Die vorliegende Arbeit befasst sich mit der Kristallzüchtung von Verbindungen, die extreme Bedingungen erfordern und deren Optimierung. Neben qualitativ hochwertigen Ausgangssubstanzen kommen dabei die Verweindung von Gasgemischen und von Hochdruck in der Kristallzüchtungskammer sowie geringe Ziehgeschwindigkeiten im Bereich von <0,4 mm/h zum Einsatz. Der Effekt von Hochdruck, insbesondere Sauerstoffhochdruck bis zu 150 bar, wurde während der Züchtung von Oxiden mit dem tiegelfreien Zonenschmelzverfahren, auf die Zusammensetzung und Kristallqualität untersucht. Im Gegensatz zu anderen Verfahren kommt es bei dieser Methode zu keinen Verunreinigungen durch Tiegelmaterial. Des Weiteren steht die Schmelze in direktem Kontakt zur Atmosphäre und kann mit dieser wechselwirken. Trotz der großen und interessanten Veränderungen die sich dadurch ergeben, gibt es hierzu bisher sehr wenige Forschungsaktivitäten. Dies kann damit erklärt werden, dass bisher die Möglichkeiten fehlten, solche Experimente durchzuführen. Um die genannten Züchtungsbedingungen zu realisieren, müssen entsprechende Anlagen bereitstehen. Ein Bestandteil dieser Arbeit war die Mitarbeit am Design sowie an den Messungen und der Erprobung von Funktionselementen für eine neu konzipierte Kristallzüchtungsanlage für Hochdruckversuche bis 150 bar.

In der Arbeit werden einfache Möglichkeiten zur experimentellen Simulation der Schnellerstarrung von Stählen durch Tauchen eines Kupferstabes in ein Stahlbad aufgezeigt. Zum Verständnis des Erstarrungsvorganges beim Tauchen wurde ein mathematisches Modell erstellt. Im Vergleich der Modellergebnisse mit den Messergebnissen wurde eine gute Übereinstimung gefunden. Mit Hilfe des Modells wurden die Erstarrungsparameter der gesamten lokalen Erstarrungszeit, der lokalen Erstarrungsgeschwindigkeit, der lokalen Abkühlgeschwindigkeit und der gesamten lokalen Erstarrungswärmestromdichte bestimmt. Durch mikroskopische Untersuchungen der Stahlerstarrungsschichten wurde die Abhängigkeit des Sekundärdendriten-Armabstandes und der Größe der nichtmetallischen Einschlüsse von den oben genannten Erstarrungsparametern gefunden. Das mathematische Modell wurde von einem rein „thermischen“ in ein „thermisch-strukturelles“ erweitert. Dazu wurden die über die metallographischen Untersuchungen erhaltenen Abhängigkeiten benutzt. Mit Hilfe der ermittelten Großen können technische Erstarrungsvorgänge der kontinuierliche, konventionellen und endmaßnahen Gießverfahren für Stähle beschrieben werden.

In the course of this PhD thesis metastable Cu50Zr50-xTix (0≤ x ≤ 10) and (Cu0.5Zr0.5)100-xAlx (5 ≤ x ≤ 8) alloys were prepared and characterised in terms of phase formation, thermal behaviour, crystallisation kinetics and most importantly in terms of mechanical properties. The addition of Al clearly enhances the glass-forming ability although it does not affect the phase formation. This means that the Cu-Zr-Al system follows the characteristics of the binary Cu-Zr phase diagram, at least for Al additions up to 8 at.%. Conversely, the presence of at least 6 at.% Ti changes the crystallisation sequence of Cu50Zr50-xTix metallic glasses and a metastable C15 CuZrTi Laves phase (Fd-3m) precipitates prior to the equilibrium phases, Cu10Zr7 and CuZr2. A structurally related phase, i.e. the “big cube” phase (Cu4(Zr,Ti)2O, Fd-3m), crystallises in a first step when a significant amount of oxygen, on the order of several thousands of mass-ppm (parts per million), is added. Both phases, the C15 Laves as well as the big cube phase, contain pronounced icosahedral coordination and their formation might be related to an icosahedral-like short-range order of the as-cast glass. However, when the metallic glasses obey the phase formation as established in the binary Cu-Zr phase diagram, the short-range order seems to more closely resemble the coordination of the high-temperature equilibrium phase, B2 CuZr. During the tensile deformation of (Cu0.5Zr0.5)100-xAlx bulk metallic glasses where B2 CuZr nanocrystals precipitate polymorphically in the bulk and some of them undergo twinning, which is due to the shape memory effect inherent in B2 CuZr. Qualitatively, this unique deformation process can be understood in the framework of the potential energy landscape (PEL) model. The shear stress, applied by mechanically loading the material, softens the shear modulus, thus biasing structural rearrangements towards the more stable, crystalline state. One major prerequisite in this process is believed to be a B2-like short-range order of the glass in the as-cast state, which could account for the polymorphic precipitation of the B2 nanocrystals at a comparatively small amount of shear. Diffraction experiments using high-energy X-rays suggest that there might be a correlation between the B2 phase and the glass structure on a length-scale less than 4 Å. Additional corroboration for this finding comes from the fact that the interatomic distances of a Cu50Zr47.5Ti2.5 metallic glass are reduced by cold-rolling. Instead of experiencing shear-induced dilation, the atoms become more closely packed, indicating that the metallic glass is driven towards the more densely packed state associated with the more stable, crystalline state. It is noteworthy, that two Cu-Zr intermetallic compounds were identified to be plastically deformable. Cubic B2 CuZr undergoes a deformation-induced martensitic phase transformation to monoclinic B19’and B33 structures, resulting in transformation-induced plasticity (TRIP effect). On the other hand, tetragonal CuZr2 can also be deformed in compression up to a strain of 15%, yet, exhibiting a dislocation-borne deformation mechanism. The shear-induced nanocrystallisation and twinning seem to be competitive phenomena regarding shear band generation and propagation, which is why very few shear offsets, due to shear banding, can be observed at the surface of the bulk metallic glasses tested in quasistatic tension. The average distance between the crystalline precipitates is on the order of the typical shear band thickness (10 - 50 nm) meaning that an efficient interaction between nanocrystals and shear bands becomes feasible. Macroscopically, these microscopic processes reflect as an appreciable plastic strain combined with work hardening. When the same CuZr-based BMGs are tested in tension at room temperature and at high strain rate (10-2 s-1) there seems to be a “strain rate sensitivity”, which could be related to a crossover of the experimental time-scale and the time-scale of the intrinsic deformation processes (nanocrystallisation, twinning, shear band generation and propagation). However, further work is required to investigate the reasons for the varying slope in the elastic regime. As B2 CuZr is the phase, that competes with vitrification, it precipitates in a glassy matrix if the cooling rate is not sufficient to freeze the structure of the liquid completely. The pronounced work hardening and the plasticity of the B2 phase, which are a result of the deformation-induced martensitic transformation, leave their footprints in the stress-strain curves of these bulk metallic glass matrix composites. The behaviour of the yield strength as a function of the crystalline volume fraction can be captured by the rule of mixtures at low crystalline volume fractions and by the load bearing model at high crystalline volume fractions. In between both of these regions there is a transition caused by percolation (impingement) of the B2 crystals. Furthermore, the fracture strain can be modelled as a function of the crystalline volume fraction by a three-microstructural-element body and the results imply that the interface between B2 crystals and glassy matrix determines the plastic strain of the composites. The combination of shape memory crystals and a glassy matrix leads to a material with a markedly high yield strength and an enhanced plastic strain. In the CuZr-based metastable alloys investigated, there is an intimate relationship between the microstructure and the mechanical properties. The insights gained here should prove useful regarding the optimisation of the mechanical properties of bulk metallic glasses and bulk metallic glass composites.