Volume 97, Issue 4

Partitioning phase transformations in the solid state are principally subjected to two processes that take place: the redistribution, through long-range diffusion, of the partitioning element, and the lattice transformation taking place at the interface. Consequently, the usual approximation to consider one of these two processes as controlling the rate of the phase transformation is of limited accuracy. For a more accurate description, the so-called mixed-mode character of partitioning phase transformations is to be taken into account. In the present study, it is shown that the mixed-mode character can be quantified and that it has a significant effect on the kinetics. By means of examples involving either substitutional (Mo in Ti) or interstitial (C in Fe) partitioning elements, it is shown that a gradual change of the character of the transformation occurs during the phase transformation, shifting from initially interface-controlled (which implies the largest interface velocity) towards more diffusion-controlled.

The liquid – liquid interfacial tension and the density difference of the coexisting phases in the immiscible alloy (Al 34.5 Bi 65.5 ) 95 Si 5 (wt.%) have been measured between 873 and 1238 K. The interfacial tension equals (73.0 ± 3.6) mN/m at 873 K and it decreases with increasing temperature proportionally to (1 – T/T C ) δ . The critical-point exponent δ = 1.38, and the critical temperature T C = 1410 K were determined from the experimental data.

Surface tension and density measurements by maximum bubble pressure and dilatometric techniques were carried out in an extensive range of temperature on liquid alloys close to the ternary eutectic Sn3.3Ag0.76Cu with different Sb content. It has been found that the addition of Sb to Sn, Sn–Sb, to binary eutectic Sn–Ag and to Sn3.3Ag0.76Cu decreases the surface tension and density. The values of the surface tension calculated by the Butler model and by the method based on the binary alloys surface tension data with the ternary and quaternary correction factors were compared with the experimental results. The best agreement between the measured and the calculated values was observed for the model comprising the binary data with the correction factors.

The velocity of sound has been measured as a function of temperature to investigate structural changes in the molten As–Sb alloy. The overall profile of temperature dependence of the sound velocity changes smoothly from Sb to As. The results indicate that a tendency to change structures in the supercooled state develops with increasing As fraction, and confirm the previous conclusion that molten As changes rapidly its structure in the undercooled state and is already in the high-temperature side of an assumed liquid – liquid transition at the melting point. By confirming the structural transition in molten As, it is possible to mark the elements in the periodic table that undergo a crossover liquid –liquid transition around the melting point under the saturating vapor pressure. Observed trends for liquid – liquid transitions are discussed on the basis of the Peierls distortion mechanism.

A simple thermodynamic model, based on the formation of self-associates, has been used to correlate the bulk and surface properties of mixing of liquid Ga–Tl and Zn–Cd alloys at temperatures of 973 K and 900 K, respectively. The model enables us to explain the positive deviation of the properties of mixing of the two liquid alloys from the ideal solution behavior. Our calculations indicate that the degree of phase separation in Zn–Cd, as measured using the Warren –Cowley short-range order parameter, is less than that in Ga–Tl.With regards to the surface, we show that the surface decomposition in Zn–Cd liquid alloys is greater than in Ga–Tl. We explain the contrast in the bulk and surface properties in terms of the possible roles of the entropic and enthalpic effects on the properties of mixing.

The origins of the S-shaped entropy of mixing and order – disorder transformation in Cd –Na liquid alloys have been investigated within a first-principle approach of a pseudo-potential perturbation scheme with the Gibbs–Bogoliubov variation technique.We find that the deepening of the structure part of the entropy around the stoichiometric composition ( c Na ≈ 0.33) is responsible for the S-shaped entropy and suggests the existence of short-range order in the liquid state. The role of the local-field correction function, G ( q ), of conduction electrons on the entropy of mixing is examined. The impact of G ( q ) on partial structure factors is more visible towards low q ( q → 0) than at higher q .

Glass transition behavior of metallic glasses under some extraordinary conditions (such as under high pressures) remains unexplored. Conventional measurements of glass transition temperature, T g , are very difficult to perform under these extraordinary circumstances. In the present paper, we introduce an indirect approach to characterize glass transition, using enthalpy recovery experiments. With annealing deeply relaxed glassy samples and subsequent DSC measurements, a variation of enthalpy change upon heating with annealing temperature can be obtained. The variation of enthalpy change, a signature of glass transition, was found to correlate well with the directly measured DSC curves for the glass transition. This unique method was successfully applied in determining T g of several metallic glasses under hydrostatic high pressures and compression stresses.

Several Pd-base alloys display a large tendency for glass formation that renders them especially suitable for investigations concerning (metastable) equilibrium properties of the deeply undercooled liquid including the glass transition, since the detrimental interference of premature crystallization can be avoided rather easily compared to other alloy systems. Here, thermodynamic, dynamic, and transport properties were analysed and compared to an Al-rich marginal glass former in the context of a possible relation between fragility, thermodynamic excess and kinetic stability against crystallization. Additionally, the possibility for liquid-phase separation occurring in the undercooled state of bulk glass-forming alloys is analysed as a function of thermal history, and critically discussed with respect of alternative mechanisms that are often summarized as ‘short-range ordering’.

The different transformations occurring during the crystallisation of amorphous Mg 80 Cu 10 Y 10 and Mg 83 Ni 9 Y 8 alloys have been elucidated. The formation of several unknown metastable phases was observed. The present work confirms that such phases exhibit long-period structures originated by the periodic alignment of (Y, TM)-layers (TM: transition metal). The long-period stacking-ordered phase, identified in the crystallised Ni-containing alloy, is stable at higher temperatures than that in the Cu-containing alloy.

The crystallization behavior of undercooled liquid Pd 40 Cu 30 Ni 10 P 20 metallic alloy has been studied with the aid of differential scanning calorimetry, optical metallography, and electron microscopy methods. The crystallization occurs as a single-stage process starting preferably on the surface of casting voids in the sample volume and on the outer sample cutting planes. The linear crystal growth velocity of crystallizing cells was measured at several constant temperatures in the range 590 –620 K. It is constant at constant temperatures and its temperature dependence is in good agreement with the theoretical concepts of Uhlmann using the temperature dependence of the viscosity of the alloy studied. A differential scanning calorimetry study of isothermal overall volume transformation kinetics of samples pre-annealed at 600 K for 600 s was carried out at several constant temperatures. The experimental data are interpreted on the basis of the equation of Kolmogorov-Avrami-Johnson-Mehl for the case of three-dimensional growth of a fixed number of pre-existing nuclei. The possibility to revitrify the crystallized samples by their melting in differential scanning calorimeter, followed by quick cooling with cooling rate of 5 K τ – 1 is shown. The volume transformation kinetics of revitrified samples has been also studied.

Magnesium-ytterbium alloys were investigated in order to determine the enthalpy of formation of YbMg 2 and to check the phase diagram of the system. The system was investigated by using differential scanning calorimetry, electron microscopy, electron microprobe analysis, and X-ray diffraction analysis. The enthalpy of formation was measured by high temperature drop calorimetry. A value of – 13.5 ± 0.5 kJ/mol of atoms was obtained for the standard formation enthalpy of YbMg 2 .

A thermodynamic description of the Mg–Al – Sr ternary in the Mg– Al-rich region was developed based on the microstructures, the crystal structures and compositions of the phases in three samples in the as-cast and 673 K annealed states. The gross compositions of the three alloys were identified for investigation based on a calculated isotherm using the description obtained from those of the constituent binaries via extrapolation. The calculated liquidus projection using the developed description was found to be consistent with the primary phases of solidification obtained from two additional alloys in the as-cast state. In addition, the model-calculated isotherm and liquidus projection are able to rationalize results reported in a recent publication for this ternary. This paper demonstrates the power of using a preliminary calculated phase diagram for selecting key alloys for experimentation, on basis of which a reliable thermodynamic description can be developed.

Zinc vapor pressures over binary Pt–Zn alloys were determined in the composition range from 63 to 84 at.% Zn for temperatures between 840 –1150 K using an isopiestic method. The phase boundaries of four different phase fields in this composition range could be fixed based on the experimental data. Thermodynamic activities and partial molar enthalpies of zinc within these phase fields were derived from the vapor pressure data and the composition dependence of the zinc activity was calculated for 1023 K. The integral Gibbs energy at 1273 K was derived as a function of composition by Gibbs–Duhem integration, combining the activity data of the present investigation with earlier results from the literature.

The enthalpy of mixing of the ternary Cu –Sn–Zn system was measured using a Calvet-type calorimeter. The investigations were carried out along the three cross sections Cu : Sn = 1 : 2, 1 : 1, and 2 : 1. Zn was added in the concentration range between 0 and 50 at.%. The calorimetric results were compared with EMF data investigated in an earlier work by Peng and Mikula [1]. The results of the calorimetric measurements show very good agreement in the copper-rich region, compared with the EMF measurements, but in the Cu : Sn = 1 : 2 cross section the results do not fit perfectly. The difference between both measuring methods was at the most Δ H = 300 J/mol at the cross section Cu : Sn = 1 : 1 and Δ H = 150 J/mol at the Cu : Sn = 2 : 1 cross section.

The partial enthalpies of dissolution of pure Ni,Wand Ta in liquid ternary Ni –Ta –W alloys have been determined at (1773 ± 5) K using a high temperature isoperibolic calorimeter. Measurements were performed in Ni-rich alloys (from 80 to 100 at.% Ni) along sections with constant Ta .W atomic ratios 1 : 0, 2 : 1, 1 : 2, and 0 : 1. The partial enthalpies and thereby the integral enthalpy of mixing of these ternary alloys are calculated from the partial enthalpies of dissolution using SGTE Gibbs energies for pure elements as reference. The obtained thermochemical data confirm that in the investigated Ni-rich alloys the binary interactions between Ta and W as well as the ternary Ni –Ta –W interactions are negligibly small. Due to this the variation of the integral enthalpy of mixing of the ternary alloys is well described as linear combination of the constituent Ni –Ta and Ni –W binaries. Such behaviour of the ternary liquid alloys is related to a very low probability of new ternary stable phases to occur in solid state. This prediction is confirmed by differential thermal analysis, scanning electron microscopy, and transmission electron microscopy of the as-solidified and annealed samples obtained as last alloy compositions in the series of calorimetric dissolution.

Site preference and thermodynamic properties of the ternary Laves phase Ti(Fe 1 – x Al x ) 2 with MgZn 2 -type have been studied employing Rietveld refinement of X-ray and neutron powder diffraction data, X-ray single crystal data, and isoperibolic drop calorimetry techniques. A detailed relation between lattice parameters of Ti(Fe 1– x Al x ) 2 and the Al content in the Laves phase has been presented. The Rietvelt refinement revealed that Ti atoms occupy the 4f site in the MgZn 2 -type, whilst Fe and Al atoms randomly share the 2a and 6 θ site. Heat of formation has been measured for Ti(Fe 0.5 Al 0.5 ) 2 . For the ab initio density functional theory applications, a large number of structural models were investigated to calculate the concentration dependent (Al) heats of formation, magnetic structural stabilities, lattice parameters, and site occupancies, which are in good agreement with experiment.

Systematic vaporisation studies on the Ni –Al system with 0.32 ≤ x (Ni) ≤ 0.50 were carried out in the temperature range 1077 – 1592 K by Knudsen effusion mass spectrometry. While Al(g) was the vapour species observed over all the samples, Ni(g) was also detected over samples with the compositions x (Ni) = 0.50 and 0.49. From the p (Al) – T relations, activities of Al, and subsequently activities of Ni (by Gibbs–Duhem integration) were obtained. Moreover, partial and integral molar enthalpies and Gibbs energies of formation were evaluated. The variation of activities and other thermodynamic properties with composition are discussed. The enthalpy and Gibbs energy of formation of Ni 0.5 Al 0.5 (s) at 298.15 K were obtained as ΔfHmo=${{\Delta }_{\text{f}}}H_{\text{m}}^{\text{o}}=$– 62.1 ± 4.2 kJ mol –1 and ΔfGmo=−57.7±1.2  kJ${{\Delta }_{\text{f}}}G_{\text{m}}^{\text{o}}=-57.7\pm 1.2~\,\text{kJ}$mol – 1 , respectively. This is the first time that a systematic mass spectrometric investigation has been carried out on the aluminium rich region of Ni –Al system where the vacancy concentration in the NiAl phase can be as high as 10 %.

In this paper, the basic principles of and some recent advances in rapid solidification are briefly considered, with main emphasis on the formation of single-phase metastable states during meltquenching in Ni-rich Zr –Ni and Hf–Ni alloys. Factors controlling nanocrystalline-forming ability are discussed in detail, with particular view on the preparation of single-phase nanocrystalline alloys by the melt-quenching process.

The processing temperature of chemical and physical vapor deposition could be successfully lowered by applying inductively coupled plasma without deteriorating the film quality. Despite the low process temperature, the deposition rate was higher than that observed in conventional deposition processes, and the impurity content could be kept low. Some examples of inductively coupled plasma applicationse. g. TiB 2 , TiN, TiO 2 , and SnO 2 films are shown.