Photo by ©FAU/Erich Malter

OPEN FAU

Online publication system of Friedrich-Alexander-Universität Erlangen-Nürnberg

The online publication system OPEN FAU is the central publication platform for Open Access publishing for all members of Friedrich-Alexander-Universität. Qualified works from research and teaching may be published here free of charge, either as a primary or secondary publication. The full texts are permanently available worldwide and are findable and citable via catalogues and search engines.


To search for documents in OPEN FAU, please select "Search" (via the magnifying glass at the top right); this will provide you with various search options. If you want to publish a document, go to "Login" and "My Publications". Then drag you document into the field provided and enter the metadata. In just a few steps, you can submit your document. Please note our guidelines, the publication contract and FAQs.

 

Recent Submissions

Doctoral thesis
Open Access
Immunmonitoring während Transarterieller Chemoembolisation beim hepatozellulären Karzinom und Korrelation mit dem klinischen Therapieerfolg
(2024) Schmid, Jonas; Kremer, Andreas
Objectives Hepatocellular carcinoma (HCC) is one of the most common carcinomas worldwide and one of the leading causes of cancer death. Overall, despite new therapeutic options, there is still a poor prognosis with a wide individual variation in response to different forms of therapy. Causes of the varying response to therapy in individual patients remain unclear. Immunological changes that are influenced by therapy and are conducive to successful treatment could be contributing factors. Interventional therapies should be emphasized in this context, as they induce rapid tumor cell death, which leads to a strong acute phase reaction with associated release of inflammatory mediators. These can trigger further immunological processes that presumably stimulate or even initiate tumor-specific defense mechanisms. For this reason, we performed immune monitoring periinterventionally to transarterial chemoembolization (TACE). The aim of this study was to analyze different immune parameters in relation to the response to therapy and to draw conclusions about the importance of individual cell groups and the associated signaling pathways in regard to the tumor-specific immune response. Furthermore, new prognostic markers should be identified. Design and Methods Twenty-one HCC-patients treated with TACE at University Hospital Erlangen were included. Plasma, serum and peripheral blood mononuclear cells (PBMC) were obtained immediately before TACE and on the two subsequent days. Clinical data were collected pseudonymized from the electronic medical record. This included parameters such as etiology, tumor diameter, tumor focus number and clinical scores. In addition relevant laboratory values such as transaminases, cholestasis levels and alpha-1-fetoprotein (AFP) were collected. The obtained patient samples were analyzed by flow cytometry and enzyme-linked immunosorbent assay (ELISA). The percentage of different lymphocyte subpopulations with their activation and proliferation levels and corresponding cytokines were determined. For evaluation, patients were divided into two groups according to progression-free survival (PFS). We looked for relevant correlations among the recorded and determined parameters with each other and with progression-free survival by using Mann-Whitney U test, Wilcoxon signed-rank test and Kaplan-Meier estimator. Observations and Results The 21 included subjects were on average 74.1 (±10.0) years old, 15 (71.4%) were male. Liver cirrhosis was present in a majority (17 subjects, 81.0%), of which the Child-Pugh score was A in 12 patients (66.7%), B in 5 (27.8%) and C in one (5.6%). The Model for End-Stage Liver Disease (MELD)-score averaged 11.7 (±4.8). The etiology of liver cirrhosis was ethyltoxic in 7 patients (33.3%), hepatitis C in 3 (14.3%), ethyltoxic in combination with hepatitis C and ethyltoxic in combination with nonalcoholic steatohepatitis (NASH) one each (4.8%) and primary biliary cholangitis (PBC) in 2 (9.5%). In seven subjects (33.3%) etiology was unclear. Six subjects had Barcelona Clinic Liver Cancer Staging System (BCLC) A (28,6%), 14 had BCLC B (66,7%) and one had BCLC D (4,8%). The median tumor diameter was 42.3 mm (±17.6 mm) and the median AFP level was 421.7 ng/ml (±657.1 ng/ml). Median PFS amounted 83 days and median overall survival was 207 days. A good response to therapy was considered a progression-free survival of more than 180 days, which was present in 16 subjects (76.2%). Significant changes at day one and two after TACE compared to baseline levels were seen in several parameters. There was a significant decrease in total lymphocyte count (baseline: 51.57 ±13.95%, day 2: 37.11 ±15.47%, z = 3.148, p = 0.002), CD4+ T-cells (baseline: 58.76 ±18.14%, day 2: 55.17 ±18.75%, z = 2.107, p = 0.035), Ki67+ CD8+ T-cells (baseline: 2.80 ±1.94%, day 2: 2.25 ±1.83%, z = 2.675, p = 0.008) and AFP (baseline: 421.71 ±657.12 ng/mL, day 2: 413.76 ±644.51 ng/mL, z = 2.417, p = 0.016). A significant increase was observed in total leukocyte count (baseline: 4.92 ±1.80 thousand/µL, day 1: 6.43 ±2.48 thousand/µL, z = -2.556, p = 0.011; day 2:7.74 ±3.74 thousand/µL, z = -2.589, p = 0.001), glutamate oxaloacetate transaminase (GOT) (baseline: 63.19 ±29.00 U/L, day 1: 154.95 ±134.47 U/L, z = -3.659, p < 0.001; day 2: 357.90 ±413.64 U/L, z = -3.464, p = 0.001), glutamate pyruvate transaminase (GPT) (baseline: 40.38 ±23.34 U/L, day 1: 67.15 ±48.55 U/L, z = -3.698, p < 0.001; day 2: 142.25 ±155.36 U/L, z = -3.517, p < 0.001), interleukin (IL)-6 (baseline: 8.60 ±11.48 pg/mL, day 1: 26.96 ±26.18 pg/mL, z = -3.678, p < 0.001; day 2: 151.29 ±186.08 pg/mL, z = -3.516, p < 0.001), C-reactive protein (CRP) (baseline: 12.97 ±21.08 mg/L, day 2: 42.81 ±49.79 mg/L, z = -2.192, p = 0.028) and CD8+ terminal differentiated T effector memory cell (Temra) (baseline:47.77 ±22.70%, day 2: 51.01 ±24.16%, z = -2.059, p = 0.040) were observed. However, the extent of change did not correlate with treatment response. There were no significant changes at day 1 or 2 after TACE compared with baseline levels in IL-1β, IL-8, IL-17A, transforming growth factor β 1 (TGF-β1), lactate dehydrogenase (LDH), gamma-glutamyl transferase (γ-GT), alkaline phosphatase (AP), T- cells, HLA-DR+ T-cells, Ki67+ T-cells, CD4+ T-cells, HLA-DR+ CD4+ T-cells, Ki67+ CD4+ T-cells, CD4+ central memory T cell (Tcm), CD4+ effector memory T-cell (Tem), CD4+ Temra, CD4+ naive T-cell (Tnaive), CD8+ T-cells, HLA-DR+ CD8+ T-cells, CD8+ Tcm, CD8+ Tem, CD8+ Tnaive, B-cells, Ki67+ B-cells, natural killer cells (NK-cells), Ki67+ NK-cells, regulatory T-cells (Treg) and Ki67+ Treg. Significantly longer PFS was correlated with lower baseline levels of TGF-β1 (PFS >180 days: 2023.08 ±1173.88 pg/mL, PFS ≤180 days: 5030.15 ±2458.92 pg/mL, logrank p = 0.019) and higher levels of HLA-DR+ CD8+ T-cells (PFS >180 days: 3.83 ±2.51%, PFS ≤180 days: 3.25 ±1.86%, logrank p = 0.027) and CD8+ Temra (PFS >180 days: 49.68 ±19.25%, PFS ≤180 days: 47.18 ±24.23%, logrank p = 0.031). A lower baseline level of TGF-β1 (PFS >180 days: 2023.08 ±1173.88 pg/mL, PFS ≤180 days: 5030.15 ±2458.92 pg/mL, p = 0.023) and a higher number of Ki67+ NK-cells (PFS >180 days: 9.12 ±4.77%, PFS ≤180 days: 5.40 ±2.36%, p = 0.048) correlated with a better treatment response. Baseline levels of the following parameters did not correlate with PFS: IL-1β, IL-6, IL-8, IL-17A, albumin, creatinine, bilirubin, INR, AFP, LDH, GOT, GPT, γ-GT, AP, leukocytes, CRP, MELD, lymphocytes, T-cells, HLA-DR+ T-cells, Ki67+ T-cells, CD4+ T-cells, HLA-DR+ CD4+ T-cells, Ki67+ CD4+ T-cells, CD4+ Tcm, CD4+ Tem, CD4+ Temra, CD4+ Tnaive, CD8+ T-cells, Ki67+ CD8+ T-cells, HLA-DR+ CD8+ T-cells, CD8+ Tcm, CD8+ Tem, CD8+ Tnaive, CD8+ Temra, B cells, Ki67+ B-cells, NK-cells, Ki67+ NK-cells, Treg and Ki67+ Treg. In addition, significantly higher baseline TGF-β1 levels existed in less favorable grading (G1: 3561.41 ±1920.77 pg/mL, G2/G3: 6243.24 ±1333.18 pg/mL, p = 0.050), more advanced BCLC-score (BCLC A: 2761.91 ±1409.21 pg/mL, BCLC B/C/D 5143.09 ±2631.47 pg/mL, p = 0.040) and incomplete devascularization (complete: 3170.62 ±1989.05 pg/mL, incomplete: 5966.44 ±2385.08 pg/mL, p = 0.011). There was no significant correlation of baseline TGF-β1 levels with the other clinical classifications Child-Pugh, etiology, tumor size, and MELD. Furthermore, no correlation of the baseline level of the other determined parameters with these clinical parameters was observed. Conclusions An important previously identified mechanism for inadequate treatment response in HCC is depletion of CD8+ T-cells, decreased NK-cell function and increased activity of Treg. In this study, the correlation of an increased baseline level before TACE of HLA-DR+ CD8+ T-cells and of Ki67+ NK-cells with a longer PFS confirmed this finding. TGF-β1 is thought to play a key role in the development of immune cell exhaustion and activation of Treg. The higher baseline level of TGF-β1 correlating with shorter PFS, less favorable grading, advanced BCLC score and incomplete devascularization confirmed this assumption. These observations could be used in clinical practice as easily available prognostic markers or as target for immunomodulatory therapy.
Doctoral thesis
Open Access
Modeling and Simulation of Bulk Metallic Glass Crystallization During Laser Powder Bed Fusion
(2024) Yang, Zerong; Körner, Carolin
Bulk metallic glasses (BMGs) exhibit many exceptional mechanical properties, making them promising structural materials in diverse industries. These properties derive from their characteristic amorphous structure. However, conventional processing technologies, due to limited cooling rates, face challenges in preventing crystallization to maintain amorphous structures. This limitation restricts the fabrication of complex BMG structures with substantial section thickness. Consequently, it remains challenging to realize the full spectrum of attractive properties offered by BMGs through conventional processing technologies. In contrast, laser powder bed fusion (PBF-LB/M), with its rapid cooling rates at the solidification front, shows great promise for manufacturing BMG parts. Yet, to achieve success, it is crucial to develop optimal process strategies that can effectively minimize both crystallization and process defects. Unfortunately, the current lack of understanding regarding BMG crystallization during PBF-LB/M forces us to heavily depend on trial-and-error experiments for process development. This approach, however, is often not only time-consuming but also costly. This thesis aims to attain a profound and comprehensive understanding of the crystallization behavior of BMGs during PBF-LB/M and develop a numerical tool that can significantly assist in the PBF-LB/M process development. The numerical work builds upon our in-house simulation software, SAMPLE2D, originally developed for simulating the electron beam powder bed fusion process. Consequently, this thesis focuses on two key modeling aspects: laser beam absorption and crystallization. Laser beam absorption was modeled using a ray-tracing approach, considering reflection, refraction, and absorption. The emphasis was on calculating the laser beam absorption coefficient, accounting for its dependence on the angle of incidence and temperature, based on fundamental physical laws. Model validation involved comparing simulation-predicted melt pool geometries with experimentally observed ones, revealing a satisfactory agreement between the two. Modeling of the crystallization behavior serves as the core focus of this thesis. Experimentation provided insights into primary physical effects to consider in the modeling efforts. The modeling approach revolved around the utilization of a phenomenological phase transformation model, specifically the Nakamura model. For numerical implementation, a two-step Euler method was devised. Model parameters were determined through isothermal crystallization experiments, providing essential inputs for the simulation. Experimental validation was achieved by comparing single-track-multi-pass and multi-layer PBF-LB/M results with simulation outcomes. Throughout the study, the glass-forming alloy Zr59.3Cu28.8Al10.4Nb1.5 (at.%) was used as a representative example. The robust agreement between experimental and simulation results allowed for the extensive application of the extended SAMPLE2D, enabling a comprehensive investigation of PBF-LB/M process conditions, encompassing beam- and powder bed-related parameters, as well as scanning strategies. Numerical insights unveiled that BMG crystallization during PBF-LB/M is a localized phenomenon primarily induced by short-range in-situ heat treatment. Furthermore, a general process development guideline was proposed: First, utilize an appropriate energy input to minimize porosity; Second, within the parameter space capable of fabricating dense parts, minimize the energy input and select parameters that yield a more even energy distribution; Third, optimize the scanning strategy to further reduce crystallization. Notably, the line jumping scanning strategy was highlighted as an effective means of minimizing crystallization while maintaining relative density. Therefore, this thesis not only deepens our comprehension of BMG crystallization during PBF-LB/M but also furnishes the industry with a readily applicable numerical tool to aid in practical PBF-LB/M process development.
Doctoral thesis
Open Access
Epitaktische Kobaltoxidfilme auf Au(111): Wachstum, Eigenschaften und Adsorption organischer Moleküle
(2024) Ammon, Maximilian Michael; Schneider, M. Alexander
Molekulare Selbstanordnungen auf Oberflächen sind für die Grundlagenforschung faszinierend und potenziell technologisch nutzbar. Vielversprechend ist dabei die Funktionalisierung von Oberflächen mit Porphyrinen, die für metallische Substrate weitgehend verstanden ist. Für Anwendungen wie z.B. die Photovoltaik ist es jedoch notwendig, die Moleküle elektrisch vom Substrat zu entkoppeln. Dies kann unter anderem durch die Verwendung isolierender oder halbleitender Substrate wie Metalloxide erreicht werden. Ein mögliches und relevantes Metalloxid ist Kobaltoxid, das sich in chemischer Hinsicht durch die Variabilität der Oxidationszahlen der Metallionen auszeichnet. Die bekannten Volumenkristalle CoO (Kochsalzstruktur) und Co3O4 (Spinellstruktur) sind Halbleiter. Diese Arbeit beschreibt das epitaktische Wachstum von Kobaltoxid auf der Au(111) Oberfläche, das hauptsächlich mit Rastertunnelmikroskopie und Beugung langsamer Elektronen untersucht wurde. Durch sorgfältige Kontrolle der Präparationsparameter konnte erstmals das epitaktische Wachstum von CoO mit Wurtzitstruktur, CoO mit Kochsalzstruktur und Co3O4 mit Spinellstruktur konsistent beschrieben und gegeneinander abgegrenzt werden. Anschließend wird auf die Adsorption von 5,10,15,20-Tetrakis(p-Cyanophenyl)Porphyrin auf der (100)-Oberfläche von Kochsalz-CoO eingegangen. Kochsalz-CoO(100) kann auf Au(111) als etwa 1 nm dicker, halbleitender Film gewachsen werden, der die Oberfläche nur partiell bedeckt. Dies ist ein ideales Modellsystem, um die Adsorption mit Rastertunnelmikroskopie zu untersuchen. Nach Adsorption bei Raumtemperatur bilden die Moleküle eine mit dem Substrat kommensurable Selbstanordnung, die durch ein Gleichgewicht zwischen Molekül-Molekül- und Molekül-Substrat-Wechselwirkung erklärt werden kann. Mittels Tunnelspektroskopie konnten zwei Molekülvarianten in der Selbstanordnung nachgewiesen werden, von denen die eine durch Metallierung mit Kobaltatomen aus dem Film entsteht.
Doctoral thesis
Open Access
On the edge of active systems: The theory of bead-spring microswimmers
(2024) Ziegler, Sebastian; Smith, Ana-Sunčana
Microswimmers are autonomous active agents, able to self-propel in viscous fluids. As an example of active systems, microswimmers continuously dissipate energy and thus are not subject to the restrictions of thermal equilibrium. While the emerging complex single-swimmer and collective behaviors have been studied for various systems in experiments, numerically or theoretically, the physical principles governing locomotion in the viscous regime are not yet fully understood. In particular, the impact of a swimmer's elasticity and the resulting altering of the swimming stroke on the swimmer dynamics is not explored in detail and requires further investigation. This thesis, which builds on six peer-reviewed publications [P1 - P6], focuses on bead-spring swimmers consisting of spherical beads connected by harmonic springs and driven by bead forces. In contrast to previous microswimmer models with a prescribed swimming stroke, bead-spring swimmers are able to adapt their stroke in dependence on their environment, which is an important ability of also biological microswimmers. Since the associated equations of motion for bead-spring swimmers are non-linear in the positions of the beads, we have developed a versatile perturbative approach to calculate the swimmer behavior [P1]. It compares very well to numerical results and presents a refinement of previous analytical results in the literature. Inspired by experiments with driving forces prescribed in the laboratory reference frame, we study the behavior of the triangular bead-spring swimmer under external driving forces. This swimmer undergoes translations as well as rotations and relaxes towards one of six stable orientations, which are equidistantly interspersed with unstable steady states [P1]. The stable and unstable states are associated with opposite swimming directions and a puller- and pusher-like dipolar time-averaged flow field, respectively. Employing the perturbative approach, it is generalized that the magnitude of the time-averaged flow field is proportional to the swimming velocity. Even more generally, we show that this relation can be proven for any swimmer with only two internal degrees of freedom being excited [P4]. In the publications [P2, P3], the frequency-dependent swimming velocity of the triangular bead-spring swimmer is compared to that of the magnetocapillary triangular swimmer studied in Lattice-Boltzmann simulations. Both the bead-spring swimmer and the magnetocapillary swimmer in Lattice-Boltzmann simulations show an optimum in the propulsion speed with respect to the frequency of the applied forces or the magnetic field, respectively. However, the optima are associated with different time scales in both systems. In the bead-spring system, inertia is generally neglected, and the time scale associated with maximal velocity emerges from the interplay of viscous drag and spring forces. In the case of the magnetocapillary swimmer, however, the time scales associated with the optimal swimming velocity of the magnetocapillary swimmer are shown to depend strongly on the beads' inertia [P2, P3]. The study of bead-spring swimmers is extended to the collective behavior of two linear three-sphere bead-spring swimmers in publication [P4]. As the main result, two such swimmers benefit mutually in a large fraction of the parameter space due to effects resulting from the time-averaged flow field each swimmer produces. This so-called passive interaction is complemented by active interaction effects, which result from the interplay of the swimmer's own activity with the time-dependent flow fields produced by nearby swimmers. We show that active interactions are particularly relevant for swimmers' rotations due to interactions with other swimmers. The rotations are responsible for the complex long-term trajectories of two interacting swimmers, which can be understood from the instantaneous rotations experienced by the swimmers [P4]. The results for two interacting swimmers in our framework, which uses an analytical expression for the hydrodynamics, are compared to simulations of the bead-spring swimmer using a coupled Lattice-Boltzmann and immersed boundary method [P5]. While a good agreement between both methods is observed for a single swimmer, there is a qualitative mismatch on which of two interacting collinear swimmers moves faster. Preliminary results suggest that this mismatch can be resolved by assuming a time delay associated with flow propagation, known to exist in Lattice-Boltzmann simulations. Studies on single swimmers are eventually extended to the behavior of bead-spring swimmers in viscosity gradients as a common environment of motile bacteria and other microorganisms. To this end, a framework to calculate the mobility matrix for spherical particles in asymptotically small viscosity gradients, based on the Lorentz reciprocal theorem and a reflection method, is presented in [P6]. It is demonstrated that a linear viscosity gradient gives rise to interactions between the particles that do not decay with distance as long as the separation between the particles is smaller than the size of the viscosity gradient by orders of magnitude. While previous studies typically considered infinitely large viscosity gradients, we show that the relative positioning of the particles within the finite-size linear gradients substantially affects their interaction behavior. With these results at hand, the behavior of the triangular bead-spring swimmer is studied within asymptotically small linear viscosity gradients. For an equilateral swimmer with equal beads, the swimmer generally aligns either parallel or anti-parallel to the viscosity gradient. Whether the swimmer eventually swims towards higher viscosity, i.e., behaves viscophilic, or lower viscosity, i.e., behaves viscophobic, depends strongly on the swimmer's parameters. However, the swimmer behaves generally viscophilic and viscophobic in the limits of very stiff and very soft springs, respectively. By exploiting the self-developed versatile perturbative framework for bead-spring swimmers, this thesis significantly advances the understanding of microswimming by linking flow field and swimming velocity as well as providing the relation between stroke-based and elastic force-based swimmers. Furthermore, the emergence of the collective behavior from single-swimmer properties is further elucidated, as well as the impact of swimmer rigidity on its behavior in a viscosity gradient. Beyond the fundamental results obtained, the methods presented in this thesis lay the foundations for future studies. Especially by allowing for a swimmer's response to external stimuli, our approach is well-suited to study, e.g., swimming algae cells or artificial magnetocapillary swimmers on a liquid interface, which will be addressed in the future. Furthermore, the swimmer's responsiveness is a vital ingredient in the design of smart microscopic devices for, e.g., medical applications, and our results might be helpful in their development.
Doctoral thesis
Open Access
Selbstverstärkendes Lernen als Beitrag zur Automatisierung der Anpassungskonstruktion
FAU Studien aus dem Maschinenbau : 441, (FAU University Press, 2024) Dworschak, Fabian; Wartzack, Sandro; Franke, Jörg; Hanenkamp, Nico; Hausotte, Tino; Merklein, Marion; Müller, Sebastian; Schmidt, Michael; Wartzack, Sandro
Die Automatisierung beschreibt die Übergabe von Aufgaben vom Menschen an eine Maschine. Die Aufgaben der Anpassungskonstruktion umfassen das Festlegen der Merkmalsausprägungen zum Erfüllen neuer Anforderungen bei gleichbleibender Wirkstruktur. Sie sind herausfordernd, wenn Merkmale gleichzeitig konkurrierende Eigenschaften beeinflussen. Je nach Kontext können sie durch Optimierungsalgorithmen oder datengetriebene Ansätze automatisiert werden. Datengetriebene Ansätze setzen eine Datenbasis voraus, die alle relevanten Informationen für das Erkennen der Zusammenhänge zwischen Merkmalen und resultierenden Eigenschaften enthält. Optimierungsalgorithmen müssen bei einer Änderung der Randbedingungen erneut zur Konvergenz gebracht werden. Für die Schnittmenge der Anpassungsaufgaben, für die sich die Randbedingungen durch Anforderungsänderungen regelmäßig ändern und gleichzeitig keine ausreichende Datenbasis vorliegt, stellt diese Arbeit das Selbstverstärkende Lernen vor. Es beschreibt das Lernen eines Agenten auf Basis von Erfahrungen mit einer Umgebung. Hier wird diese durch Konstruktionsumgebung der Produktentwickelnden im Rahmen der Anpassungskonstruktion repräsentiert. Der Agent passt die Merkmale an und erhält die Produkteigenschaften als Feedback. Ziel der Arbeit ist nicht das Entwickeln eines neuen Algorithmus für Selbstverstärkendes Lernen, sondern das Erforschen der Voraussetzungen für Selbstverstärkendes Lernen in der Produktentwicklung, das Definieren der Rolle der Produktentwickelnden bei dessen Einsatz sowie das Erforschen der Übertragbarkeit der gesammelten Erfahrungen von einer Anpassungsaufgabe auf nachfolgende.