14 Externe wissenschaftliche Einrichtungen
Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/15
Browse
12 results
Search Results
Item Open Access Vortex-Kern-Korrelation in gekoppelten Systemen(2014) Jüllig, Patrick; Schütz, Gisela (Prof. Dr.)In der vorliegenden Arbeit wurden strukturierte ferromagnetische Dreischichtsysteme zum einen auf ihre statische in-plane- sowie out-of-plane-Magnetisierungsverteilung als auch auf deren dynamisches Verhalten hin untersucht. Die sowohl quadratischen als auch kreisförmigen Strukturen bestanden aus zwei ferromagnetischen Lagen mit einer Dicke von jeweils 50nm, welche durch eine nicht magnetische Cu-Zwischenschicht getrennt waren. Die Dicke dieser Zwischenschicht variierte schrittweise von t(Cu)=3nm bis 15nm. Als Magnetmaterialien kamen für die untere Schicht Kobalt (Co) und für die obere Schicht das magnetisch isotrope Permalloy (Ni80Fe20) zum Einsatz. Die lateralen Abmessungen sowie das Aspektverhältnis der beiden Einzelschichten wurden so gewählt, dass der Vortexzustand die stabile Domänenkonfiguration ist. Somit resultierten zwei vertikal übereinander angeordnete Vortexkonfigurationen, sodass deren Wechselwirkung sowohl im statischen als auch im dynamischen Fall untersucht werden konnte. Aufgrund der gewählten Cu-Schichtdicke von mindestens 3nm war gewährleistet, dass die Kopplung der in-plane-Schichtmagnetisierung hauptsächlich durch die elektrostatische Streufeldenergie beeinflusst wurde und somit der Beitrag der Oszillatorischen Zwischenschichtaustauschwechselwirkung vernachlässigt werden konnte. Im Falle zweier vertikal übereinander angeordneter Vortexstrukturen kann man bezüglich der Zirkulation C (beschreibt die Orientierung der in-plane-Magnetisierung) und der Polarisation P (beschreibt die Orientierung der out-of-plane-Komponente des Vortexkerns) unter Berücksichtigung der Symmetrie vier verschiedene Konfigurationen voneinander unterscheiden: Die beiden Fälle, bei denen C und P jeweils bzw. orientiert sind, sowie die beiden Fälle, bei denen lediglich C oder P parallel ausgerichtet ist. Der erste Schritt dieser Arbeit bestand in der Probenpräparation. Als Strukturierungsverfahren kamen zum einen das Ionenstrahlätzen und zum anderen die Elektronenstrahllithographie zum Einsatz. Anhand von Röntgenbeugungsexperimenten konnte herausgefunden werden, dass beide Schichtmaterialien, sowohl das Permalloy als auch das Kobalt, eine polykristalline, fasertexturierte Schichtstruktur mit einer fcc-Gitterstruktur aufwiesen. Diese Erkenntnisse waren vor allem für die korrekte Parameterwahl für die nachfolgend durchgeführten mikromagnetischen Simulationen von großer Bedeutung. Messungen der Oberflächenrauigkeiten mittels des AFM ließen darauf schließen, dass neben dem Beitrag der Streufeldenergie ebenso korrelierte bzw. unkorrelierte Zwischenschichtrauigkeiten zur gegenseitigen Ausrichtung der in-plane-Schichtmagnetisierungen beitrugen. Mit Hilfe von SQUID-Messungen bei T=40K an unstrukturierten Co/Cu/Py-Dreischichtsystemen konnte nachgewiesen werden, dass erst für Proben mit Cu-Schichtdicken ab t(Cu)=2,0nm beide ferromagnetische Materialien chemisch voneinander getrennt vorlagen und keine direkte ferromagnetische Kopplung aufgrund von sogenannten Pinholes auftrat. Somit konnte geschlussfolgert werden, dass erst ab einer Dicke von t(Cu) größer gleich 2,0nm eine vollständig geschlossene Cu-Schicht vorlag. Die ersten statischen in-plane-Messungen am STXM zeigten, dass Proben, welche im as-sputtered Zustand eine undefinierte metastabile Mehrdomänenkonfigurationen aufwiesen, durch einen Entmagnetisierungsprozess in den stabilen Vortexzustand überführt werden konnten. Neben antiparallel gekoppelten Systemen bezüglich der Zirkulation C wurden mit einer ähnlich hohen Wahrscheinlichkeit Proben mit einer parallelen Ausrichtung der in-plane-Magnetisierung gefunden. Dies zeigte, dass die Kopplung der Schichtmagnetisierungen nicht allein durch die Streufelder realisiert wurde, sondern ein weiterer Beitrag hinzukam, dessen Ursache mit hoher Wahrscheinlichkeit in den Zwischenschichtrauigkeiten zu finden war. Statische mikromagnetische Simulationen an quadratischen Co/Spalt/Py-Strukturelementen haben gezeigt, dass die in-plane-Magnetisierungsverteilung der Systeme mit C=parallel eine merklich verzerrte Landaustruktur aufwies. Zudem lag bei Konfigurationen mit P=antiparallel ein lateraler Shift bezüglich der Gleichgewichtspositionen der Kerne vor, was aufgrund der Interaktion der out-of-plane-Streufelder zu erwarten war. Dies spiegelte sich auch in der Energiebetrachtung wieder, wobei die beiden Systeme mit der Konfiguration C=parallel deutlich höhere Gesamtenergien aufwiesen als diejenigen mit C=antiparallel. Allgemein lagen im Falle von parallelen Kernpolarisationen die Energiewerte etwas niedriger als bei antiparallel ausgerichteten Kernen. Die dynamische Anregung der ferromagnetischen Schichtsysteme wurde experimentell mittels eines in-plane-Magnetfeldpulses realisiert, welcher durch die lineare Stripline generiert wurde. Die Pulsdauer betrug je nach Element 0,5 bis 1,6ns, und bezüglich der Pulsamplitude mussten Feldstärken von B(Puls)=3,1mT bis zu 6,0mT angelegt werden, um eine Gyrationsbewegung beobachten zu können.Item Open Access Grain growth and texture evolution in copper thin films(2010) Sonnweber-Ribic, Petra; Arzt, Eduard (Prof. Dr. phil)An improved basic understanding of mechanisms causing grain growth and texture evolution in Cu thin films contains the potential to improve performance and reliability of components and devices. In this work, the influence of film thickness, strain and temperature on grain growth and texture evolution in Cu thin films was investigated. By varying the parameters, information about the underlying mechanisms were revealed. The 0.5 to 10 micrometer thick Cu films were deposited on 125 micrometer thick polyimide substrates (Kapton®, DuPont) using a UHV magnetron sputtering system. For detailed observation of grain growth and texture evolution an EBSD-based in situ testing appliance was constructed. This system allowed the simultaneous observation of grain growth and texture evolution, giving new insight into growth kinetics and details of grain growth. In a first step, Cu thin films of thicknesses in between 0.5 and 10 micrometer were deposited on polymer substrates and annealed at 330°C for 30 min. Their resulting texture and microstructure were investigated by EBSD. A texture transition from (111) to (100) was observed at film thicknesses between 3 and 5 micrometer. The experimental findings were explained by the texture evolution model of Thompson and Carel. A significant observation which cannot be explained by a purely energetic argument is the broad texture transition. In order to get more information about the critical role of strain energy, uniaxial tensile tests were carried out on 3 micrometer thick films. In contrast to theoretical predictions, various tensile tests revealed no influence of strain on grain growth behaviour. Neither at room temperature nor at elevated temperatures, further (100) grain growth was observed. In a next step, the abnormal growth of individual (100) oriented grains was recorded for more than 24 hours at temperatures between 90 and 118°C. Annealing was carried out inside a Leo 1530-VP SEM equipped with a heating facility. Detailed analysis of grain growth and estimates of the possibly acting driving forces indicated that the reduction of dislocation density played an important role for abnormal grain growth. A further hint for the critical importance of defect density was given by the HWHM of the (100) texture fraction. Nevertheless, it was not clear why this driving force favours the growth of (100) oriented grains. A possible answer could be given by the strain energy release maximization (SERM) model developed by Lee. In addition, when analysing the activation energy for grain growth, they were found to possess a higher grain boundary mobility, supporting the preferred growth of (100) oriented grains. A new texture map, considering dislocation density as driving force, was constructed. Assuming dislocation density to play a significant role for grain growth and texture evolution in Cu thin films, the influence of deposition parameters is pointed out.Item Open Access Synthesis and characterization of carbon nanotube reinforced copper thin films(2006) Otto, Cornelia; Arzt, Eduard (Prof. Dr.)Two model composites of copper and carbon nanotubes were fabricated by very different deposition methods. Copper electrodeposition in a plating bath containing nanotubes created a 3D matrix of randomly oriented CNTs within a thick, 20 micron Cu film. In contrast, sandwiching a layer of well-separated nanotubes between two sub-micron sputtered Cu layers produced a 2D-composite with nanotubes lying parallel to the substrate surface. These composites, which were mechanically tested using various techniques, proved to be well suited to explore the nature of the CNT/Cu matrix interface. Columns approximately 600 nm in diameter and 1.4 microns high were cut from the sputter-deposited composite and microcompression tested in a nanoindenter. No influence of the presence of nanotubes on the stress-strain-curves was observed, which was attributed to the low nanotube content. On the other hand, microscopic analysis showed an influence of the nanotube on the copper immediately surrounding it, resulting in funnel-like depressions on the column surface. In addition to deformation by slip, twinning was observed in some columns, which has never before been reported in the literature for micron-sized columns. Macroscopic tensile tests were performed on the electrodeposited films and the samples with the highest carbon content showed an increase in toughness of over 100% with respect to the CNT-free control samples produced by the same method. Finally, short copper electrodepositions into carbon nanotube carpets revealed large regions with conformally coated nanotubes. Until now, it was assumed that copper would not wet the nanotubes and that the interfacial strength between copper and CNTs would be low, since copper does not form a carbide. However, these experiments all revealed clear evidence of adhesion exceeding the copper shear strength. To our knowledge, this is the first time such strong adhesion was demonstrated between a nanotube and a metal matrix. We attribute this unexpected, but highly desirable adhesion and wetting behaviour to the defect structure in the nanotubes used. Most of the experiments were done with nitrogen-doped carbon nanotubes, which are known to be rich in defects. The nanotube carpets used in the last experiment were not doped but had a high defect density due to the synthesis method used. As a good adhesion between fiber and matrix is a prerequisite for the successful use of carbon nanotubes in metal matrix composites, these results are very encouraging. The composites and methods presented here provide a foundation for further studies needed to understand the nanotube-metal interaction in more detail and thus ultimately for successful metal-carbon nanotube composites.Item Open Access Microstructural changes and intermetallic compound formation in metallic bilayers(Stuttgart : Max-Planck-Institut für Intelligente Systeme (ehemals Max-Planck-Institut für Metallforschung), 2016) Rossi, Paul J.; Mittemeijer, Eric Jan (Prof. Dr. Ir.)This thesis investigates interdiffusion and intermetallic compound (IMC) formation, as well as their effects on the microstructure, in metallic thin-film bilayers. The investigation focuses on bilayers based on the Ag-Sn and Ag-In binary systems, which are technologically important as basis for lead free solders. Due to the enhanced diffusional mechanisms in these systems, diffusion occurs readily even at room and low temperatures. The proceeding interdiffusion eventually leads to IMC formation in the bilayers, allowing for the investigation of the kinetics of IMC formation and the associated microstructural changes at room and low temperatures. The combination of the properties special to thin films with the diffusional mechanisms in the binary Ag-Sn and Ag-In systems leads to interesting effects, such as the dependence of IMC formation on the stacking sequence in the bilayers. The obtained experimental results for both systems could be explained using thermodynamic and kinetic models. Experimental characterization of the bilayers mainly relied on X-ray diffraction (XRD) and electron microscopy. In order to investigate the effect of the deposition process on IMC formation and the microstructure of the bilayers, different physical vapor deposition (PVD) techniques, especially thermal evaporation and magnetron sputtering, were used for the preparation of the bilayers. During investigation of the Ag-Sn system it was found that ambiguity exists among the published crystal structures of the Ag3Sn IMC. Therefore, the crystal structure of Ag3Sn has been reinvestigated using high-resolution XRD in connection with Rietveld refinements.Item Metadata only Internal precipitation of nitrides in iron-based alloys(Stuttgart : Max-Planck-Institut für Intelligente Systeme (ehemals Max-Planck-Institut für Metallforschung), 2016) Steiner, Tobias; Mittemeijer, Eric J. (Prof. Dr. Ir.)X‐ray diffraction has become a standard method of microstructural analysis. However, in systems with a complex microstructure, the interpretation of the measured diffractograms may not be very straightforward. The influence of the precipitation of fine alloying element nitrides and the changes in precipitation morphology that occur upon continued nitriding on the shape and position of XRD peaks have been identified. The thus obtained quantitative model has been applied to a variety of precipitation systems and in general good agreement of predicted values and experimental results is found. The precipitation of finely distributed alloying element nitrides is the main strengthening mechanism in the diffusion zone of nitrided parts. Various alloying elements having an affinity for N show considerably different nitriding behavior. Cr shows a strong N‐affinity and therefore readily precipitates in the presence of N. However, Mo has a weak strength of interaction with N and reacts only slowly. In order to better understand the nitriding behavior of nitrided steels containing both Cr and Mo, the nitriding behavior of ternary Fe‐Cr‐Mo model alloys is investigated. The complex precipitation sequence of ternary mixed Cr-Mo-nitrides has been identified and the role of the Cr/Mo-ratio of the alloy is exposed. X‐ray diffraction has become a standard method of microstructural analysis. However, in systems with a complex microstructure, the interpretation of the measured diffractograms may not be very straightforward. The influence of the precipitation of fine alloying element nitrides and the changes in precipitation morphology that occur upon continued nitriding on the shape and position of XRD peaks have been identified. The thus obtained quantitative model has been applied to a variety of precipitation systems and in general good agreement of predicted values and experimental results is found.Item Open Access Nitriding of Fe-Mo alloys and maraging steel : structure, morphology and kinetics of nitride precipitation(2012) Selg, Holger; Mittemeijer, E.J. (Prof. Dr. Ir.)The nitriding behaviour of Fe-Mo alloys was investigated in terms of sequence of nitride precipitation and its morphology and kinetics. Upon nitriding Fe-1at.% Mo alloy cubic Mo2N-type precipitates develop as platelets of several hundreds of nm length and thickness of a few atomic layers. Upon prolonged treatment a discontinuous precipitation reaction initiating at ferrite-matrix grain boundaries takes place replacing the submicroscopical Mo2N-type platelets by a lamellar ferrite/MoN (hexagonal crystal structure) microstructure. The morphology of the formed compound layer consisting of iron-nitride at the surface is strongly dependant of the defect density of the matrix: a compact layer develops in case of cold rolled specimen having a high defect density, whereas a plate-like morphology develops in case of recrystallized specimens having a low defect density. The development of the nitrogen concentration-depth profiles in maraging steel upon nitriding can be successfully modelled with a combined diffusion and precipitation kinetic model, provided that the role of excess nitrogen is recognized. The model employs the following fit parameters: the composition parameter of the nitride (expressing the presence of immobile, i.e. adsorbed excess nitrogen at the nitride/matrix interface), the solubility of nitrogen in the matrix (recognizing the presence of mobile, i.e. additionally in the strained matrix, dissolved excess nitrogen), the diffusion coefficient of nitrogen in the matrix and the solubility product of the nitride precipitates.Item Open Access Structural and mechanical investigations of biological materials using a Focussed Ion Beam microscope(2005) Orso, Steffen; Arzt, Eduard (Prof. Dr. phil.)Biological materials have been evolved over millions of years of evolution to fulfil the requirements posed by the organism and environment. A closer inspection of these materials reveals that they are composites with a highly hierarchical structure. A detailed understanding of the behaviour and function of these materials is possible only if the structure and the mechanical properties down to the smallest level of the hierarchy are known. This requires specimens of very small scale to be analysed. This thesis describes the development and application of a novel technique for the quantitative investigation of both the three-dimensional structure and the mechanical properties of biological materials. This technique allows the micromechanical testing in bending and tension of samples of a few tens of micrometers in length and a few micrometers or less in diameter. It uses a Focussed Ion Beam system (FIB) as an in situ laboratory for structural investigations, sample preparation and sample fixation. Mechanical tests are carried out in situ in a FIB and a scanning electron microscope (SEM). Advantages of this method are that samples from larger objects can be prepared site-specifically using the FIB, and that testing in tension is possible without end effects due to gripping, since the samples are affixed by metal ‘tapes’ deposited using the FIB. Forces are measured with a piezoresistive Atomic Force Microscope (AFM) tip attached to a micromanipulator for high precision positioning. The displacement is determined from micrographs taken during the test. The mechanical properties of three different polymeric and biological materials and structures were measured in bending in situ inside an SEM: polyimide (Kapton®), horse hair (keratin) and spruce wood cell wall material (cellulose-fibre composite). Four different biological materials were tested in tension in situ in a FIB: a single element (seta) of the hairy attachment system of a beetle Gastrophysa viridula, wind-receptor hairs from the filiform sensor of crickets (Acheta domesticus) (both chitin-fibre composites), natural spider silk from the garden cross spider (Araneus diadematus) and artificial spider silk (protein fibres). Some of the biological samples could be tested for the first time using the newly designed testing method. They showed exceptional high mechanical properties when compared to technical materials.Item Open Access The initial oxidation of magnesium(2004) Kurth, Matthias; Mittemeijer, E.J. (Prof. Dr. Ir.)Pure Magnesium samples have been oxidised in an UHV chamber under controlled conditions. Pressure range was 10-10 Torr to 10-7 Torr, temperature range was 273 K to 435 K. The samples have then been investigated with XPS, Ellipsometrie and HERDA. Additionally, furnace oxidations at 750 Torr and 673 K have been carried out and investigated with XPS. From the XPS measurements data concerning layer thickness, composition, oxidation state and binding state have been gained. The ellipsometrie measurements yielded additional data concerning layer thickness as well as the size of the band gap of the developing oxide. With the HERDA measurements, the oxygen content within the oxide layer has been determined yielding additional information about composition and layer thickness. The layer thickness as a function of time have then been modelled with a kinetic growth model of Fromhold & Cook. For the refinement of the XPS data concerning layer thickness and composition, the pronounced plasmon excitations that occur in magnesium have been determined with two different procedures which have been developed in the methodical part of this work. The layer thickness and composition values have thus been corrected. Results: Two oxidation stages could be identified: a strong increase for the first few Langmuirs (1L = 1s * 10-6 Torr), followed by a „saturation“ region which was about 1.2 nm to 1.5 nm in magnitude. XPS and ellipsometry results have thereby been in very good agreement. The composition of the developing oxide showed a clear deviation from stoichiometric MgO, mainly caused by an oxygen deficiency; this deficiency has also been confirmed with the HERDA measurements. The Mg/O ratio as a function of layer thickness showed a continous decay starting from very high values for the thinnest layers (>~2.5) down to a saturation value of about 1.4, even for larger layer thicknesses gained with the furnace oxidations. The determination of the size of the band gap which has been gained from the ellipsometry measurements showed a clearly reduced value compared with MgO (~2.5 eV vs. 7.8 eV). The modelling of the kinetic data with the Fromhold & Cook model resulted in good agreement of fit and experimental data. Thereby, fit parameters have been the work function for the interface oxide/oxygen as well as the energy barrier for cation movement (here: Mg cations). For both parameters reasonable values have been gained. Furthermore, a dependence from the oxidation stage corresponding to the changing composition could be realised. In the methodical part of this work mentioned above, the plasmon excitations in Mg have been calculated, for the first time, separately for bulk and surface plasmons. The reconstruction of a Mg spectrum with physically reasonable functions yielded 35 % and 11 %, respectively, relative to the excitation of the main line. With a second method, which is based upon experimental data and which has been further developed here (especially the error estimation procedure), both excitations could have been determined to 17 % and 6 %, respectively.Item Open Access Keimbildung und Wachstum metallischer Whisker(2013) Weissmayer, Lisa; Strunk, Horst (Prof. Dr.)Die vorliegende Dissertation leistet einen Beitrag zur Aufklärung von Keimbildung und Wachstum mittels physikalischer Gasphasenabscheidung hergestellter Whisker aus Kupfer, und bei ausgewählten Züchtungsexperimenten zum Vergleich aus Silber. Als Substrate für die Züchtung der Whisker dienten einkristalline Siliziumwafer der Orientierungen (100), (110) und (111), welche mit einer 60 nm dicken Deckschicht aus amorphem Kohlenstoff beschichtet sind. Die Whisker wurden bei erhöhten Temperaturen (903-973 K) mittels MBE gezüchtet. Anordnung, Form (Längen, Durchmesser, Winkel zur Substratoberfläche) und Flächendichte der gewachsenen Whisker wurden hauptsächlich mit Hilfe der Rasterelektronenmikroskopie untersucht. Zweierlei Substrate wurden verwendet: mit abgeschiedener Kohlenstoffschicht unstrukturiert und nachträglich strukturiert durch Löcher oder Goldkolloide. Die Löcher sind künstlich hergestellte Nukleationsstellen, die in den Löchern entstandenen Whisker belegen das von Hofacker [Hofacker, L., Universität Stuttgart, 2009. Stuttgart.] aufgestellte Keimbildungsmodell. Die Goldkolloide sind ebenfalls Nukleationsstellen und bestätigen die Verallgemeinerung des Keimbildungsmodells: Whisker entstehen an punktuellen Bereichen erhöhter Oberflächenenergie umgeben von Bereichen niedriger Oberflächenenergie, welche das laterale Wachstum des Keims verhindern. Auf unstrukturierten Substraten beeinflusst die Morphologie der Kohlenstoffschicht die Flächendichte an Keimstellen. Diese Keimstellen besitzen – abweichend zu den künstlich generierten - unterschiedliche Oberflächenenergien, welche zu verschiedenen Keimbildungsarbeiten und somit Keimbildungsgeschwindigkeiten führen, was sich in einer Zunahme der Flächendichte an Whiskern mit steigender Beschichtungsdauer äußert. Neben der Keimbildungsarbeit beeinflussen die Desorption und Oberflächendiffusion der Adatome sowie die Abscheiderate die Keimbildungsgeschwindigkeit. Die ersten drei Prozesse sind thermisch aktiviert und führen bei einer konstanten Abscheiderate zu einer maximalen Keimbildungsgeschwindigkeit bei einer bestimmten Temperatur, die sich in einer maximalen Flächendichte an Whiskern äußert. Mit zunehmender Beschichtungsdauer wachsen die Whisker in Länge und Durchmesser. Das Dickenwachstum ist vor allem auf direktes Auftreffen der Adatome aus der Gasphase auf die Oberfläche des Whiskers zurückzuführen, das Längenwachstum hingegen erfolgt hauptsächlich durch die Einlagerung von Adatomen, die auf der Kohlenstoffschicht ankommen und über Oberflächendiffusion den wachsenden Whisker erreichen. Mit steigendem Lochdurchmesser führen die Wachstumsprozesse – übereinstimmend mit den experimentellen Ergebnissen auf strukturierten Substraten - zu einem sinkenden Aspektverhältnis des entstehenden Whiskers bei konstanter Beschichtungsdauer. Die zu Vergleichszwecken gezüchteten Silberwhisker untermauern die allgemeine Gültigkeit der Aussagen zu Keimbildung, Dicken- und Längenwachstum. Qualitativ zeigen Kupfer- und Silberwhisker gleiches Verhalten. In der Literatur anerkannte Modelle zum Wachstum PVD-gezüchteter Whisker sind nicht geeignet, das Wachstum der in dieser Arbeit gezüchteten Whisker zu beschreiben. Dittmar und Neumann [Dittmar, W. et al., Z. Elektrochem., 1960, 37, 428-430.] und Blakely und Jackson [Blakely, J.M. et al., J. Chem. Phys., 1964, 41, 3139-3149.] vernachlässigen den wichtigen Einfluss der Diffusion der Adatome auf dem Substrat als Beitrag zum Längenwachstum. Des Weiteren stoßen die genannten Modelle, wie auch das von Ruth und Hirth [Ruth, V. et al., J. Chem. Phys., 1964. 41, 3139-3149], an ihre Grenzen, wenn die Whisker durch Keimbildung auf den Seitenflächen in ihren Durchmessern zunehmen, wie die experimentellen Ergebnisse im untersuchen Fall zeigen. Die Arbeit hat den wichtigen Einfluss des Substrats auf die Entstehung und das Wachstum der Whisker gezeigt. Mit den Erkenntnissen dieser Arbeit ist bei bekannten Oberflächenenergien von Substrat, Deckschicht und Whiskermaterial, sowie den Aktivierungsenergien von Oberflächendiffusion und Desorption der Adatome bei der Züchtungstemperatur eine Voraussage möglich, ob Whisker bei geeigneter Substratstrukturierung entstehen und wachsen können. Dazu muss das Substrat bevorzugte Keimstellen aufweisen und das Zusammenspiel von Oberflächendiffusion und Desorption der Adatome zu einem schnelleren Längen- als Dickenwachstum führen. Mit der systematischen Herstellung bevorzugter Keimstellen durch Strukturierung der Deckschicht weist die Arbeit einen Weg für die ortsgenaue Züchtung von Whiskern auf, welcher eine wichtige Voraussetzung für deren mögliche Nutzung ist. Weitere Forschung ist allerdings erforderlich, um auf atomarem Niveau die Keimbildung und die Einbauwege zusätzlicher Atome beim Wachstum zu klären.Item Open Access Nanoscale adhesion of individual gecko spatulae explored by atomic force microscopy(2006) Huber, Gerrit; Arzt, Eduard (Prof. Dr. phil.)Attachment mechanisms of animals that can cling to walls and even walk on ceilings have drawn a significant amount of scientific and public attention. The gecko is one of the heaviest and best clinging animals and it has developed intricate hierarchical structures consisting of toes (millimeter dimensions), lamellae (400-600 µm size), setae (micron dimensions) and spatulae (~ 200 nm size). At first this work gives the reader a theoretical background of the techniques used and the underlying physical principles. By means of these techniques the adhesion force for individual spatulae on glass at ambient conditions could be measured and was found to be about 10 nN. This became only possible using the milling facility of a focused ion beam microscope for specimen preparation. The pull-off force was additionally measured as a function of various parameters (air humidity, surface chemistry and surface roughness) and it turned out that the gecko adhesion was remarkably influenced by the presence of water. The pull-off forces were proportional to the relative humidity varied inside an air tight container and increased with decreasing water droplet contact angle of the wafer used. The data obtained were modeled theoretically to explain the observed adhesion phenomena. Two physical theories were presented which are based on concepts of macrocapillarity and the effect of water monolayers on the van der Waals interaction. Both theories showed good agreement with the experimental data. The pull-off forces were also sensitive to the substrate topography. In cases where the surface roughness was in the critical range of the spatula size, presumably imprecise contact formation led to a distinct minimum of the measured adhesion values compared to smoother or rougher surfaces. Furthermore the mechanical properties of single setae could be determined for the first time. The hairs were mechanically tested by three methods: (a) in situ tensile tests using a focused ion beam microscope, (b) three-point bending tests using atomic force microscopy (AFM) and (c) nanoindentation. The results presented in this work shed new light on the nanomechanisms of gecko’s attachment and will help in the rational design of artificial bio inspired attachment systems.