Volume 88, Issue 3

Microsegregation of Ti and Mn in a nitrogen-poor Fe –Ti– Mn–C alloy has been investigated experimentally and explained by a theoretical model. The formation of TiN precipitates in the residual melt of a nitrogen containing alloy has been detected and described by the model of homogeneous nucleation. The microsegregation is very high at the end of the solidification process, but the segregation zones are strongly widened during cooling in the delta ferrite. Precipitation of Ti nitride takes place in regions of the melt where the concentration product [%Ti][%N] exceeds the solubility product L TiN by more than a factor 10. The same supersaturation level is required by the theory of homogeneous nucleation.

By means of systematic measurements of single orientations the microtexture and the orientation relations in the area of partly recrystallized shear bands in deformed (112)[111̅] single crystals of pure copper have been investigated. In the analysis the distribution function of orientations and orientation differences of various kinds were calculated. In the calculations of these orientation characteristics the direct method and also for comparison the commonly used series expansion method were applied. To enable a comparison of the calculated results, both methods require some definite modifications which are explained in the paper. The results of the microtexture analysis are discussed concerning details of the texture changes and the orientation relations between the deformed matrix and the recrystallized grains.

The creep behavior of the near α Ti alloy IMI 834 was investigated in the temperature regime around 600 °C, comparing a fully lamellar and two bi-modal microstructures consisting of 15 and 40 vol.% equiaxed primary α phase in a lamellar matrix of transformed β phase. The width and length of the α lamellae of all three microstructures was varied by changing the rate of cooling from the recrystallization temperature. The width of the α lamellae decreased continuously with increasing cooling rate. The creep resistance exhibited the generally observed dependency to increase slightly with increasing cooling rate, to pass over a maximum and then to decrease drastically. This cooling rate dependent creep resistance is tentatively thought to be controlled by two competing mechanisms, an increasing amount of deformation strengthening as a result of the continuously decreasing interface distance and an increasing rate of recovery of dislocations at interfaces as a result of the drastically increasing density of interfaces which depends in a first approximation on the reciprocal interface distance. The observed lower creep resistance of the bi-modal as compared to the fully lamellar structure is thought to result from the alloy element partitioning effect reducing the strength of the lamellar matrix of the bi-modal in comparison to the fully lamellar microstructure.

The effects of cold working and recrystallization on the homogenization of a Cu-50 wt.% Ni alloy have been investigated by electron probe microanalysis and other complementary techniques. The occurrence of homogenization has been explained by the synergism of 3 factors: deformation alters interdendritic spacing and diffusion distances, crystalline defects enhance diffusion and microstructure sweeping by grain boundaries during recrystallization assists solute redistribution by grain boundary diffusion.

A molten Al-4 wt.% Cu as well as a Al-13 wt.% Si alloy have been mixed mechanically with particulate of SiC, Al 2 O 3 , or graphite. After the completion of mixing, each mixture was poured into a permanent mould to solidify. To overcome the problem of non-wettability that exists between the investigated particulate and the molten aluminum alloys the particulate was chemically treated by impregnation in a solution containing Na + ions. The loading of SiC or Al 2 O 3 particulate in the produced composites can be as high as 40 wt.%, and for graphite particles it can be 20 wt.%. The mixing time required to introduce and distribute the investigated particles into the molten matrix was as low as five minutes to recluce chemical reactions at the interfaces between them. Processing details and parameters controlling this technique are described. Metallographic examinations as well as tensile tests were carried out to characterize the microstructure, the distribution of the particles and the strength of these composites. The results display that the composites made by this technique have good microstructure and tensile properties.

The creep crack growth and the fractal dimension of grain-boundary fracture ( D f or D ̄ f ) were examined using surface notched specimens with different fractal dimensions of the grain boundaries ( D gb ) in austenitic 21Cr–4Ni–9Mn steel. The value of D f estimated in the scale range larger than about one grain-boundary length decreased with crack growth and with increasing bulk stress (gross section stress), while the value of D f was higher in the specimen with higher D gb value. The value of D ̄ f estimated in the scale range smaller than about one grain-boundary length was close to the value of D gb in each specimen and did not change with crack growth and with the bulk stress (gross section stress). The growth rate of creep crack was lower and the threshold stress intensity factor for crack growth was higher in the specimen with the higher D gb value.

The paper deals with the energy release rate G of nonlinear double cantilever beam (DCB) specimens. Often the equivalence of the basic definition G=−(1/B)∂H/∂l $G=-(1/B)\partial H/\partial l$( B : width of the specimen, H : Helmholtz energy, l : crack length) and the expression G*=(1/B)∂W/∂l ${{G}^{*}}=(1/B)\partial W/\partial l$( W : elastic energy) is assumed. The validity of this assumption is investigated. It is concluded that for the nonlinear DCB specimen the two expressions are not equivalent so that the basic definition for G should be used. The paper further comments upon an article [1] in which nonetheless G * is used for the nonlinear DCB specimen and, contrary to expectation, good agreement with experimental data is obtained. The reason for this agreement is indicated.

The relative partial molar Gibbs energy ΔG¯Al $\Delta {{\bar{G}}_{\text{Al}}}$of Al was measured in the Fe-rich range (0.4 ≤ x Fe ≤ 1) of the Al –Fe system between 620 and 800 °C by the emf method using solid galvanic cells with a single-crystal CaF 2 electrolyte. The relative partial molar excess Gibbs energy of Al in Fe at infinite dilution is ΔxsG¯Al∞=−76 ${{\Delta }_{\text{xs}}}\bar{G}_{\text{Al}}^{\infty }=-76$kJ/mol at 727 °C. The Gibbs energy of formation in the concentration G¯ $\bar{G}$range 0.36 ≤ x Fe ≤ 1 was obtained from ΔG¯Al $\Delta {{\bar{G}}_{\text{Al}}}$and from △ Fe after Gibbs-Duhem integration and yields at 727 °C: △ f G 0 〈Al1.80Fe〉 = −25 kJ/g-atom, ΔfG0〈AlFe〉=−26.1 kJ/g-atom and ΔfG0 ${{\Delta }_{\text{f}}}{{G}^{0}}\langle \text{AlFe}\rangle =-26.1~\text{kJ}/\text{g-atom and }{{\Delta }_{\text{f}}}{{G}^{0}}$〈AlFe 2.85 〉 = −18.8 kJ/g-atom. The composition Al 1,80 Fe is the Fe-rich phase boundary of Al 2 Fe at 727 °C. The composition AlFe 2.85 is the Fe-rich end member of the CsCl type solid solution ranging between x Fe = 0.49 and 0.74. X-ray microanalysis of an Al 0.241 Fe 0,759 sample annealed at 727 °C revealed a two-phase region with Al 0.26 Fe 0.74 precipitates in the Al 0.21 Fe 0.79 matrix which is the Al-rich end member of the W type Al xFe1- x solid solution at this temperature.

Thermochemical assessments of phase equilibria in the binary and ternary alloy subsystems which describe the addition of Ag to Al–Cu–Mg ternary alloys have been undertaken. The assessed Gibbs free energies show that the ternary intermetallic AlMgAg, silver-rich T-Al(Ag,Cu)Mg and AgMg 3 phases are introduced into Al-rich alloys of the Al–Cu–Mg ternary system for atomic ratios Ag/Mg < 1. For Ag/Mg > 1 the B2 phase AgMg and the Mg-based hcp solid solution become stable. Phase diagrams computed from the assessed Gibbs free energies for Al–Cu– Mg alloys with 0.1 at.% Ag agree with observations of the enhanced stability of the phase T-Al 32 (Ag,Cu,Mg) 49 relative to the h-Al 2 Cu phase compared with the ternary Al–Cu–Mg system, and the enrichment of Ag in the T phase in the Al–Cu–Mg–(Ag) quaternary system compared with its Cu content in the Al–Cu–Mg ternary system. The assessment accepts the Ω precipitation as a growth morphology of h-Al 2 Cu and strongly suggests that the ternary intermetallic AlMgAg rather than AgMg 3 is the precursor on which it nucleates.

Phase equilibria in selected phase regions of the Fe –Cr – Mn system were investigated at 1073, 1173, 1273 and 1373 K applying wavelength dispersive X-ray microanalysis and electron diffraction phase identification on foils. Binary and ternary phase equilibria were analysed in the (fcc + bcc), (fcc + sigma) and (fcc + bcc + sigma) regions.

A Knudsen effusion technique combined with mass-spectrometric analysis of the evaporated species was used to investigate the gaseous phase and to determine the thermodynamic properties of alloys of calcium with antimony in the temperature range from 933 to 1669 K. Concentration of antimony varied from 50.2 to 63.5 mol.%. The thermodynamic properties of the liquid phase were approximated by the association solution model, assuming that four types of associates Ca 3 Sb 2 , Ca 3 Sb, Ca 2 Sb and CaSb can exist in the melt. The Gibbs energies of formation of compounds from solid components were expressed by the equations: Δ f G C a 11 S b 10 = − ( 21884442 ± 14595 ) − ( 43.47 ± 12.34 ) T ,  and   Δ f G C a 5 S b 3 = − ( 919276 ± 5110 ) + ( 34.24 ± 4.75 ) T ,   J   mo l − 1 $$\begin{align}& {{\Delta }_{\text{f}}}G\left( \text{C}{{\text{a}}_{11}}\text{S}{{\text{b}}_{10}} \right)=-(21884442\pm 14595) \\ & \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ -(43.47\pm 12.34)T,\text{ and} \\ & \text{ }{{\Delta }_{\text{f}}}G\left( \text{C}{{\text{a}}_{5}}\text{S}{{\text{b}}_{3}} \right)=-(919276\pm 5110) \\ & \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ +(34.24\pm 4.75)T,~\text{J}~\text{mo}{{\text{l}}^{-1}} \\ \end{align}$$ The obtained data were used for the computation of the part of the Ca–Sb phase diagram, corresponding to the composition range studied.

By the methods of differential thermal, X-ray, microscopy and electron microprobe analysis phase equilibria in the Tirich corner of the Ti–Si –Al system were studied. Projections of solidus and liquidus surfaces as well as an isothermal section at 1300 °C were constructed. It was shown that at the solidus and 1300 °C Ti 5 (Si,Al) 3 (Z) phase coexists with all of the Ti– Al-based phases, resulting in wide two-phase (β + Z, α + Z, γ + Z) and narrow three-phase (α + β + Z, α + γ + Z) regions. The last two ones result from invariant four-phase peritectic and eutectic reactions, respectively. Coordinates of peritectic ( U 1 ) and eutectic ( E ) points were determined to be: U 1 – 1420 °C, ∼49Ti–4Si– 47Al; E – 1415 °C, 48Ti –4Si– 48Al. The solidus surface has its temperature maximum at 1545 °C corresponding to the invariant pseudobinary eutectic L↔β + Z with eutectic point at 65Ti – 8Si– 27Al. The isopleth at 10 at.% Si is given. The shape of the Ti 5 (Si,Al) 3 homogeneity range is discussed.