Volume 89, Issue 10

Time-dependent inelastic deformation during high-temperature low-cycle fatigue has been studied by subjecting the ODS superalloy MA 6000 to discretized hysteresis loops with holdtimes at constant load. The inelastic deformation during the holdtime, recorded as a function of stress and temperature, was expressed by rate equations. This procedure gives insight into the material behaviour at various points of a stress-strain hysteresis. It is then applied to predict the stress-strain response under isothermal and thermomechanical fatigue loading. From the deformation analysis and a comparison between calculated and experimental results, conclusions concerning the internal stress evolution and deformation mechanisms acting during cyclic loading can be drawn.

The chemical compositions of γ and γ′ phases in the alloy SC16 were determined by means of energy-dispersive X-ray spectroscopy in the transmission electron microscope (TEM/EDS) after heat treatment at different temperatures. According to these measurements, single-phase alloys with compositions identical to those of the γ and γ′ phases in SC16 were produced. The temperature dependence of the lattice parameters of these alloys was deduced (i) from the measured values of the lattice parameters at room temperature and (ii) from the measured unidirectional thermal expansion up to 1323 K. These measurements show that the lattice misfit between the γ and γ′ phases in unconstrained condition at the different elevated temperatures at which their equilibrium composition in the alloy SC16 was determined, is positive. The results are compared with the misfit values in the alloy SC16 reported in literature for constrained and unconstrained condition and also with the predictions from models on rafting.

Self-similar coarsening of particles dispersed in a matrix is well known experimentally and theoretically as Ostwald ripening. In this paper we present experimental and numerical material on Cu–Co dispersions looking for the inverse effect: Does the copper matrix surrounding cobalt particles also coarsen self-similar during liquid phase sintering as the particles do? The experimental investigations were performed with Cu–Co alloys of various volume fractions of solid phase (0.25 to 0.7). Green compacts of both elemental powders were sintered at 1473 K for times between 5 minutes and 70 hours. The coarsening of the matrix was analysed in plane section of the specimens. The experimental results show that the matrix coarsens self-similar. A numerical modelling is presented clarifying the experimental results.

The electrical resistivity of a copper wire was measured repeatedly in situ between ambient temperature and 1064 °C and from this temperature down to ambient temperature. The measured curves were fitted by a 2nd order polynomial between 400 and 700 °C and extrapolated to high temperatures. The difference in electrical resistivity between the extrapolated and the measured curve is attributed to vacancy type defects, the concentration of which increases with increasing temperature. An effective formation energy of vacancies of EVeff = 1.20 eV $ E_{\rm{V}}^{{\rm{eff}}} \; = \;1.20\;{\rm{eV}} $ and an electrical resistivity of ρ V = 1.45 · 10 –7 Ωcm due to vacancies at the melting temperature is obtained. These results are approximately in agreement with electrical resistivity data already reported in 1977. The analysis of these data confirms the results obtained in the last forty years on the formation energy of vacancies, on the activation energy of migration of vacancies and of divacancies, and on the binding energy of divacancies. The values are E 1 V F = 1.03 ± 0.01 e V , E 1 V M = 1.06 ± 0.01 e V , E 2 V M = 0.64 e V , a n d E 2 V B = 0.20 e V $ E_{{\rm{1V}}}^{\rm{F}} \; = \;\left( {1.03 \pm 0.01} \right)\;{\rm{eV, }}E_{{\rm{1V}}}^{\rm{M}} \; = \;\left( {1.06 \pm 0.01} \right)\;{\rm{eV, }}E_{{\rm{2V}}}^{\rm{M}} \; = \;0.64\;{\rm{eV}},\;{\rm{and }}\, E_{{\rm{2V}}}^{\rm{B}} \; = \;0.20\;{\rm{eV}} $ . In addition the formation and the migration entropies of single vacancies and of divacancies are obtained. For the electrical resistivity of a vacancy a value of ρ 0 = 1.9 · 10 –4 Ω cm is obtained, assuming an effective vacancy concentration at the melting temperature of copper of CVeff = 7.6 · 10–4 $ C_{\rm{V}}^{{\rm{eff}}} \; = \;7.6\;\cdot\;10^{-4} $ (Hehenkamp et al.). We show that the Matthiessen rule is fulfilled and that the electrical resistivity of a vacancy is approximately the same at 4 K and at the melting temperature of copper.

The CHAPEL technique (continuous hydrogen analysis by pressure evaluation in liquids) measures directly the hydrogen pressure built up in an evacuated sensor volume immerged in an aluminum melt. In copper melts, the gas phase in the CHAPEL sensor consists of a mixture of hydrogen and carbon monoxide. A modified sensor system equipped with a PdAg diffusion cell measures continuously the total pressure and in addition the hydrogen partial pressure. The total pressure in the evacuated sensor volume increases slowly over many hours depending on the parameters defined by the experimental setup and the conditions of the melt. However, the hydrogen pressure approaches an equilibrium state after some time. Transport parameters for hydrogen and oxygen in the melt can be studied by doping the melt with gases introduced by a separate doping cell or after reactions of the melt or the coal bed on top of the melt with H 2 O or organic materials.

The thermal oxidation process of γ-TiAl is discussed regarding kinetic and thermodynamic aspects. A simple thermodynamic model allows the calculation of diffusion paths and leads to a better comprehension of the role of thermodynamic and kinetic factors in diffusion processes. The calculations allow a reproduction of the experimental results. Moreover, it is possible to predict results in regions not yet experimentally determined. The advantage of potential phase diagrams compared with concentration phase diagrams is discussed for diffusion processes using the Ti–Al–O system as example.

The phase equilibria of the ternary Al–AlCo–lNi system are demonstrated using isothermal sections at 1170, 1100, 1050, 900, 850, 730, and 600 °C. These are based on metallographic and TEM investigations and on several temperature-concentration sections, which are investigated using thermal analyses. The Scheil reaction scheme is used for explanation of the data. Eleven phases from the binaries Al–Co and Al–Ni and the three ternary phases Y 2 (Co 2 NiAl 9 ), X and D were found at room temperature. The binary high-temperature phase Co 4 Al 13 (h) is stabilized in the ternary down to room temperature by Ni addition (Y 1 ). The phase field of D is composed of several modifications which can hardly be distinguished by light microscopy. These modifications are not considered in the present paper. 45 three-phase regions can be derived from the reaction scheme in the Al–AlCo–AlNi subsystem.

The Mn–Sc binary system has been investigated experimentally in the entire composition range. Two new phases Mn 23 Sc 6 and MnSc 4 have been identified. Three invariant reactions have been determined: L ↔ Mn 23 Sc 6 + (Mn), L + Mn 2 Sc ↔ Mn 23 Sc 6 and L ↔ Mn 2 Sc + MnSc 4 . Based on the experimental data from this work and the sparse literature data, a thermodynamic assessment has been performed. The calculated phase diagram is in fair agreement with the measured values.

The partial and integral enthalpies of mixing of ternary Al– Cu–Zr alloys have been measured by high temperature calorimetry at 1450 to 1485 and 2010 to 2080 K. The experimental data reflect the strong chemical interaction between the components. The most negative integral enthalpy of mixing (– 51.3 ± 1.5 kJ mol –1 ) shows up in the binary Al–Zr system at about 47 at.% Zr. A least-squares treatment of the enthalpy of mixing results in the following relationships (in kJ mol –1 ): Δ H = x C u x Z r α C u − Z r + x C u x A l α C u − A l + x A l x Z r α A l − Z r + x C u x A l x Z r α C u − A l − Z r $$\begin{array}{l} \Delta H\; =& \;x_{{\rm{Cu}}} x_{{\rm{Zr}}} \alpha_{{\rm{Cu-Zr}}} + x_{{\rm{Cu}}} x_{{\rm{Al}}} \alpha _{{\rm{Cu-Al}}} + x_{{\rm{Al}}} x_{{\rm{Zr}}} \alpha_{{\rm{Al-Zr}}} \\ & + x_{{\rm{Cu}}} x_{{\rm{Al}}} x_{{\rm{Zr}}} \alpha _{{\rm{Cu-Al-Zr}}} \\ \end{array}$$ α Cu− Zr=− 40.9+675.0xCu− 926.6xCu2+2541.7xCu3− 2895.3xCu4+1202.6xCu5 $$\begin{array}{l} \alpha _{{\rm{Cu-Zr}}} \; = \;& - 40.9 + 675.0\;x_{{\rm{Cu}}}-926.6\;x_{{\rm{Cu}}}^{\rm{2}} + 2541.7\;x_{{\rm{Cu}}}^{\rm{3}} \\ &-2895.3\;x_{{\rm{Cu}}}^{\rm{4}} + 1202.6\;x_{{\rm{Cu}}}^{\rm{5}} \\ \end{array}$$ α Al− Cu=− 75.6− 63.8xAl+231.8xAl2− 130.6xAl3 $$\alpha _{{\rm{Al-Cu}}} \; = \;-75.6-63.8\;x_{{\rm{Al}}} + 231.8\;x_{{\rm{Al}}}^{\rm{2}} -130.6x_{{\rm{Al}}}^{\rm{3}}$$ α Al− Zr=− 177.3− 170.0xZr+230.1xZr2 $$\alpha _{{\rm{Al-Zr}}} \; = \; - 177.3-170.0\;x_{{\rm{Zr}}} + 230.1\;x_{{\rm{Zr}}}^{\rm{2}}$$ α Al− Cu− Zr=− 459.5− 1395.8xZr+1576.7xZr2− 1853.9xAl+1301.9xAl2+986.9xZrxAl $$\begin{array}{l} \alpha _{{\rm{Al-Cu-Zr}}} \; = \;&-459.5-1395.8\;x_{{\rm{Zr}}} + 1576.7\;x_{{\rm{Zr}}}^{\rm{2}} \\ &- 1853.9\;x_{{\rm{Al}}} + 1301.9\;x_{{\rm{Al}}}^{\rm{2}} + 986.9\;x_{{\rm{Zr}}} x_{{\rm{Al}}} \\ \end{array}$$ The entropy of mixing is estimated on the basis of the ?H data using an empirical relation for excess entropy of mixing.

Thermochemical calculations and considerations have been applied to describe the interdiffusion and formation of interface phases in the diffusion couple Ti–SiC for the case of a thin Ti layer on a SiC substrate. With the aid of activity diagrams and relevant thermochemical calculations possible diffusion paths are discussed and their progress simulated. It is shown that the emergence and disappearance of TiC 1- x during interdiffusion can be interpreted thermochemically. TiC 1- x and Ti 5 Si 3 C x appear as initial phases to be joined later by Ti 3 SiC 2 . Then TiCn 1- x and Ti 5 Si 3 C x disappear and TiSi 2 emerges. The final decomposition of TiSi 2 and Ti 3 SiC 2 can either take place with excess carbon available from the substrate or the gas phase or by simultaneously abstracting silicon from the system. In both cases TiSi 2 decomposes first, followed by Ti 3 SiC 2 . Besides SiC, only (Ti, Si)C 1- x is then left as a stable compound.

Bei quasistatischer Zug- und Druckverformung unterscheiden sich die Nennspannung-Totaldehnung-Kurven kugelgestrahlter und ungestrahlter Proben aus vergütetem 42 CrMo 4 außer in der Streck- und Stauchgrenze sowie in den Lüdersdeformationsbereichen nur wenig voneinander. Ausgehend von einem Rand-Kern-Verbundmodell wird unter Berücksichtigung der in den oberflächennahen Werkstoffbereichen kugelgestrahlter Zylinderproben auftretenden axialen und tangentialen Eigenspannungstiefenverteilungen ein Mehrbereichsmodell entwickelt, das sowohl das einachsige Zug- als auch das Druckverformungsverhalten dieser Proben quantitativ zu beschreiben gestattet. Die berechneten Übergänge von elastischen in elastischplastische Verformungszustände stimmen vorzüglich mit den experimentellen Beobachtungen überein. Auch die mit dem Modellansatz berechneten zug- und druckverformungsbedingten Änderungen der axialen uund tangentialen Makroeigenspannungen sind in bester Ubereinstimmung mit den entsprechenden Meßwerten.