Band 99, Heft 7

The constitution of the ternary system Al–Si–Ti is reinvestigated over the entire composition range using X-ray diffraction, energy dispersive X-ray analysis, differential thermal analysis up to 1500 °C and metallography. The ternary phases τ 1 and τ 2 are confirmed and their crystal structures and solid state phase equilibria are corroborated. The compositions observed confirm previous literature. The phase τ 1 (I4 1 /amd, Zr 3 Al 4 Si 5 -type) melts incongruently at 953 °C < T < 1000 °C into L + τ 2 + TiSi 2 . The phase τ 2 (Cmcm, ZrSi 2 -type) melts incongruently at 1338 °C into L + TiSi + TiSi 2 . The liquidus surface is characterized by a eutectic trough, which descends monotonically from the pseudobinary maximum e max1 L = βTi + Ti 5 Si 3 at 1540 ± 5 °C through a series of transition reactions and ends in the ternary eutectic L = τ 1 + (Si) + (Al) at 573 °C.

The eutectic reaction between the NiZr and NiZr 2 compounds in the Ni – Zr binary phase diagram was evaluated according to the previous experimental measurements and thermodynamic calculations. A discrepancy between the calculations and the measurements exists. In order to determine definitely this eutectic temperature, Ni – Zr alloys of the compositions 65, 68, 72 and 74 wt.% Zr were prepared by melting in an arc furnace under a purified argon atmosphere. The phase transformation temperature was measured by Differential Scanning Calorimetry using both alumina and zirconia crucibles. At heating rates of 5, 10 and 20 K min − 1 , the eutectic temperature between the NiZr and NiZr 2 compounds was determined consistently to be 1022 ± 5 °C. Zirconia crucibles can be used in Differential Scanning Calorimetry measurements of zirconium-containing alloys but only below 1170 °C.

Tensile experiments on micro-sized polycrystalline copper wires with diameters ranging from 14 μm to 50 μm were performed using a recently developed fibre tensile module capable of high accuracy and flexibility. This module is able to fit into a scanning electron microscope for in-situ deformation studies. In this study the influence of the gauge length, wire diameter and grain size on the tensile properties is analysed. In-situ experiments performed in a scanning electron microscope revealed clearly that the occurrence of “pop-in” events in the load-displacement curves are related to slip. The results are compared to deformation studies of micro-sized copper samples in the literature and discussed taking in-situ scanning electron microscope observations into account.

NiAl polycrystals with different textures have been compressed at room temperature with and without confining pressure to strains of 6 and 30 %, respectively. Grain orientation dependent dislocation densities have been determined by X-ray line profile analysis using the multiple whole profile procedure. The dislocation density as well as the stored energy in hard grains, <100> oriented with respect to the compression axis, generally is about half of that in soft grains, which are randomly oriented. Taking into account appropriate individual dislocation contrast factors it has been found that in hard grains slip mainly takes place by <111> screw dislocations while in soft grains there is about a 50 % contribution of mixed type <100> dislocations. The results are discussed with respect to the inhomogeneity of deformation of plastically anisotropic materials and its effect on recrystallization.

The serrations observed on the stress–strain curves of a Fe-22Mn-0.6C wt.% steel present all the phenomenological features associated with dynamic strain ageing, including negative strain rate sensitivity and thermal activation. We show that the activation energy is, however, inexplicably low and that the dynamic strain ageing contribution to work-hardening is limited, i. e. mechanical twinning is the dominant strain hardening mechanism. Finally, novel in-situ experiments are used to study the dynamic behaviour of intense strain localisation bands which form on tensile samples. In spite of the localisation behaviour seen in tensile tests, stamped parts do not present any surface defects.

Using a new approach to modeling bainite nucleation, the kinetics of isothermal bainite formation have been calculated under the assumption of displacive growth. The nucleation rate is assumed to depend on the number density of potential nucleation sites N i , a factor λ accounting for autocatalytic nucleation, and an activation energy Q*. Compatible with the theory for athermal martensite nucleation, N i is assumed to be proportional to the driving pressure. Analogous to the Koistinen – Marburger model for martensite formation, the average volume of bainitic sub-units is assumed to be constant over the extent of the transformation and the growth of sub-units is very fast, and thus the change in fraction is directly related to the nucleation rate of bainite. The model results in an analytical expression for the fraction bainite as a function of time that contains only two adjustable parameters: a (temperature independent) autocatalytic parameter λ and a rate parameter κ, which has a temperature dependence that is mainly governed by Q*. The calculations are compared with experimental fraction curves measured isothermally with dilatometry for the carbon steels C45, C50 and C60 at a range of temperatures. From the best agreement between the calculations and the experimental data it follows that Q* decreases linearly with temperature, which is consistent with other bainite nucleation models. By austenitizing steel C60 at different temperatures it is found that λ depends on the austenite grain size: when the austenite grain size is increased, λ becomes larger.

An equation, based on the Eyring equation of viscosity, has been recently derived by Kaptay to extrapolate dynamic viscosity of pure liquid metals to the viscosity of liquid metallic alloys. The results of its application to ternary metallic alloys, calculated on the basis of the known concentration dependence of the integral enthalpy of mixing and excess molar volume of investigated alloys, are presented in this paper. As an example, viscosities of ternary Au – Ag – Cu alloys have been calculated in the temperature interval 1373 – 1573 K and compared with literature data.

In the present work, an unalloyed ductile cast iron containing 3.50 C wt.%, 2.63 Si wt.%, 0.318 Mn wt.% and 0.047 Mg wt.% was austempered for various times at both the conventional austenitizing temperature (single phase austenitic region) and the different intercritical austenitizing (partially austenitizing) temperature ranges to investigate the effect of parent austenite dispersion at the austenitizing temperature on the nature of austenite decomposition during austempering. A microstructure map was created to illustrate the transformation of products quantitatively as a function of austempering time for a particular austenitizing temperature and austempering time. It was concluded that in addition to proeutectoid and bainitic ferrite, the new ferrite (also called epitaxial ferrite) introduced into the partially austenitized structure during austempering and new ferrite formation and its content was dependent on the scale of the parent austenite dispersion at the austenitizing temperature, variation in austenitizing temperature and austempering time. The new ferrite formed only in the structure with coarse parent austenite dispersion.

The effect of trace additions (0 – 0.1 wt.%) of titanium on the microstructure, mechanical properties and creep resistance of an AM50 alloy have been studied. The microstructure was examined using optical microscopy and scanning electron microscopy. Results showed that the microstructures of AM50 alloy were obviously refined by trace Ti addition, although the actual content of Ti was only about 0.0025 – 0.0079 wt.%. The refining ability of titanium was ascribed to its large growth restriction factor, and it can be described as a constitutional undercooling driven mechanism. As-cast samples were subjected to tensile tests at room temperature, 150 °C and 200 °C. A considerable increase in tensile properties was observed due to grain refinement, especially the ultimate tensile strength and elongation. In addition, the creep resistance of AM50 alloy was dramatically increased by Ti addition.

Molybdenum-base alloys have excellent high temperature mechanical properties and joining of these alloys as heat sink materials is envisaged in many structural applications of fusion reactors. TZM is an important Mo-base alloy and diffusion bonding experiments were carried out in a hot vacuum press by sandwiching interlayers, Ti, Ni, Mo and Ta between two pieces of the TZM alloy in the temperature range 900 to 1200 °C. Detailed metallurgical analysis was carried out using optical microscopy, scanning electron microscopy and by electron probe microanalyser. All the sandwiched diffusion couples consisting of TZM/X/TZM, (X = Ti, Ni, Mo, Ti) showed defect-free microstructures at the bonded interface at the optimised bonding conditions. Ti seems to be a more suitable interlayer compared to other interlayers in terms of bonding temperature and applied pressure. The TZM/Ni interface was marked by the presence of the intermetallic compound NiMo. The concentration profiles for diffusion couples TZM/Ti were asymmetric, whilst TZM/Mo and TZM/Ta were symmetric about the Matano Interface. The interdiffusion coefficients calculated using the Boltzmann–Matano method for TZM/Ti diffusion couples showed decreases in the values with increase in the Mo concentration.

The microstructure and the magnetic properties of mechanically milled Sm–Co nano powders were investigated using X-ray diffraction, transmission electron microscopy and vibrating sample magnetometry. Microstructural studies revealed that after 40 h milling the powder nearly becomes amorphous with a small fraction of nanocrystallites embedded in the amorphous matrix. With increase in milling time up to 16 h, the diffracted crystallite size decreased exponentially and thereafter it remained constant. The grain size of 40 h milled powder was found to be ∼2 – 10 nm.

Martensitic AISI 420 stainless steel was plasma nitrided using a DC N 2 – H 2 pulsed discharge, after thermal treatment at two different tempering temperatures. Sliding wear behavior was determined by means of a ball-on-disk tribometer at room temperature and without lubrication. Wear volume was calculated from the information provided by an optical profilometer and wear scars were analyzed by Scanning Electron Microscopy. The results indicated that the nitrided samples exhibited improved wear resistance when compared to the untreated specimens, but corrosion resistance was diminished due to the presence of the outer nitrided layer. However, when this layer was removed mechanically considerable improvement in the corrosion resistance was achieved without impairing the wear resistance of the whole system, which was 200 % higher compared to the wear resistance of the non-nitrided specimens.

The investigations aim at the control of the formability and the strength of hot extruded brass alloys through choosing the appropriate hot extrusion parameters. Different α-brass and α/β-brass alloys, namely CuZn10, CuZn20, CuZn37 and CuZn40Pb2, were hot extruded and subsequently investigated by means of microscopy, X-ray diffraction and mechanical testing. The investigations show that the strength and formability of the extrudates are strongly affected by the hot extrusion parameters which determine the grain sizes and the textures of the extrudates. Additionally in the cases of CuZn37 and CuZn40Pb2, multiple phase transformations arise during the hot extrusion process and affect the resulting properties. The bcc β-phase content especially depends on the extrusion temperature, deformation degree and deformation rate.

Surface texture influences friction and transfer layer formation during sliding. In the present paper three well-defined surface textures, namely unidirectional, 8-ground, and random, were employed on steel plates. Experiments were conducted on a tribological couple – zinc pin against steel plate – using an inclined pin-on-plate sliding tester. It was observed that the occurrence of stick-slip motion, the transfer layer formation and the value of coefficient of friction as well as its two components, namely adhesion and plowing, depend primarily on surface texture. The effect of surface texture on coefficient of friction was attributed to the variation of the plowing component of friction for different surfaces. The plowing component of friction was highest for the surface texture that promotes plane strain conditions while it was lowest for the texture that favors plane stress conditions at the interface. Dimensionless quantifiable roughness parameters were formulated to describe the degree of plowing and hence the plane strain/stress type deformations taking place at the asperity level.

Nitriding of recrystallised and cold-rolled Fe-4.65 at.% Al alloy at 550 °C in an ammonia – hydrogen gas mixture of nitriding potential 0.104 atm − 1/2 indicated that nucleation of the thermodynamically (most) stable hexagonal (wurzite) AlN in the ferrite matrix of recrystallised specimen is difficult. The competition between the precipitation of nitrogen gas (dissociation of nitrogen supersaturated ferrite into pure ferrite and nitrogen gas) at the grain boundaries and the precipitation of hexagonal (wurzite) AlN led, via the coalescence of pores filled with nitrogen gas, to the emergence of microcracks along the grain boundaries and the formation of nitride-precipitate free zones near the grain boundaries of the recrystallised specimens. Upon continued nitriding the initially partially nitrided grains became fully nitrided and the microcracks disappeared.