Volume 88, Issue 8

Concerning small scattering objects (aggregates) in a very thin layer, their characterization by small-angle scattering of X-rays in the transmission mode is difficult because the recorded intensity is very weak, owing to the very short X-ray path in the layer. Grazing incidence X-ray scattering circumvents this difficulty, precisely because the X-ray path inside the thin layer to be analyzed is considerably increased, while the background remains sufficiently low. Full potentiality of this technique is obtained when using a synchrotron source (flux, collimation and choice of the wavelength in order to avoid fluorescence or to perform anomalous measurements) and when patterns are recorded with two-dimensional detectors, e.g. gas detectors or imaging plates (IPs). Several examples of small inclusions in thin surface layers are shown in order to highlight the advantage of the technique. For a carbon-silver deposited layer with non-isotropic nanoprecipitates, a comparison between transmission and grazing-incidence modes is made, showing the significant advantage of the latter.

The influence of predeformation on the precipitation and on the resulting mechanical properties in an Al–Zn–Mg alloy was studied using a combination of complementary experimental techniques, namely transmission electron microscopy, differential scanning calorimetry, small-angle X-ray scattering and microhardness. The thermomechanical treatment was designed to show a significant effect of deformation on the precipitation process. It was observed that the presence of dislocations during precipitation promoted fast nucleation of the equilibrium η phase in the η 2 orientation instead of nucleation of the η ′ phase with subsequent transformation to η 1 and η 4 outside dislocations. The large growth rate of precipitates lying on dislocations provoked a growing precipitatefree zone around dislocations. This resulted in a lower value of peak hardness and a faster overall coarsening kinetics.

The effect of Cu and Zr on the precipitation and reversion processes in high strength Al–Zn–Mg alloys was investigated by compression and microhardness tests and by electron microscopy. The addition of Cu increases the supersaturation in the quenched alloy strongly changing thereby the upper temperature limit of Guinier-Preston (GP) zone formation, the reversion of GP zones and the transition from GP zone to intermediate phase. The grain size of the Zr containing alloy is by more than an order of magnitude smaller than that of the Al–Zn–Mg and the Al–Zn–Mg–Cu alloys. Because of the large fraction of material near grain boundaries, the concentration changes caused by segregation and grain boundary precipitation have a pronounced effect on the precipitation process in the Zr containing alloy.

In situ synchrotron radiation small-angle and 100 Bragg scattering experiments have been performed to examine the microstructural change of preaged Al–Li alloys during heating at a constant rate. When heating an Al-7.9 at.% Li sample preaged at 423 K, the average radius decreases first and then increases at higher temperatures. The decrease in the radius is hardly observable for Al-11.8 at.% Li aged at 423 K and Al-7.9 at.% Li aged at 333 K. The difference in the change of average precipitate size is explained on the basis of the kinetics by combining isothermal dissolution and coarsening data. The present results suggest that a coarsening during differential scanning calorimetry experiments is rather common, and it should be considered in more detail when analysing such experiments.

An experimental method is described by which the interfacial concentration of a remelting interface can be measured after quenching. The interface position during remelting of an Al-0.5 at.% Fe alloy in a liquid Al-21 at.% Mg alloy is made visible by interdendritic FeAl 3 precipitates. Measured solute concentrations in the solid phase at the remelting interface show that the local equilibrium is not established for interface velocities in the order of 10 μm/s, and deviations of more than 50% of the equilibrium composition are observed.

The effect of adding Cu, Fe, and Ni to the magnetically hard τ phase in Mn–Al was studied by investigating isothermal sections at 1000 and700 °C. The Mn–Cu–Al alloys show no τ but only hexagonal ε at 1000 °C. Likewise, in Mn–Fe–Al and Mn–Ni–Al only hexagonal ε exists at 1000 °C. The Cu alloys exhibit a hitherto unobserved splitting of ε (and therefore τ) into two phases having the same structure but different Cu contents. One of them can be understood as a continuation of the metastable binary ε which can be undercooled even more effectively in the ternary. The other is the stable extension of ε into the ternary. This apparent decomposition of ε is described with the help of an isopleth at 55 Mn, 45 Al. Above ~ 10 % Cu the τ phase is lost and is replaced by κ, resulting in a drop in Curie temperature and a loss of magnetic hardness. None of these three elements is a true stabilizer of the magnetic τ phase; it always forms by transformation from the hexagonal high-temperature phase ε. This precludes processing in which fully dense magnets are obtained by sintering magnetically oriented powders at high temperatures. The Cu alloys offer an advantage for the type of processing applied to Mn–Al–C because they stabilize the τ precursor ε to temperatures below 700 °C.

Guinier-Preston zones in Al-3 at.% Ag and Al-1.75 at.% Cu were investigated by diffuse wide-angle and small-angle scattering of X-rays and neutrons. On the basis of the separated short-range order scattering, model crystals were generated. In modelled Al–Ag no microstructural differences are found between η zones and ε zones. The average Ag concentration of the zones consisting of ~ 375 (η phase) and 255 atoms (ε phase) is (80 ± 10) at.% and slightly higher in the shell (by 10 to 15 %) than in the core. For Al–Cu short-range order scattering is nearly unmodulated along the 〈100〉 streaks. Modelled crystals show dominantly zones with a Cu fraction of (84 ± 2) at.% while the Cu fraction in the matrix is 0.35 at.%. About 90% of the zones are monolayers.

The effects of short-range order (SRO) and stacking-fault energy (SFE) on the mode of slip and its transition, with increasing solute concentration, from homogeneous to heterogeneous and planar slip are examined. In particular, Cu–Al and Cu–Mn are considered (with Cu–Zn and Cu–Ni for comparison), where reliable SRO data are available for similar compositions and heat treatments as used for the deformation studies. The diffuse antiphase boundary energy is estimated from calculations (based on SRO data) and from transmission electron micrographs of dislocation groups. The diffuse antiphase boundary energy provides a destructible contribution to the glide plane resistance for the motion of dislocations and appears to be the primary reason for the mentioned transitions of the slip mode. The SFE strongly affects the fine structure of slip on the microscale through the cross-slip probability.

The deformation behaviour of the pressure die cast Mg alloys AZ91 and AS21 was investigated in uniaxial compression and tension at constant stress σ as well as in compression at constant strain rate ε˙$\dot \varepsilon $ in the temperature range 70°C ≤ T ≤ 135 °C, which is of interest for the use of Mg alloys in automobiles. At ε˙=10−3s−1$\dot \varepsilon = {10^{ - 3}}{{\text{s}}^{ - 1}}$, the yield stress is only weakly dependent on T , whereas the maximum deformation resistance is strongly influenced by T . Normalization with the diffusion coefficient D and the shear modulus G of pure Mg produces an essentially unique relationship between normalized ε˙$\dot \varepsilon $ and normalized σ for the maximum deformation resistance. Comparison of this relation with that of pure Mg points to solute hardening by Al in AZ91 and AS21 and to additional particle hardening in AS21. In the range of high stresses the higher Al content of AZ91 leads to stronger solute hardening and consequently superior resistance against deformation and stress relaxation compared with AS21.

The influence of the spinodal decomposition in a Cu–Ni–Fe alloy on the microstructure and on the mechanical behaviour has been investigated by TEM and hardness measurements as well as by measurements of the yield strength. A correlation between the early stages of decomposition, defined by TEM investigations, and the yield strength has been found. By means of a model, based on thermal activation, the amount of hardness can be predicted for different times and temperatures.

The constitution of a total of 42 binary, ternary and quaternary alloys with compositions around those of the intermetallic compounds Ni 3 (Al, Si, Ti), (Ni, Co) 3 (Al, Ti) and (Ni, Co) 3 (Si, Ti) has been studied in the temperature range 973 to 1393 K. The object of this investigation was to find alloy systems in which fine, spherical, homogeneously distributed, disordered f.c.c. particles can be precipitated in an L1 2 -long-range ordered matrix. Such alloys are suitable model systems for investigating precipitation hardening and precipitation toughening of L1 2 -ordered intermetallic compounds. Two well suited alloys are: Ni 69 Co 9 Al 18 Ti 4 and Ni 58 Co 20 Si 10 Ti 12 .

The deformation behaviour of the single crystal superalloy SC16 has been studied under low cycle fatigue (LCF) loading at 1223 K, 1123 K and 298 K with a strain amplitude range of 0.4% to 1.0% and a constant strain rate of 10 −3 s −1 . Microstructural investigation by transmission electron microscopy of the deformed specimens revealed that the deformation mechanism depends on strain amplitude as well as on temperature of deformation. At large strain amplitudes at elevated temperatures, the γ′ precipitates are sheared by the matrix dislocations leading to stacking fault (SF) formation within the precipitates, while at small strain amplitudes deformation is confined within the γ matrix. At room temperature, the planar faults in γ′ precipitates are complex. The deformation under cyclic loading is compared with that under monotonic loading. The threshold stresses for shearing of γ′ precipitates at different temperatures are estimated.

Following a short evaluation of the advantages and weaknesses of materials based on NiAl for high temperature applications, it is proposed that compositionally graded design of components might be a viable solution to the problem of parts which are subjected to varying loads at both low and high temperature. Following this principle, specimens have been prepared by powder metallurgical techniques with composition and microstructure gradually changing from NiAl reinforced with ternary Laves phases (with Nb and Ta) for the hot section, to superalloy IN 738 for the low temperature section of the component. These specimens have been subjected to thermal fatigue testing between 200 and 1100 °C. Crack initiation and propagation have been observed by SEM and quantitatively evaluated. The advantage of the graded solution with respect to crack arrest is obvious. The different factors governing thermal fatigue behavior of graded components are discussed in detail.

In der Vorliegenden Arbeit wird ein Ansatz vorgestellt, bei dem durch die genaue Untersuchung der Mikrostruktur einer Flugturbinenschaufel auf die tatsächliche Betriebsbelastung geschlossen werden kann. Die Charakterisierung der einkristallinen Turbinenschaufel aus der Nickelbasis-Superlegierung CMSX-6 wurde mit Hilfe unterschiedlicher, sich ergänzender Methoden durchgeführt. Dabei handelt es sich um die konventionelle Raster- und Transmissionselektronenmikroskopie zur Untersuchung der Änderung der Gefügemorphologie. Zur Bestimmung der Gitterparameteränderungen und der damit Verbundenen lokal wirkenden Gitterverzerrungen wurden ein spezielles hochauflösendes Röntgenverfahren sowie die konvergente Elektronenbeugung im Transmissionselektronenmikroskop angewandt. Die Untersuchungen des Gefüges der Turbinenschaufel zeigten eine teilweise sehr starke Veränderung der Mikrostruktur von der im Ausgangszustand vorliegenden würfelförmigen Morphologie der γ′ -Ausscheidungsteilchen hin zu einer γ/γ+′-Plattenstruktur, die auch als Floßstruktur bekannt ist. Mit den beiden Verfahren zur Bestimmung der lokalen Gitterparameterverteilungen wurden die Gitterverzerrungen der beiden Phasen ermittelt, die auf weitreichende, verformungsinduzierte innere Spannungen sowie auf thermisch induzierte innere Spannungen zurückzuführen sind. Zur Bestimmung der lokal herrschenden Temperaturen wurden mit Hilfe von Finite-Elemente-Berechnungen die Gitterverzerrungen der in der Turbinenschaufel vorliegenden Mikrostruktur für verschiedene Belastungszustände ermittelt und mit den an der Turbinenschaufel gemessenen Verzerrungen verglichen. Die Untersuchungen haben erwiesen, daß sowohl die Mikrostruktur wie auch die Gitterparameterverteilung der Turbinenschaufel während der Betriebsbelastung im wesentlichen durch die Fliehkräfte und die auftretenden, sehr hohen Metalltemperaturen charakteristische Änderungen erfahren.