Vienna (2) Technische Universität Wien, Institute of Chemical Technologies and Analytics, Getreidemarkt 9, 1040 Vienna
Over the last decades, double perovskite oxides are attractive materials due to their complex magnetic
and electrical behaviour. Their promising physical and electronic properties are interesting for a wide
range of applications. Among these properties, also the large negative magnetoresistance of
La2CoMnO6 (LCM) [1], which is known as the colossal magnetoresistance (CMR), is of great interest.
This effect is an metal-insulator transition describing the change of the resistance in the presence of a
magnetic field.
These days, transmission electron microscopy (TEM) equipped with energy filters are powerful tools
and its main advantage for probing band gaps is the high spatial resolution. It offers an opportunity to
investigate the influences of the physical consequences of the CMR effect on the electron energy loss
spectrometry (EELS) signal in TEM. The observation and detection of the change of the band structure
in the low energy range and the magnetic behaviour of the material with chemical sensitivity [2] by
using energy loss magnetic chiral dichroism (EMCD) will be presented.
The dielectric response is contained in the low loss spectrum, which is exhibited in Fig. 1. The 40 keV
and 200 keV spectra are shown at a temperature of 85 K in Fig. 1A. It is obvious that there is a difference
in the energy loss range of 1.5 eV to 4 keV. On the other hand, concerning the comparison of the 40
keV valence EELS (VEELS) spectra at different temperatures (Fig. 1B), an intensity variation is caused
by the CMR effect and not by the Čerenkov effect. The EMCD investigations are performed at 200 keV.
The CMR effect causes a magnetisability in a magnetic field of less than approximately 0.5 T [1]. The
chemical sensitivity of EMCD is shown in Fig. 1C, the EMCD effect can be observed at the Co edge in
the 85 K experiment.
Figure 1: (A) Unprocessed VEELS spectra recorded at 85 K at a sample thickness of 0.3 λ using 40 keV
and 200 keV, respectively. (B) Low loss spectrum recorded at room temperature (RT) and at 85 K using
40 keV electrons. The insertion shows the divergences between the RT and the 85 K spectrum. (C)
Normalized EELS spectrum of the LCM layer. The Co edge shows induced chiral electronic transitions
at 85 K.
[1] Mahato, R. N., Sethupathi, K. & Sankaranarayanan, V. (2010) Journal of Applied Physics, 107, 09D714. [2] Ennen, I., Löffler, S., Kübel, C., Wang, D., Auge, A., Hütten, A. & Schattschneider, P. (2012) Journal
of Magnetism and Magnetic Materials, 324, 2723-2726. The authors kindly acknowledge financial support by the Austrian Science Fund (FWF):[F4501-N16, F4509-N16].
33
Reaction of Ni and C thin films studied by TEM and SEM
Semir Tulić (1), Viera Skákalová (1), Thomas Waitz (1), Gerlinde Habler (2), Marián Varga (3), Alexander
Kromka (3), Viliam Vretenár (4), Mária Čaplovičová (4)
(1) Faculty of Physics, University of Vienna, Boltzmanngasse 5, 1090 Vienna, Austria (2) Department of Lithospheric Research, University of Vienna, Althanstrasse 14, 1090 Vienna, Austria
(3) Institute of Physics, Czech Academy of Sciences, Cukrovarnická 10, Prague 6, Czech Republic (4) Slovak University of Technology, Center for Nanodiagnostics, Vazovova 5, 812 43 Bratislava, Slovakia
This work focuses on the catalytic reaction of thin Ni films (thickness 20 and 500 nm) with
nanocrystalline diamond (NCD; grains ~200 nm in diameter). The films are deposited on Si substrates
by magnetron sputter deposition to yield a Ni-NCD-Si sequence. After reaction by annealing at a
temperature of 900 °C, samples are studied prior and after the removal of any residual Ni surface layer
by using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). TEM
specimens were prepared in a cross-sectional geometry by focused ion beam (FIB) thinning using
protective layers of Au and Pt. After annealing of the samples with 500 nm thick Ni films (denoted NCD-
500), columns of NiSi2 are covered with a surface layer of C (see Figs. 1(a) and (b)). While graphite is
observed directly at the interface with NiSi2, with increasing distance from this interface, the layered
structure of graphite seems to become more disordered, finally yielding amorphous C (a-C) (see Fig.
1(c)). Graphite is arising by a catalytic reaction of the Ni with diamond [1]. NiSi2 is arising by a reaction
of Ni diffusing along the grain boundaries of the NCD towards the Si substrate. In addition, Si and C
have interacted to form nanocrystallites of SiC. After annealing of the samples with 20 nm thick Ni
films (denoted NCD-20), isolated Ni nanoparticles arise by dewetting of the Ni film; catalytic etching
by the Ni nanoparticles causes the formation of grooves in the NCD (see Fig. 2(a)) On top of the NCD,
a continuous layer of a-C is observed (see Figs. 2(b) and (c)). Since the catalytic reaction of Ni and C is
expected to yield the formation of graphite [1], in the present case the a-C might have formed by
radiation damage during the specimen preparation via FIB [2].
Figure 1: NCD-500. (a) SEM image of NiSi2 columns decorated with Ni nanoparticles. (b) TEM images
of (a) C-NiSi2-SiC-Si reaction nanostructures and (c) graphite gradually changing to amorphous C.
Figure 2: NCD-20. (a) SEM image of NCD showing grooves. TEM images (b) of the reaction layers and
(c) the interface between the NCD and a-C.
[1] H. Mehedi, et al., Carbon, 59, 448-456, 2013.
[2] R. Colby, et al., Diamond and Related Materials, 19 (2), 143-146, 2013.
We kindly acknowledge financial support by the Austrian Science Fund (FWF), Czech Science Foundation GACR
and Slovak Scientific Grant Agency VEGA:[AI0234421, 16-34856L, 1/1004/15].
34
Analytical FIB-SEM Tomography without Compromises
Fabián Pérez-Willard (2), Giuseppe Pavia (2), Wolfgang Schwinger(1)
(1) Carl Zeiss GmbH, Laxenburger Str. 2, AT-1100 Vienna, Austria
(2) Carl Zeiss Microscopy GmbH, Carl-Zeiss-Str. 22, DE-73447 Oberkochen, Germany
In materials research the capability to analyse comprehensively the microstructure of a specimen in
three-dimensions is becoming increasingly important. In this context X-ray and FIB-SEM microscopy –
the focus of this work – play a key role as they enable researchers to understand structural changes
caused by processing or use of a material across different relevant length scales [1]. As a result,
materials with better properties and performance can be developed more efficiently.
While FIB-SEM tomography provides its best spatial resolution of a few nm voxel size, when performed
at low accelerating voltages, usually between 1 and 2 kV, energy dispersive spectroscopy (EDS)
requires at least a factor of two to three larger landing energies for the excitation of the characteristic
EDS fingerprint. In the past, analytical FIB-SEM tomography was always performed at the lowest
acceleration voltage still compatible with the EDS analysis, thus sacrificing spatial resolution in SEM
imaging.
Recently, a software solution within the Atlas 5 tomography environment has been developed, which
allows automatic switching between two different sets of SEM conditions: A first one at low voltage
and current for the acquisition of high-resolution electron images and smallest possible voxel sizes.
And a second at a much higher voltage and current, for high-throughput EDS mapping with lower
spatial resolution and larger voxel sizes [2].
In this contribution, we will present some materials science examples to illustrate the advantages of
this new approach.
Figure 1: Exemplary slice from a FIB-SEM tomography dataset on a lead free solder sample (courtesy
of M. Cantoni, EPFL Lausanne). Electron imaging (left) was done at 1.8 kV using Inlens SE detection
with a voxel size of (10 x 10 x 10) nm³. The EDS maps were acquired every tenth slice at 6 kV with a
voxel size of (40 x 40 x 100) nm³.
[1] Merkle, A. et al. (2014) Microscopy and Analysis, 28(7), 10-13.
[2] Cantoni, M. et al. (2016) Proceedings of the 16th European Microscopy Congress, Lyon, France.
35
Dealing with light refraction in 3D mapping in combined Raman/SEM
Robert Striemitzer (1,2), Peter Pölt (1,2) ,Harald Fitzek (1),
(1) Institute of Electron Microscopy and Nanoanalysis, Graz University of Technology, NAWI Graz,
Steyrergasse 17, 8010 Graz
(2) Graz Centre for Electron microscopy, Steyrergasse 17, 8010 Graz
In recent years Raman-microscopy and scanning electron microscopy (SEM) were integrated in one
instrument, and the advantages of correlative Raman-microscopy and SEM could be demonstrated
[Zitat Raman/SEM Paper]. One advantage of a combined Raman/SEM is that depth profiling and 3D-
maps are possible using confocal Raman microscopy. Unfortunately, the sample has to be immersed
in oil for confocal microscopy as otherwise the refraction of light at the sample surface will lead to a
quick deterioration of the depth resolution with focus depth and a compressed depth scale [Evrall
1,2]. While it is possible to mathematically correct the compression of the depth scale, the lost
resolution cannot be recovered. Therefore, we are trying to make immersion possible in the vacuum
of a SEM chamber.
A measurement with the least information loss would occur with light not changing medium at all or
a change to media with the same refractive index (Reflection is negligible in both cases). To get such a
case usually oil with the preferred refractive index is dropped on the sample and an oil immersion
objective is used. Our approach is to introduce a thin film with a refractive index close to the
immersion oil between the sample and the oil. If the contact between the sample and the film is
satisfactory this would allow for immersion to be used in vacuum. The first experiments using
different methods are displayed in Fig. 1 and compared to the standard oil approach and measuring
at ambient air.
Figure 1: Visualisation of polymer multilayers (PS, PMMA, PS, PMMA) on a PET substrate measured
with different techniques: Space between sample and objective filled with a) Oil, b) Air, c) Tape and
Oil, d) Silicone Layer and Oil
The thickness obtained by measuring with oil corresponds with the known thickness of 52.5 µm. The
closest condition to vacuum (n = 1) is air (n � 1.0001) and its result is compressed as expected. The
results coming from measurements with a protective adhesive tape and a silicone layer are within
uncertainty range equivalent to the measurement with oil.
[1] Worobiec, A., Potgieter-Vermaak, S., Brooker, A., Darchuk, L., Stefaniak, E., & Van Grieken, R. (2010),
Interfaced SEM/EDX and micro-Raman Spectrometry for characterisation of heterogeneous environmental
particles— fundamental and practical challenges, Microchemical Journal, 94(1), 65-72 [2] Evrall, N. (2004), Depth profiling with confocal Raman microscopy, Part I. Spectroscopy-Springfield then Eugene then Duluth, 19, 22-33 [3] Evrall, N. (2004), Depth profiling with confocal Raman microscopy, Part II. Spectroscopy-Springfield then Eugene then Duluth, 19, 16-27
Oil PMMA PS PET
Air Tape Glue Silicone
a) Oil b) Air c) T ape d) Silicone Layer
36
SEM Characterization of functionalized Carbon Nanotubes
Philipp Siedlaczek, Gerald Singer, Gerhard Sinn, Leon Ploszczanski, Helga Lichtenegger
University of Natural Resources and Life Sciences Vienna, Institute of Physics and Material Sciences, Peter-
Jordan-Straße 82, 1190 Vienna, Austria
Carbon nanotubes are high performance materials which exhibit superior mechanical, electrical and
thermal properties. Due to their outstanding metallic and semi-conducting behaviour, research on a
large scale is currently done for electronic applications. Semiconductors, batteries, capacitors and
photovoltaic systems are just a few future prospects of their field of application. However, attention
is also paid to their mechanical properties since the measured tensile strength can achieve up to
63GPa, which is multiple times more than steel. This quality can be used to reinforce carbon fibre
composites substantially.
In order to utilize carbon nanotubes adequately in composites, the nanotubes have to be modified.
Their strong nonpolar character refuses any interaction with epoxy resins, the important polymer for
adhering carbon fibre mats. Thus, chemical treatment of nanotubes is necessary to attach functional
groups, such as carboxylic groups and/or amine groups, at the surface of the molecule via oxidation to
induce intermolecular linkages to epoxy groups or amine groups of the hardener. After the reaction,
the mechanical properties along with thermal and electrical conductivity can be exploited to a great
extent. Common oxidation methods contain aggressive chemicals such as sulfuric acid, nitric acid,
potassium permanganate and sodium nitrate in large quantities. The exhaust gases and side products
are hazardous and toxic. Too harsh treatments additionally damage the carbon nanotubes and
subsequently reduce its mechanical properties tremendously.
This master thesis follows a gentler, more economic and environmental friendlier approach with H2O2
as an oxidizing agent. Analysis with Scanning Electron Microscopy (SEM) and Energy Dispersive X-Ray
Spectroscopy (EDX) offer a powerful method to determine successful oxidation techniques.
Furthermore, it is possible to analyse the degree of destruction and the expected ability to form
crosslinks with epoxy resins.
37
FEBID Based Direct-Write of 3D Plasmonic Gold Structures
Robert Winkler(1), Franz Schmidt(1,2), Ulrich Haselmann(1), Jason Fowlkes(3,4), Philip Rack(3,4), Harald
Plank(1,5)
(1) Graz Centre for Electron Microscopy, Steyrergasse 17, 8010 Graz (2) Institute of Physics, Karl-Franzens-University, Universitätsplatz 5, 8010 Graz
(3) Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee, USA (4) Department of Materials Science and Engineering, University of Tennessee, Knoxville, Tennessee, USA
(5) Institute for Electron Microscopy, Graz University of Technology, Steyrergasse 17, 8010 Graz
During the last decade, resonant optics attracted enormous interest in science and technology as this
research field provides deep insights in fundamental physics but also an increasing number of
applications ranging from filters over waveguides towards sensor devices. While several techniques
for the fabrication of metallic structures have been introduced [1] the direct-write fabrication of highly
defined structures on the nanoscale, especially for complex three-dimensional geometries on non-flat
surfaces is still an intractable challenge. Focused Electron Beam Induced Deposition (FEBID) recently
has taken a huge step forward in terms of fabrication of predictable and complex three-dimensional
geometries, leveraging this technique into the status of a nano-printer [2,3] not only for 2D but also in
real 3D on almost any substrate material and morphology. Beside the reliable shape performance on
the nanoscale high purity of the material is essential for plasmonic activity. Therefore, the direct usage
of FEBID structures for plasmonic investigations is impossible due to the high carbon impurities of
about 90 at.% [4].
In this contribution we focus on the fabrication of complex freestanding FEBID structures (Fig. 1) in
general and in particular for the plasmonic investigation. In this context an alternating point sequence
approach is presented that enables 3D-nanoprinting beyond current limitations. In the following we
briefly show a purification process to Au precursor Me2Au(acac) as an ideal material for plasmonics
leading to pure Au structures in general. Applying the described purification process with adapted
parameter to such 3D deposits leads to compact and pure Au structures. In the last step, we present
TEM based EELS measurements to show plasmonic activity. This demonstrates the potential of FEBID
as fabrication method for free-standing, three-dimensional plasmonic structures on practically any
given surface. By that, the field of resonant optics can be expanded by yet unknown 3D architectures
in combination with regions which were very complicated to access in the past.
Figure 1: 3D-nanoprinting of plasmonic active FEBID-structures: First, complex 3D-nanoarchitectures
are reliably fabricated (left). After that a purification step utilizing electron stimulated reactions with
water vapor is introduced to transfer the Au-C deposition into pure gold as shown via TEM
characterization (center). Finally, STEM-EELS investigation revealed plasmonic activity (right).
[1] L. Hirt, A. Reiser, R. Spolenak, and T. Zambelli, (2017) Adv. Mater. 1604211. [2] J. D. Fowlkes, R. Winkler, B. Lewis, M. Stanford, H. Plank, P. D. Rack (2016) ACS Nano, 10, 6163-6172. [3] R. Winkler, F. P. Schmidt, U. Haselmann, J. D. Fowlkes, B. B. Lewis, G. Kothleitner, P. D. Rack, H. Plank
(2017) ACS Appl. Mater. Interfaces, 9, 8233-8240. [4] A. Botman, J. J. L. Mulders, R. Weemaes, S. Mentink (2006) Nanotechnology 17, 3779-3785.
38
Direct-Write Fabrication of Electric and Thermal High-Resolution Nanoprobes
on Self-Sensing AFM Cantilever
J. Sattelkow(1), J. Fröch(1,2), R. Winkler(2), U. Radeschnig(2), C. Schwalb(3,4), M. Winhold(3,4), A.
Deutschinger(3,4), T. Strunz(3,4), V. Stavrov(5), G. Fantner(6), E. Fantner(6), H. Plank(1,2,*)
1 Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology, Graz, AUSTRIA, 2 Graz Centre for Electron Microscopy, Graz, AUSTRIA, 3 GETec, Vienna, AUSTRIA, 4 SCL Sensor Tech., Vienna, AUSTRIA, 5 AMGT, Bovegrad, BULGARIA, 6 Laboratory for Bio- and Nano-Instrumentation, EPFL, Lausanne, SWITZERLAND,
* corresponding author: [email protected]
Atomic Force Microscopy (AFM) has evolved into an essential part in research and development due
to its quantitative 3D surface characterization capability and additional AFM modes, which provide
laterally resolved electric, magnetic, chemical, mechanical, optical, or thermal properties of the sample
surface. Our Partner GETec Company has introduced an AFM system (AFSEM®) providing a high-
resolution AFM tube scanner, which can be integrated into standard Scanning Electron Microscopes
(SEM) / Focused Ion Beam Microscopes (FIB) / Dual Beam Microscopes (DBM). The use of self-sensing
cantilever eliminates an optical detection system as it uses stress-strain elements for the electric
readout according to the cantilever motion. The application of the self-sensing technology, however,
prevents traditional tip fabrication or subsequent modification such as large area coating with
conductive or magnetic materials. Hence, a method is needed to allow the highly localized, functional
tip fabrication and / or modification according to AFM mode related requirements.
Based on this motivation we here demonstrate a concept, which aims on the Focused Electron Beam
Induced Deposition (FEBID) based fabrication of specialized AFM tips for electric and thermal nano-
probing via the AFSEM®. A common self-sensing cantilever (SS-CL) platform as substrate will be
modified via FEBID towards two different functionalities. For electric nano-probes, Pt-C nano-pillars
are first fabricated and then purified by our gas assisted purification approach as demonstrated by
Geier et-al.[1] This contribution discusses chemical / structural aspects of the FEBID high-resolution
tips together with Conductive-AFM (C-AFM) measurements to demonstrate the capabilities of this
approach. For thermal nano-probes we take advantage of platinum resistivity response on varying
temperatures. While basically shown in the past [2], our approach uses our recently developed
simulation approach for 3D architecture [3] to enable highly precise, freestanding tri- and tetrapod
architectures. To maximize the mechanical stability in X-Y-Z during scanning, finite element simulations
using COMSOL® have been applied and finally fabricated via FEBID. This contribution discusses the
current development state of the thermal probes ranging from simulation driven geometry
optimization over detailed 3D characterization, post-growth curing, [4] and purification towards nano-
mechanic characterization.
[1] B. Geier, C. Gspan, R. Winkler, R. Schmied, J. Fowlkes, H. Fitzek, P. Rack, H. Plank; Rapid and Highly Compact
Purification for Focused Electron Beam Induced Deposits: A Low Temperature Approach Using Electron
Stimulated H2O Reactions; Phys. Chem. Chem. Phys. (2014); 14009
[2] I.W. Rangelow, T. Gotszalk, N. Abedinov, P. Grabiec, K. Edinger; Thermal Nano-Probe; Microelectron Eng
(2001); 737
[3] J. Fowlkes, R. Winkler, B. Lewis, M. Stanford, H. Plank., P. Rack; Simulation Guided 3D Nanomanufacturing
via Focused Electron Beam Induced Deposition; ACS Nano (2016); in revision
[4] H. Plank, G. Kothleitner, F. Hofer, S.G Michelitsch., C. Gspan, A. Hohenau, J. Krenn; Optimization of Postgrowth
Electron-Beam Curing for Focused Electron-Beam-Induced Pt Deposits.; J. Vac. Sci. Tech. B Microelectronics
Nanometer Structures (2011); 051801
39
POSTER
40
41
Nanoscale studies of mechanical properties of rat bones around
biodegradable implants
Martin Meischel(1), Stefanie Stanzl-Tschegg(1), Tilman Grünewald(1), Lisa Martinelli(2), Annelie
Weinberg(2), Helga Lichtenegger(1)
(1) University of Natural Resources and Life Sciences Vienna, Institute of Physics and Materials Science, Peter-Jordan-Straße 82, 1190 Vienna, Austria
(1) Medical University of Graz, Department of Orthopedic Surgery, Auenbruggerplatz 5, 8036 Graz, Austria
Biodegradable materials made of magnesium and its alloys have been in the focus of scientist of
various disciplines during the last years. Especially in medical technology, the use of degradable
implants is becoming more and more important. This study is concerned with the mechanical
properties of rat bone around a partially degraded Mg implant. The effect of implant degradation and
bone healing on for example the modulus of elasticity and the hardness of bone is of special interest
to assess bone quality after implantation.
For this purpose, several regions around the implant are examined by means of nano-indentation. In
order to illustrate the course of implant degradation and bone regeneration, samples with varying
implant degradation time have been investigated. The implants consisted of WZ21, which is a
magnesium alloy with 2 wt.% of Yttrium [1].
Our results show that both hardness and indentation modulus decreased at the bone/implant
interface as well as in newly formed bone replacing the implant and were higher in the intact cortical
bone, as concluded from our nanoindentation maps with 20 micrometre resolution. It is also observed
that the hardness and indentation modulus change with progressive bone healing.
Figure 1: Left: Overview of the rat bone in BSC mode; Right: detail of the left bone with the values of
the equivalents Vickers hardness (kp / mm²).
[1] Gunde, P., Hänzi, A.C., Sologubenko, A.S., Uggowitzer, P.J., 2011. High-strength magnesium alloys for
degradable implant applications. Materials Science and Engineering: A 528, 1047-1054.
42
Convergent-Beam EMCD: Efficient Magnetic Measurements on the Nanoscale
Stefan Löffler(1), Walid Hetaba (2)
(1) TU Wien, USTEM & IFP, Wiedner Hauptstraße 8-10, 1040 Vienna (2) FHI-Berlin, Faradayweg 4-6, 14195 Berlin, and MPI-CEC, Stiftstraße 34-36, 45470 Mühlheim an der Ruhr,
Germany
Energy-loss magnetic chiral dichroism (EMCD) [1] is a widely available tool for studying magnetism in
the TEM. In its classical form, it employs incident and outgoing plane waves. This not only gives rise to
poor spatial resolution, but also to an infamously low signal-to-noise ratio (SNR). With the high
popularity of STEM-EELS, we investigate the theoretical possibilities and limitations of convergent-
beam EMCD.
To that end, we performed extensive multislice calculations using the mixed dynamic form factor
approach for modelling the inelastic scattering effects [2]. We investigated the dependence of both
the EMCD effect and the SNR on the convergence and collection angles for different detector positions,
namely for the “classical” position on the Thales circle through the diffraction spots, on the intersection
of the (elastic) diffraction disks, and adjacent to the (elastic) diffraction disks (see Fig. 1).
We found that the best EMCD effect and the optimal SNR could be produced for intermediate
convergence and collection angles similar to the Bragg angle. There, the SNR could be several times
larger than for the “classical” detector geometry, thus allowing for more reliable measurements and
shorter acquisition times. In addition, the fact that intermediate angles suffice allows to use this
method even in non-aberration corrected microscopes. Moreover, the spatial resolution can be
improved significantly over the traditional plane-wave approach.
Convergent-beam EMCD not only provides a better spatial resolution than “classical” EMCD, it also
features an improved efficiency and SNR. Thus, it will make EMCD measurements quicker and more
reliable, paving the way for a more widespread application of EMCD.
Figure 1: Expected EMCD effect (left) and SNR (right) for convergent beam EMCD with the collection
area touching the diffraction disks (see inset in the left panel) for 10 nm thick Fe).
[1] Schattschneider, P., Rubino, S., Hébert, C., Rusz, J., Kuneš, J. Novák, P. Carlino, E., Fabrizioli, M., Panaccione, G.
& Rossi, G. (2006) Nature, 441, 486–488.
[2] Löffler, Stefan. PhD Thesis, TU Wien 2013.
We kindly acknowledge financial support by the Austrian Science Fund (FWF): J3732-N27.
43
Vortex Filter EMCD: Towards an Alternative EMCD Approach
Thomas Schachinger (1), (2), Stefan Löffler (1), (2), Andreas Steiger-Thirsfeld (1), Michael Stöger-Pollach (1),
(2), Peter Schattschneider (2)
(1) USTEM, TU Wien, Wiedner Hauptstraße 8-10, 1040 Vienna
(2) Institute of Solid State Physics, TU Wien, Wiedner Hauptstraße 8-10, 1040 Vienna
In inelastic scattering events in magnetic materials, the target atom transfers orbital angular
momentum (OAM) in units of ħ to the probing electron. Thus, the outgoing electron wave resembles
an atom-sized electron vortex beam (EVB) carrying either plus or minus ħ of OAM. Additionally, these
spin polarized transitions show an asymmetry of the OAM transfer to the probing electron, such that
detecting the relative OAM intensities in the outgoing wavefield is a measure of the (local) magnetic
moment distribution in the sample. This is called energy-loss magnetic chiral dichroism (EMCD).
However, it has been argued that, upon elastic propagation of atomic-scale EVBs through a crystal
lattice, strong deviations to the original OAM content of an incident EVB can be observed, thus
rendering the direct detection of an EMCD signal an experimental challenge [1,2].
Traditionally, holographic vortex masks (HVMs) have been used in the condenser system of a TEM to
produce EVBs [3]. In this study, we propose an alternative EMCD setup, not relying on an
interferometric setup using the crystal itself as a beam splitter, but instead employing a HVM as a post-
specimen vorticity filter, see Fig.1 (a). We pin down experimental artefact sources, like astigmatism
and defocus variations between opposing vortex orders m = ±1, see Fig. 1 (b) and (c), and investigate
under which conditions the OAM transfer to the sample is negligible using multislice simulations. For
example, in amorphous materials the OAM exchange is relatively weak over an extended range of
specimen thicknesses, see Fig. 1 (d), enabling an EMCD signal detection, see Fig. 1 (e).
Figure 1: (a) Vortex filter EMCD principle; the EMCD effect leads to an intensity asymmetry in the
centre of two opposing vortex orders m = ±1, which are produced by a post-specimen HVM. (b)
influence of objective astigmatism on the difference signal. (c) the effect of a defocus variation
between m = ±1. (d) Simulations of an amorphous Fe-based alloy for different beam positions (see 5 x
5 nm² inset), show negligible OAM exchange of an atomic-sized vortex with the sample. (e) inelastic
simulation of the setup given in (a) for an amorphous Fe-based alloy for different numbers of scattering
atoms (10,100,1k,10k) exemplifying the effect of sample thickness and beam size on the asymmetry of
m = ±1 EVBs and thus the detectability of an EMCD effect.
[1] Löffler, S., Schattschneider, P. (2012) Acta Crystallographica Section A, 68, 443–447. [2] Rusz, J., Bhowmick, S. (2013) Physical Review Letters, 111, 105504. [3] Verbeeck, J., Tian H., & Schattschneider, P. (2010) Nature, 467, 301-304. [4] Schachinger, T., Löffler, S., Steiger-Thirsfeld, A., Stöger-Pollach, M., Schneider, S., Pohl, D., Rellinghaus, B., Schattschneider, P. (2016) Ultramicroscopy, submitted. We kindly acknowledge financial support by the Austrian Science Fund (FWF):[I543-N20, J3732-N27].
44
Flash-annealed CuZr based bulk metallic glass studied by electron microscopy methods
Christoffer Müller(1), Christian Ebner(1), Christoph Gammer(1), Konrad Kosiba(2), Benjamin Escher(2),
Simon Pauly(2), Jürgen Eckert(3), Christian Rentenberger(1)
(1) Physics of Nanostructured Materials, Faculty of Physics, University of Vienna, Vienna, AUSTRIA (2) Institute for Complex Materials, IFW Dresden, Dresden, GERMANY
(3) Erich Schmid Institute of Materials Science, Österr. Akademie der Wissenschaften, Leoben, AUSTRIA
Bulk metallic glass (BMG) is an amorphous material with no long-range order. Still, topological and
chemical short-range or medium-range order is expected to occur. To circumvent the limited ductility
of BMG, the concept of heterogeneous microstructure by forming composites has recently been used.
One route to achieve a composite structure is thermal treatment of the BMG.
Here we present the structure of flash-annealed CuZr based BMG. During the flash-annealing process
Cu44Zr44Al8Hf2Co2 samples are modified by heating to different target temperatures above glass
transition temperature (439°C) and subsequent rapid cooling in a water bath.
SEM observation of the sample heated to 642°C reveals crystallites of different size in the sample. Fig.
1a is a TEM image of a FIB lamella prepared from a single crystallite. The associated diffraction pattern
(DP) with superlattice reflections indicates the presence of the B2 ordered structure. It is interesting
to note that in crystalline CuZr based materials, devitrified from the amorphous structure, Cu10Zr7 and
CuZr2 structures are expected to occur.
To obtain structural information of the amorphous phase of the as-cast state and flash-annealed
samples (heated to 540°C, 619°C, 642°C) variable resolution dark field (DF) fluctuation electron
microscopy was applied. Tilted DF images show structural correlations in form of speckles in the image
(Fig. 1b). By varying the objective aperture size the normalized variance of the intensity as a function
of spatial resolution can be determined. This can be used to calculate the correlation lengths of the
differently treated samples (Fig. 1c).
Figure 1: a) TEM image of single crystallite embedded in the amorphous structure. The DP containing
superlattice reflections indicates the formation of B2 structure by flash-annealing up to 642°C. b) Tilted
DF image showing intensity variations as a result of local structural correclations. c) The correlation
length of the medium range order from the amorphous structure increases with the target
temperature of the flash-annealing treatment.
We kindly acknowledge financial support by the Austrian Science Fund (FWF): [I1309, J3397].
45
Understanding surface enhanced Raman spectroscopy using accurate
simulations of electric nearfields
Harald Fitzek(1), Jürgen Sattelkow(1), Peter Pölt (1,2)
(1) Institute of Electron Microscopy and Nanoanalysis, Graz University of Technology, NAWI Graz, Steyrergasse 17, 8010 Graz
(2) Graz Centre for Electron microscopy, Steyrergasse 17, 8010 Graz
Surface enhanced Raman spectroscopy (SERS) is a powerful technique that uses metal nanostructures
(Au, Ag, Cu) to gain single molecule sensitivity at best [1]. Most of the enhancement is attributed to
the enhancement of the electric field near the surface of the metal, because the Raman signal in this
case scales approximately proportional to the fourth power of the electric field strength [2]. Since the
enhancement of the electric nearfield is frequency dependent, the enhancement of the Raman signal
can be different for the various Raman bands of the absorpt molecule. This can lead to significantly
different band intensities from the same molecule positioned on different substrates. Therefore a
detailed specification of electric nearfields is crucial for the understanding of surface enhanced Raman
spectra.
We were using the AFM to precisely determine the geometry of Au nanostructures on flat substrates.
From the geometry we calculated the electric near fields using a homemade Matlab implementation
of the discrete dipole approximation (DDA). The DDA was chosen because it relies on no assumptions
other than the target geometry and the dielectric functions of the materials involved [3]. An exemplary
calculation is shown in figure 1. Simultaneously we performed measurements of the enhancement
factors for all the Raman-bands of a test molecule, in order to determine how accurate our predicted
enhancement factors are.
Figure 1: (left, bottom) AFM image of an Au-island film on a silicon wafer; (left, top) Dipole representation of
the Au-island film (blue dots) and the Si-Substrate (red dots) used for the simulation; (right) Simulated E4-
approximation of the local enhancement of the Raman signal. The colourbar is plotted using a logarithmic
scale.
[1] Xu, H., Bjerneld, E.J., Käll, M. and Börjesson, L. (1999), Spectroscopy of Single Hemoglobin Molecules by
Surface Enhanced Raman Scattering, Phys. Rev. Lett. 83, 4357
[2] Garcia-Vidal, F.J. and Pendry, J.B. (1996), Collective Theory for Surface Enhanced Raman Scattering, Phys. Rev.
let. 77, 1163
[3] Yurking, M.A. and Hoekstra, A.G. (2007), The discrete dipole approximation: an overview and recent
developments, Journal of Quantitative Spectroscopy and Radiative Transfer, 106(1), 558-589
46
Preparation of Transmission Electron Microscopy Samples by Mechanical
Techniques in Combination with Low-Voltage Ion Milling
Cornelia Trummer(1), Martina Dienstleder(2), Gerald Kothleitner(1)
(1) Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology, Steyrergasse 17, 8010
Graz, Austria
(2) Graz Centre of Electron Microscopy, Steyrergasse 17, 8010 Graz, Austria
For decades, samples for Transmission Electron Microscopy (TEM) analysis have been produced by
standard mechanical pre-preparation like cutting, grinding, polishing, and dimpling with a subsequent
broad argon ion milling step. [1] But the progressive development of high-performance TEM needs a
development of the sample preparation as well because it is one of the limiting factors for high-end
TEM analysis down to the atomic scale. To generate a sample, which is only a few atom layer thin and
with the least preparation artefacts, the thinning of the samples with focused low energy argon ions
(such as provided by the NanoMill® instrument by Fischione) seems to be a viable way.
Daily practise shows, upon first examination in a TEM, that it is very complicated to perform a low
energy argon ion milling of defined areas, especially on samples pre-prepared classically. Therefore,
we developed an elaborated preparation and analysis procedure as described in this study. To test
this procedure, low-voltage milling is executed on different mechanically preprepared samples like
planar, dimpled, and cross-sectioned samples. The resulting thickness reduction is measured via
energy filtered TEM (EFTEM; t/λ method [2]).
Based on the results of this study two different preparation and examination procedures are given
specifically for the Fischione NanoMill® to thin and clean mechanically pre-prepared samples.
Figure 1: Relative thickness map of a certain sample position of a mechanically pre-prepared Si-disc
before (left) and after (right) milling with the NanoMill®. By comparison, the thinning effect is clearly
visible, showing that the elaborated preparation and analysis procedure works.
[1] Jeanne Ayache, Hrsg., Sample preparation handbook for transmission electron microscopy: techniques
(New York: Springer, 2010). [2] Egerton, R. F. (2009) Reports on Progress in Physics, 72, 016502
47
In situ studies of high-purity mono- and bimetallic nanostructures in
experiment and simulation
Daniel Knez(1,2), Martin Schnedlitz(3), Maximilian Lasserus(3), Gerald Kothleitner(1,2), Andreas W.
Hauser(3), Wolfgang Ernst(3), Ferdinand Hofer(1,2)
(1) Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology, Graz, Austria (2) Graz Centre for Electron Microscopy, Graz, Austria
(3) Institute of Experimental Physics, Graz University of Technology, Graz, Austria
For the application of metallic nano structures, knowledge about their thermodynamic properties and
thermal stability is of paramount importance[1]. To this end, we performed in situ heating experiments
using metallic nanowires and bimetallic clusters using aberration corrected scanning transmission
electron microscopy (STEM).
The nanoparticles used for our investigations were synthesized within superfluid helium
nanodroplets (composed of 103 to 1010 helium atoms), under ultra-high vacuum (UHV) conditions[2].
This approach provides exceptional advantages over conventional methods, like sequential
addition of a wide range of materials without the use of ligands and stabilizers. Thus, nanoparticles
can be synthesized with any composition and different structures, with extremely high purity,
which cannot be achieved by other known methods [3].
For in situ heating experiments, we used a DENSsolutions Wildfire D6 holder in a probe corrected FEI
Titan3 60-300 microscope. This microscope is equipped with a Super-X detector (EDX) and a Gatan
Quantum energy filter for EELS.
This enables us to study phenomena such as Rayleigh breakup of nanowires (Figure 1a) as well as
alloying of Ni@Au core-shell clusters on the atomic scale (Figure 1b).
Figure 1: (a) Au nanowire in its initial state and after breakup at 300°C, with corresponding molecular
dynamics simulation results. The simulation correctly predicts the regions where the wire starts to
segregate (red circles). (b) Two clusters in their initial state at room temperature. The darker Ni core
can clearly be distinguished from the bright Au shell. At a temperature of 400°C, all particles were
found to be alloyed.
[1] F. Baletto, R. Ferrando, Rev. Mod. Phys. 2005, 77, 371.
[2] A. Volk, P. Thaler, M. Koch, E. Fisslthaler, W. Grogger, W. E. Ernst, J. Chem. Phys. 2013, 138, 214312.
[3] P. Thaler, A. Volk, F. Lackner, J. Steurer, D. Knez, W. Grogger, F. Hofer, W. E. Ernst, Phys. Rev. B 2014, 90.
We kindly acknowledge financial support by the Federal Ministry of Science, Research and Economy with the
project “Infrastrukturförderung 2015” and by the Austrian Research Promotion Agency (FFG) in the project
SOLABAT (853627).
48
Polymer fracture – What can the 3D reconstruction of the crack region tell us about the microscopic fracture mechanisms
Manfred Nachtnebel (1), Claudia Mayrhofer (2), Armin Zankel (1,2), Peter Pölt (1,2)
(1) Institute for Electron Microscopy and Nanoanalysis, NAWI Graz, TU Graz, Steyrerg. 17, 8010 Graz (2) Graz Centre for Electron Microscopy, Steyrerg. 17, 8010 Graz
Polypropylene (PP) is a widely used thermoplastic polymer and its various modifications make it
interesting for even highly innovative applications [1]. For the investigation of their fracture behaviour,
often tensile tests are used. Subsequently the fracture surfaces are investigated by various microscopic
methods. But this gives only limited information about the microscopic fracture mechanisms taking
place during the test. To overcome this limitation the 3D reconstruction of the cracks developing in
the polymer blends is necessary. For this purpose an in situ ultramicrotome 3ViewTM (Gatan Inc.,
Pleasanton, USA) mounted in the specimen chamber of an environmental scanning electron
microscope (ESEM) Quanta FEG 600 (FEI, Eindhoven, NL) was used to perform serial sectioning and
imaging [2]. This method enables the 3D reconstruction of the cracks and the modifier particles inside
the PP matrix, providing information about the crack surfaces and the position of the cracks related to
the distribution of the particles.
Tensile tests were stopped at a predefined force at around 25 % or 50 % yield. Subsequently part of
the fracture region was extracted and sectioned and imaged by serial block-face scanning electron
microscopy (SBEM). To create a 3D reconstruction of the fracture surfaces and the particles, the
images in the stacks had to be noise filtered and then segmented to make the different phases and
structures distinguishable. After the segmentation the cracks, EPR particles and PP matrix can be
reconstructed.
Examples of 3D reconstructions are shown in Figure 1, and they disclose detailed information about
the microscopic fracture mechanisms in the particle modified PP. It is clearly visible that different
responses to the tensile force appear for the ethylene propylene rubber (EPR) modified PP, and PP
modified with linear low density polyethylene (LLDPE). While widespread cracks inside the PP matrix
were formed, in the EPR modified samples, see Figure 1 a), in case of the LLDPE modified samples voids
were formed in only small bands perpendicular to the applied force, see Figure 1 b).
Figure 1: 3D reconstruction of the fracture region of PP modified with a) EPR (particles green, cracks
blue) and b) LLDPE (particles = green, voids = red) after the tensile tests were stopped at 50 % yield.
[1] G. Grestenberger, G. D. Potter, C. Grein, (2014) Express Polym. Lett., vol. 8, no. 4, 282–292, [2] A. Zankel, J. Wagner, P. Pölt, (2014), Micron, vol. 62, pp. 66–78, Jul. 2014.
49
Challenges in sample preparation for HRSTEM analysis
Martina Dienstleder (1), Gerald Kothleitner (2)
(1) Graz Centre of Electron Microscopy, Steyrergasse 17, 8010 Graz, Austria (2) Institute for Electron Microscopy and Nanoanalysis, TU Graz, Steyrergasse 17, 8010 Graz, Austria
The advent of fully corrected ultra-high-resolution transmission electron microscopes (TEM) enabled
access to the sub-atomic scale and has become paramount for innovations in material design relevant
to science and technology. Their vastly improved performance for the characterization of materials is
accompanied by the need to generate high-quality samples via sophisticated and innovative sample
preparation techniques. Specimen preparation [1-3] hence is the key to obtain representative insights
into morphology, structure, chemistry and functionality. However, to exploit the full potential of high-
end TEM instrumentation towards the sub-atomic scale, classical approaches and routine preparation
protocols are not sufficient. Even minor preparation artefacts such as thin amorphization layers, ion
implantation, selective milling or re-deposition can already prevent a deeper insight to a material [4-
6]. To overcome these limitations, it is first necessary to understand artifact origins, and then improve
on sample preparation procedures, which often necessitate special sequences and unusual
preparation steps.
In this light, the talk will give a short overview of current challenges in sample preparation for high-
resolution electron microscopy analysis and then shows up STEM applications that could not have been
conceived before.
Figure 1: a) FIB image of with standard parameter prepared TiO2 sample; b) STEM dark field image of
the twin boundary in a; c) Comparison of the implantation of Ga by EDX measurements before and
after improved preparation procedure; d) STEM dark field image of the lamella after improved
preparation procedure.
[1] Ayache, J., Beaunier, L., Boumendil, J., Ehret, G., & Laub, D. (2010). Sample preparation handbook for
transmission electron microscopy: techniques (Vol. 2). Springer Science & Business Media.
[2] G. Petzow “Metallographisches Keramographisches Plastographisches ÄTZEN”
[3] L. Reimer „Elektronenmikroskopische Untersuchungs- und Präparationsmethoden (1967)
[4] McCaffrey et al., (2001) Surface damage formation during ion-beam thinning of samples for transmission
electron microscopy, Ultramicroscopy 87, pp. 97–104.
[5] A. Barna et al., (1998) Ultramicroscopy, 70, p. 161
[6] C.-M. Park et al., (2004) Measurement of Ga implantation profiles in the sidewall and bottom of focused-ion
beam-etched structures Applied Physics Letters 84, 3331; DOI: 10.1063/1.1715142
50
Investigation of the non-equilibrium formation of stoichiometric precipitates
in multi-component aluminium alloys
Angelina Orthacker(1,2), Georg Haberfehlner(1), Johannes Tändl(3), Maria C. Poletti(3), Bernhard Sonderegger(3) and Gerald Kothleitner(1,2)
(1) Graz Centre for Electron Microscopy, Graz, Austria (2) Institute of Electron Microscopy and Nanoanalysis, Graz University of Technology, Graz, Austria
(3) Institute of Materials Science and Welding, Graz University of Technology, Graz, Austria
A detailed understanding of the entire precipitate forming process is necessary to ensure best
applicability and use of precipitation hardened alloys. The combination of different electron
microscopic methods, such as high resolution scanning transmission electron microscopy (HR-STEM)
linked to frozen phonon multi-slice simulations and analytical electron tomographic techniques in the
form of 3D resolved energy dispersive X-ray (EDX) voxel spectroscopy [1], enables very detailed
insights into precipitate forming processes. In case of an industrially cast aluminium alloy
(AlMg4Sc0.4Zr0.12) these insights at unprecedented spatial resolution overthrew current views [2]
that the resulting core-shell-structure of the precipitates is a direct result of the different diffusion
rates of the precipitate forming species - scandium (Sc) and zirconium (Zr) - in the matrix. Electron
microscopy based experimental results supported by thermodynamic calculations and locally
resolved 2D diffusion simulations revealed that the finally stable precipitate is determined by a
dynamic interaction of Gibbs energy, ordering of phases and the activation energy for the necessary
jump cycles. Therefore this study shows that deviations from equilibrium conditions found in
precipitates of multicomponent alloys may be explained by a space- and time-resolved investigation
of composition, order and resulting activation and Gibbs energies of the system. With the recent
developments in electron microscopy the necessary detail to perform such investigations is offered
and new insights on precipitate forming processes are enabled.
Figure 1: High resolution STEM image of a short aged (5 min at 500°C) precipitate revealing a matrix
like channel with single column width in the Sc/Zr sublattice allowing 1D diffusion of Sc and Zr
[1] G. Haberfehlner et al., Nanoscale 6 (2014), p. 14563.
[2] E. Clouet et al., Nature Materials 5 (2006), p. 482.
The authors thank the Austrian Cooperative Research Facility, the Federal Ministry of Science, Research and
Economy with the project “Infrastrukturförderung 2015” and the Austrian Research Promotion Agency FFG (TAKE
OFF project 839002) for funding.
51
High resolution episcopic microscopy (HREM): a tool for 3D imaging of organic
materials
Stefan H. Geyer, Wolfgang J. Weninger
(1) Medical University of Vienna, Center for Anatomy and Cell Biology & MIC, Währinger Str. 13, 1090 Vienna
“High resolution episcopic microscopy” (HREM) is a post-mortem 3D imaging technique gaining stacks
of images of resin embedded samples by capturing the block surface during physical sectioning. The
quality of single HREM images nearly matches the quality of images captured from hematoxylin/eosin
stained histological sections. Typical HREM data sets comprise stacks of 2,000 to 3,500 inherently
aligned digital images and have voxels sizes of 1 x 1 x 1 µm3 to 3 x 3 x 3 µm3. Digital data created from
mouse embryos had volumes of approximately 8 x 6 x 11 mm3 and voxel sizes of 3 x 3 x 3 µm3.
HREM has a broad spectrum of potential applications and was already used for examining various
organisms and materials. Besides researching the topology and morphology of organs and tissues of
various biomedical model organisms and human tissue samples, it was extensively used for
systematically phenotyping mouse embryos in the scope of the “Deciphering the Mechanisms of
Developmental Disorders” (DMDD) project. This project was launched aiming at providing phenotype
information of embryos of mutant embryonic lethal mouse lines harvested at embryonic day (E)14.5.
In our presentation we will demonstrate the field of applications of HREM and focus on the usability
of HREM data created for phenotyping E14.5 embryos created in the DMDD project.
Our presentation, briefly demonstrates the quality of HREM data generated of mouse, chick, quail,
frog and zebrafish embryos, human tissue materials and other organic materials. It focuses on the
advantages of HREM data for scoring the phenotype of E14.5 mouse embryos with gene deletions
produced in the DMDD project. The results demonstrate that HREM is an optimal technique for
visualising the morphology of a broad spectrum of organic materials and for scoring the phenotype of
genetically engineered mouse embryos.
Figure 1: Volume rendered 3D model of an E14.5 mouse embryo. Scale bar 1 mm.
52
The Electronic Phase Diagram of YSZ
Thomas Götsch(1), Alexander Menzel(1), Erminald Bertel(1), Michael Stöger-Pollach(2), Simon Penner(1)
(1) University of Innsbruck, Institute of Physical Chemistry, Innrain 80–82, 6020 Innsbruck (2) Vienna University of Technology, USTEM, Wiedner Hauptstraße 8–10, 1040 Vienna
Yttria-stabilized zirconia (YSZ) is one of the most frequently employed electrolytes in chemical sensors
or fuel cells because of its oxygen ion conductivity originating from oxygen vacancies due to the Y3+
doping of ZrO2. In solid oxide fuel cells, it is used as both the electrolyte as well as part of an anode (in
a Ni cermet). In the electrolyte, a high impedance is required, whereas a high electronic conductivity
would be favourable in the anode to obtain higher current densities. To achieve that by means of band
gap engineering, the electronic structure around the band gap needs to be known. Thus, we present a
systematic study of various electronic properties as a function of the Y2O3 concentration. For that,
electron-transparent, unsupported thin films were prepared by direct current ion beam sputtering.[1]
Figure 1: a) The unit cell height of the specimens is plotted as a function of the yttria content, revealing
the phase transition between 8 and 20 mol% Y2O3.[1] In b), the direct and indirect band gaps are
shown, exhibiting the same behaviour as the lattice parameter.
In Figure 1a, the lattice parameter c, as determined by electron diffraction, is displayed as a function
of the Y2O3 concentration in the samples. The retention of this parameter between 8 and 20 mol%
yttria can be attributed to a phase transformation between tetragonal and cubic YSZ.[1] In a later
study, this was confirmed to actually lie between 8 and 9.4 mol% Y2O3.[2] The same shape can be seen
for the band gaps, as obtained by VEELS and UPS (Figure 1b), as well as for a multitude of other
properties, such as the electron affinities or the optical properties, as calculated by Kramers-Kronig
analysis. Furthermore, the electron affinities are revealed to be negative, indicating that the vacuum
level lies below the conduction band, i.e. electrons that are excited by optical absorption can be
emitted with almost no energy barrier. This could lead to new applications, for instance regarding spin-
polarized electron microscopy or dynamic TEM.
Additionally, a small overview of the research activities of our research group, the Nanostructured
Model Catalyst Group Innsbruck,[3] is given, with a focus on selected applications utilizing microscopy
and electron spectroscopy methods.
[1] Götsch, T., Wallisch, W., Stöger-Pollach, M., Klötzer, B., Penner, S. (2016) AIP Advances, 7, 025119.
[2] Götsch, T., Schachinger, T., Stöger-Pollach, M., Kaindl, R., Penner, S. (2017) Applied Surface Science, 402, 1–
11.
[3] http://webapp.uibk.ac.at/physchem/nmci/
We kindly acknowledge financial support by the Austrian Science Fund (FWF) via grant F4503-N16.
53
Lamellae in FeAl deformed under hydrostatic pressure
Stefan Noisternig(1), Christian Rentenberger(1), Christoph Gammer(2),
Gerlinde Habler(3), H Peter Karnthaler(1)
(1) Universität Wien, Physik Nanostrukturierter Materialien, Boltzmanngasse 5, 1090 Wien (2) OEAW Leoben, Erich Schmid Inst. of Materials Science, Jahnstraße 12, 8700 Leoben
(3) Universität Wien, Dept. für Lithosphärenforschung, Althanstraße 14, 1090 Wien
We severely deformed single crystalline FeAl of B2 structure by the method of high-pressure torsion
(HPT). The samples have a diameter of 8 mm and a thickness of ≈ 0.7 mm. Two different early states
of deformation were compared: first, compression only, to obtain a hydrostatic pressure of 8 GPa and
second, HPT with a small amount of torsion at 8 GPa.
The SEM analysis shows for both states of deformation structures formed by lamellae of alternating
orientations and widths (≈ 25 µm) (cf. Fig. 1). Therefore, applying hydrostatic pressure already changes
the single crystalline structure to a lamella microstructure. Using electron backscatter diffraction the
orientations of the lamellae were analyzed leading to the result that the lamellae are separated by tilt
angle boundaries. The sample deformed by HPT exhibits higher misorientation angles between the
lamellae than the sample deformed by compression only.
By TEM a plain view specimen and a FIB cross-section specimen of the sample deformed by HPT were
analyzed. Fig. 2 shows a TEM image revealing sharp and narrow structures of lamella (≈ 50 nm in
width). They occur in addition to the structures shown in Fig. 1. The three dimensional orientation of
the sharp lamella was obtained from the two specimens prepared in perpendicular orientations.
Up to now there are no reports of deformation twinning in B2 FeAl. By comparing the possible twinning
relations for B2 structures resulting from theoretical reasoning [1] with the measured misorientations,
a twinning relation between the sharp lamellae could not be identified directly. This might be due to
an additional deformation process. To analyze this in detail, different states of deformation will be
compared by TEM.
Figure 1: SEM image of the lamella structure
using backscatter detector.
Figure 2: TEM bright-field image of sharp
narrow lamellae.
[1] Christian, J. W., Laughlin, D. E. (1988) Acta Metallurgica, 36, 1617-1642
54
Preparation Methods of Biological Samples: a Comparison of Chemical Fixation and High-Pressure Freezing (HPF)
Katharina Keuenhof (1), Marlene Brandstetter (1), Thomas Heuser (1)
(1) Electron Microscopy Facility, Vienna Biocenter Core Facilities, Dr.-Bohr-Gasse 3, 1030 Vienna
Electron microscopy (EM) is indispensable when it comes to the analysis of ultrastructural elements
of biological material. However, the right preparation is necessary to obtain an accurate
representation of cell physiology. Chemical fixation can be considered the most classical approach.
Fixatives are used to crosslink proteins as well as lipids to retain the original structure and stains are
added for increased contrast. It is a relatively easy and cost-effective procedure. On the other hand,
high-pressure freezing offers a more physical approach to structural preservation. Exertion of a very
high pressure hinders the expansion of ice crystals and thus the damage to cellular material. The
sample subsequently undergoes a process called freeze substitution in which frozen water is
replaced by liquid solvents and fixatives while slowly warming up to 0 C. The optimal method of
preparation varies by sample and structure of interest. To establish a comparison, tissue and cell
cultures from various organisms were prepared and the quality of the ultrastructure compared. One
of the specimens observed was the cyanobacterium Arthrospira fusiformis. The high-pressure frozen
sample shows more detail of the individual cell components. The thylakoids are much better
preserved and cristae can also be observed more clearly. Additionally, contrast is improved in the
HPF sample. Although chemical fixation provides easier handling and fewer variables, high-pressure
freezing showed improved preservation for most samples.
Figure 1: Method comparison on cells of the cyanobacterium Arthrospira fusiformis. Left shows the chemically
fixed sample, right the high-pressure frozen and freeze-substituted one. The form of the thylakoids depends on
the orientation of the sample. Magnification of 44000x, scale bar represents 1µm.
55
Quantitative Analysis of Internal Interfaces Structural and quantitative analysis via High resolution STEM
Evelin Fisslthaler(1), Christian Gspan (1), Georg Haberfehlner (1) , Werner Grogger (2)
(1) Graz Centre for Electron Microscopy, Steyrergasse 17, 8010 Graz (2) Institute for Electron Microscopy and Nanoanalysis, Graz University of Technology, Steyrergasse
17, 8010 Graz
The scope of the research project “Quantitative analysis of internal interfaces” is the high resolution
analysis of internal interfaces in multilayer materials for electronic devices via aberration corrected
STEM combined with HR EELS and EDX.
For this purpose, a variety of different approaches for both data acquisition and data analysis is
consequently refined to provide reliable and reproducible datasets with high accuracy in both spatial
and energetic resolution as well as in terms of quantitative reliability. Concomitantly, TEM sample
preparation methods were sufficiently enhanced and modified to provide specimens with adequate
quality.
One of the key topics of the project is the detection and analysis of interfacial layers with few- or even
sub-monolayer dimension in silicon based materials. The transition area between Si and SiO2 is of
major interest, since the properties as well as the extent of this region can be crucial for device
performance. Therefore, the ELNES signals of both silicon and oxygen are traced with high spatial and
energetic resolution to yield detailed information about the chemical composition of the few atomic
layers that form the interface between the two materials. In order to obtain the contributions of
individual silicon oxidation states from the ELNES signal, various signal optimization and fit procedures
are used. These procedures are then applied for several silicon based material systems to determine
the exact composition of the transition region in order to advance material fabrication procedures for
silicon based electronic devices.
Figure 1: Schematic illustration of transition region between Si and SiO2 (a) [1], HR-EELS of SiL-Edge in the transition area (b) [2], high resolution STEM HAADF images of the interface between silicon and SiO2 (optimized for spatial resolution (c), optimized for both spatial and energetic resolution (d))
[1] Oh, PhysRevB (2001) 205310.
[2] Batson, Nature (1993) 366:727
We kindly acknowledge financial support by the Austrian Research Promotion Agency (FFG) (project
850220/859238).
(b) (c) SiO2
Si
SiO2
Si
(d)
56
In-situ electron microscopy for heterogeneous catalysis
Walid Hetaba (1,2), Marc-Georg Willinger (2,3), Robert Schlögl (1,2)
(1) Max-Planck-Institute for Chemical Energy Conversion, Stiftstraße 34-36, 45470 Mülheim an der Ruhr (2) Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, 14195 Berlin
(3) Max-Planck-Institute of Colloids and Interfaces, Am Mühlenberg 1 OT Golm, 14476 Potsdam
Analytical electron microscopy is a versatile tool for the investigation of catalysts. It allows for imaging
the structure as well as analysing the elemental composition at the same time. However, conventional
electron microscopy is limited to performing investigations near room temperature and in the pressure
regime of the column vacuum. In order to investigate a catalyst in its active state, in-situ electron
microscopy is applied. Usually, a thorough investigation of a catalytic material includes a combination
of conventional and in-situ microscopy as well as performing simulations in order to interpret the
acquired data. In this work we show the investigation of iridium oxides as an example for this approach
and focus on the simulation and interpretation of electron energy-loss spectra.
Iridium oxides and hydroxides are promising catalysts for water splitting due to their high stability and
activity. [1] Figure 1a) shows a HRTEM image of an iridium catalyst used for electrochemical
investigation. In Figure 1b) the simulated oxygen K-edge ELNES for different iridium oxide structures
can be seen. The simulations confirm the results of experimental EELS measurements and pair
distribution function analysis of previous work. [2] This suggests that the amorphous IrOx-hydroxide
synthesized at the FHI is best described by hollandite structural motifs. Additionally, the calculated
spectra allow to distinguish the contributions of each atom at different crystallographic positions to
the total spectrum. This improves the understanding of the structure-property relationship of these
highly promising catalyst materials.
Figure 1: a) TEM micrograph of a polycrystalline iridium catalyst on a Si-substrate used for
electrochemical investigation. b) Oxygen K-edge ELNES calculation for four different iridium oxide
structures.
[1] Massue et al., ChemSusChem, DOI: 10.1002/cssc.201601817.
[2] Willinger et al., in preparation.
57
List of participants
Universität für Bodenkultur, Wien
Meischel, Martin
Plewka, Jacek
Ploszczanski, Leon
Pum, Dietmar
Siedlaczek, Philipp
Montanuniversität Leoben
Gao, Jinming
Haselmann, Ulrich
Issa, Inas
Kromout, Karoline
Völker, Bernhard
Zeiler, Stefan
Zhang, Zaoli
Fritz-Haber-Institut Berlin
Hetaba, Walid
Bio Center, Wien
Brandstetter, Marlene
Fellner, Nicole
Heuser, Tom
Jacob, Sonja
Keuenhof, Katharina
Kotisch, Harald
Resch, Günther
Serwas, Daniel
Wozelka, Lisa
IST Austria, Klosterneuburg
Borges-Meranje, Carolina
Gütl, Daniel
Kleindienst, David
Montanaro-Punzengruber, Jacqueline
Zheden, Vanessa
Johannes Keppler Universität, Linz
Groiß, Heiko
Truglas, Tia
Medizin-Universität Graz
Leitinger, Gerd
Sele, Mariella
Wernitznig, Stefan
Medizin-Universität Wien
Dibiasi, Christoph
Ellinger, Adolf
Geyer, Stefan
Hubert, Virginie
Kain, Renate
Pavelka, Margit
Schulz, Stefan
Technische Universität Brünn
Horak, Michal
Technische Universität Graz
Dienstleder, Martina
Fisselthaler, Eveline
Fitzek, Harald
Hofer, Ferdinand
Knez, Daniel
Nachtnebel, Manfred
Orthacker, Angelina
Sattelkow, Jürgen
Striemitzer, Robert
Trummer, Cornelia
Winkler, Robert
Technische Universität Wien
Bernardi, Johannes
Löffler, Stefan
Fidler, Josef
Pfeiffer, Stefan
Schachinger, Thomas
Schattschneider, Peter
Schwarz, Sabine
Shawrav, Mostafa
Steiger-Thirsfeld, Andreas
Stöger-Pollach, Michael
Wallisch, Wolfgang
Wojcik, Tomasz
Universität Innsbruck
Götsch, Thomas
Holzinger, Andreas
Oberwegser, Sabrina
Salvenmoser, Willi
58
Universität Graz
Schmidt, Franz-Philipp
Stabentheiner, Edith
Zellnig, Günther
Universität Salzburg
Herbst, Markus
Lütz-Meindl, Ursula
Minnich, Bernd
Steiner, Philipp
Zickler, Gregor
Universität Wien
Ebner, Christian
Eckhard, Margret
Goldhammer, Helmuth
Habeler, Gerlinde
Karnthaler, Peter
Meyer, Jannik
Müller, Christoph
Noisternig, Stefan
Reipert, Siegfried
Rentenberger, Christian
Tulic, Semir
Waitz, Thomas
ASEM - Firmenmitglieder:
AMETEK
Jung, Matthias
Christine Gröpl
Gröpl, Christine
Gröpl, Leopold
Diatome
Jenke, Martin
FEI
Dubovy, Klaus
Lich, Ben
GATAN
Kastenmüller, Andreas
Schweitzer, Michael
GeTec
JEOL
Raggl, Georg
Leica
Ranner, Robert
Wenger, Alfred
Science Services GmbH
Stefan Schöffberger
Videko - HITACHI
ZEISS
Perez, Fabian
Schwinger, Wolfgang
59
www.fei.com
www.gatan.com
www.jeol.de
www.zeiss.at
www.leica-microsystems.com
www.getec-afm.com
www.christine-groepl.com
www.videko.at
www.ametek.de
www.diatome.ch
60