Effects of ultrasonic irradiation and cooling rate on the solidification microstructure of Sn–3.0Ag–0.5Cu alloy

A comparative study on the microstructures of Sn–Ag–Cu alloy ingots grown by ultrasound-assisted solidification was carried out with a specific focus on the limits on the ultrasonic processing depth and time imposed by the cooling rate during the melt solidification. During air-cooling, increasing the ultrasonic power reduced the undercooling temperature and increased the solidification time, leading to β-Sn phase fragmentation from a dendritic shape into a circular equiaxed shape. The grain size was decreased from approximately 300 μm to 20 μm. When the cooling rate was increased from 4 °C/s in air to 20 °C/s in water, the macro-undercooling temperature was more greatly reduced by an increase in ultrasonic power, but the solidification time seemed to change only slightly because only a limited period for ultrasonic processing was permitted in the melt. Under both cooling rates, the microstructures were inhomogeneous along the processing depth. The functional depth and period for ultrasonic cavitation and acoustic steaming contributed to the differences in the solidification microstructures.     

Effect of Y2O3 crucible on contamination of directionally solidified intermetallic Ti–46Al–8Nb alloy

The effect of Y₂O₃ crucible on contamination of Ti–46Al–8Nb (at.%) alloy directionally solidified (DS) in a Bridgman-type apparatus was studied. Directional solidification experiments were performed in dense Y₂O₃ crucibles using different growth rates, melt temperatures and various reaction time between the melt and the crucible. The main mechanism responsible for the contamination of the DS samples is diffusion controlled dissolution of the Y₂O₃ crucible in the melt which leads to an increase of oxygen and yttrium content in γ(TiAl) + α₂(Ti₃Al) matrix and precipitation of non-metallic particles in interdendritic region. Transmission electron microscopy (TEM), energy-dispersive spectroscopy (EDS) and X-ray diffraction analysis (XRD) showed that these particles are Y₂O₃ phase. The oxygen content and volume fraction of Y₂O₃ particles increase with increasing melt temperature and reaction time. The activation energy for increase of oxygen content is calculated to be Q_O = 412.1 kJ/mol and the kinetics of this process is suggested to be controlled by long-range diffusion with the oxygen content exponent of 3. The activation energy for Y₂O₃ particle formation is calculated to be Q_Y = 421.8 kJ/mol and the time exponent is determined to be m = 0.55. Vickers microhardness measurements in lamellar γ + α₂ matrix without Y₂O₃ particles can be used as an indirect evidence of the level of contamination of DS samples with statistically identical α₂–α₂ interlamellar spacing.     

Effect of a high magnetic field on the morphological instability and irregularity of the interface of a binary alloy during directional solidification

The influence of an axial high magnetic field (up to 12 T) on the stability and morphology of the liquid–solid interface of a binary alloy has been investigated experimentally during the directional solidification of the Al–0.85 wt.% Cu and Zn–2.0 wt.% Cu alloys. Experimental results indicate that the high magnetic field caused the breakdown of a planar interface into cellular undulations and the formation of an irregular shape. Specifically, for the Zn–2.0 wt.% Cu peritectic alloy, a wavy band-like structure appears under a high magnetic field. Moreover, the high magnetic field promoted the enrichment of the solute Cu element in the diffusion boundary layer. A theory about the magnetization and solute build-up in the diffusion boundary layer under a high magnetic field for a binary alloy has been proposed. This magnetization and solute build-up could be partly responsible for the breakdown of the planar interface and the formation of the band-like structure in a peritectic alloy. Moreover, the stresses in the solid near the interface under a high magnetic field were analyzed, measured and simulated numerically. It is suggested that they are responsible to the interface irregularity, and are also capable of inducing the interface instability.     

Evolution of elongated pores at the melt–solid interface during controlled directional solidification

The evolution of elongated gaseous pores during directional solidification has been examined through a theoretical model and compared to previous experimental findings from the literature. The model is based on the observation and interpretation of a wedge-shaped structure on the solid side that extends beyond the melt–solid interface and wraps around the bubble at its equator. At the tip of the wedge is the meeting point of the melt, solid and gas. The model takes into account the competition around the meeting point between the law of segregation of the solute at the melt–solid interface on one hand, and Henry’s law, which governs the concentration of the solute at the melt–pore interface at a given pressure, on the other. It predicts VR² = constant, where V is the processing speed and R is the pore radius, and agrees well with reported experimental data.     

Columnar to equiaxed transition in directional solidification of inoculated melts

Experiments on transient directional solidification were carried out with Al–3% Si and Al–7% Si alloys to study the modification by melt inoculation of the as-cast ingot macrostructure with columnar and equiaxed zones into a completely refined equiaxed structure. Without inoculation the macrostructure consists of typical columnar and equiaxed zones, separated by a relatively thin region of columnar to equiaxed transition. As the inoculant Al–3% Ti–1% B was added to the melt in a series of experiments relatively small equiaxed grains were observed in the as-cast macrostructure. For larger inoculant additions the number of these equiaxed grains increased. The macrostructure became completely refined and equiaxed as the Ti concentration in the melt increased in the small range 0.002 < %Ti < 0.01 for Al–3% Si, and of 0.029 < %Ti < 0.075 for Al–7% Si. The larger inoculant additions for the Al–7% Si alloy are attributed to Si poisoning of the inoculant. This macrostructure modification occurred with an increasingly large fraction of fine equiaxed grains mixed with columnar grains, rather than by a significant decrease in the length of the columnar grains. Solidification paths were calculated from the measured cooling curves and superimposed on dimensionless growth maps, enabling good qualitative predictions of the macrostructure, especially the existence of the mixed region. The maps, however, predict that the position where columnar grains are completely blocked is closer to the ingot base than that observed experimentally. This discrepancy might be related to the restrictive hypothesis assumed to construct the growth maps, not valid in the present experiments, of steady-state solidification conditions.     

矩形冷坩埚定向凝固过程中钛铝熔体的流场数值计算(英文)

对矩形冷坩埚定向凝固钛铝合金熔体流场开展数值模拟研究。结合实验结果,建立熔体流场的3-D有限元模型,研究不同电源参数下熔池内流动特性。计算结果表明:熔池内存在着复杂的循环流动,在固液界面前端存在着较为强烈的径向对流,并在中部合流。熔体流动随着电流强度的增强而增强,但是宏观流动形貌并没有改变。当电流为1000A时,熔池内最大流速为4mm/s,固—液界面前端达到3mm/s。当频率从10kHz变化到100kHz时,熔池流动形貌发生明显改变,分析其影响机制。对于冷坩埚定向凝固,存在着一个最佳频率。     

Rapid phase transformation kinetics on a nanoscale: Studies of the α → β transformation in pure,nanocrystalline Ti using the nanosecond dynamic transmission electron microscope

Using the unique capabilities and high time resolution of dynamic transmission electron microscopy (DTEM), the fast kinetics of α → β-phase transformation in nanocrystalline Ti films were investigated using single-shot electron diffraction and bright-field TEM images. From quantitative analysis of the diffraction patterns, the transformation rates were determined for temperatures between transition start (1155 K) and melt temperature (1943 K). The experimental data were summarized in a time–temperature-transformation (TTT) curve with nanosecond time resolution. Theoretical TTT curves were calculated using analytical models for isothermal martensite and available thermodynamic data. Above 1300 K, there is excellent agreement between the experiment and the discrete–obstacle interaction model, suggesting that the nucleation rate and thermally assisted motion of the martensite interface are controlled by interface–solute atom interactions. However, theory predicts much slower transformation rates near the transition temperature than experiment. Experimental data fits using the Pati–Cohen model suggests that an increase in autocatalytic nucleation may partially account for the fast transformation rates at lower temperatures.     

Preparation of Al2O3–ZrO2–SiO2 ceramic composites by high-gravity combustion synthesis

By a furnace-free technique of high-gravity combustion synthesis, Al₂O₃–ZrO₂–SiO₂ ceramic composites were prepared via melt solidification instead of conventional powder sintering. The solidification kinetics and microstructure evolution of the ceramic composites in high-gravity combustion synthesis were discussed. The phase assemblage of the ceramic composites depended on the chemical composition, where both (Al₂O₃ + ZrO₂) and (mullite + ZrO₂) composites were obtained. The ceramic composites consisted of ultrafine eutectics and sometimes also large primary crystals. In the (mullite + ZrO₂) composites, two different morphologies and orientations were observed for the primary mullite crystals, and the volume fraction of mullite increased with increasing SiO₂ content. The ceramic composites exhibited a hardness of 11.2–14.8 GPa, depending on the chemical composition and phase assemblage.     

Microsegregation in high Nb containing TiAl alloy ingots beyond laboratory scale

Microsegregation in big ingots of Ti–45Al–(8–9)Nb–(W, B, Y) alloy had been studied. The composition and microstructural morphology of the large ingot exhibited significant microinhomogeneity. Three types of microsegregation were observed in as-cast microstructure of the large ingot. First is the solidification segregation (S-segregation) at interdendritic area, in which the composition is characterized by higher Al, B (boride), and Y (oxide) contents and lower Nb and W contents. Second is the β-segregation at the boundary and triple junctions among α grain due to the phase transformation of β → α. The composition at the segregation area is characterized by higher Nb and W additions that lead to the formation of β particles and γ phase. Third is the α-segregation that forms local lamellar structure composed of β, γ and α plates due to phase transformation of α → α₂ + β + γ. The microsegregation for the PAM ingot is lower than that for SM ingot in terms of the volume fraction of β phase. The reason is that the PAM melting can offer better control of pouring temperature and rather fast cooling rate by water-cooled copper crucible.     

Effects of ultrasonic treatment on microstructure and tensile strength of AZ91 magnesium alloy

In this study, the effects of ultrasonic treatment on microstructural features and tensile strength of AZ91 magnesium alloy were investigated. AZ91 melts were subjected to ultrasonic waves of different power levels for 5 min using an ultrasonic device with frequency of about 20 kHz and maximum power of 600 W and cast in sand moulds. The results showed that ultrasonic treatment of the melt prior to casting had a significant effect on the size and sphericity of α-Mg dendrites as well as on the size, continuity, sphericity and distribution of intermetallic particles formed during cooling and solidification of the alloy. Increasing the applied ultrasonic power resulted in smaller, more rounded and better distributed grains and particles. The microstructural effects were mainly attributed to the cavitation and streaming phenomena which took place during ultrasonic treatment in the melt. Tensile strength of the alloy was significantly improved by ultrasonic treatment of the melt. Discontinuity and refinement of Mg₁₇Al₁₂ particles in the ultrasonically treated samples is thought to be the main reason for this improvement. The paper also examines different possible mechanisms responsible for microstructural modification of different phases under ultrasonic treatment conditions.     

Influence of an axial high magnetic field on the liquid–solid transformation in Al–Cu hypoeutectic alloys and on the microstructure of the solid

The influence of an axial high magnetic field (up to 10 T) on the liquid–solid interface morphology and microstructure of the solid has been investigated experimentally during Bridgman growth of Al–Cu hypoeutectic alloys. It is found that the field causes the interface to become destabilized and irregular, and promotes planar–cellular and cellular–dendritic transformation. The field has a great influence on the cellular and dendrite array morphology. Indeed, the field causes severe distortion in the cellular array and enhances cell branching. The field makes the morphology of the dendrite array more complex and, with the increase of the magnetic field intensities and decrease of the growth velocities, the dendrites become broken and orientate with the 〈1 1 1〉-direction along the solidification direction instead of the 〈1 0 0〉-direction. Furthermore, the field also enlarges the primary dendrite spacing and promotes the branching of the dendrites to form higher-order arms. The above phenomena may be attributed to the change of the equilibrium partition coefficient k and the liquidus slope m_L caused by the field, the magnetic anisotropy of the α-Al crystal and the flow created by the thermoelectromagnetic convection.     

Competition of the primary and peritectic phases in hypoperitectic Cu–Sn alloys solidified at low speed in a diffusive regime

Directional solidification experiments on hypoperitectic Cu–Sn alloys have been performed at very low velocity in a high thermal gradient to ensure planar front growth of both phases. The diameter of the sample has been reduced to 500 μm to strongly reduce convection. Lamellar and fibrous peritectic cooperative growth of the primary α- and peritectic β-phases has been observed on length spanning several millimeters. For the first time in a high solidification interval peritectic alloy, a quenched interface of both phases in contact with the liquid has been obtained. An unexpectedly high volume fraction of the primary phase, which furthermore fluctuates over time, has been observed. This is attributed to the transient state of the (α + β) growth front to a steady state and the associated evolution of the large diffusion layer ahead of the solid–liquid interface.     

Cyclic twin nucleation in tin-based solder alloys

The microstructure and Sn crystal orientations of lead-free solder alloys such as near-eutectic SnAgCu have a significant influence on the mechanical response of a solder joint to service conditions. Thus solidification processes were examined in SnAgCu solder joints. Distinct evidence of sixfold cyclic growth twinning of Sn during solidification from the melt was observed in Sn–Ag, SAC and Sn–Cu solders. Three orientations of Sn grains, each having a common 〈1 0 0〉 direction, were found in each of these systems, though the morphologies of these cyclic twinned microstructures differed. Analysis of dendrite arm spacing in cyclically twined structures with a beach ball morphology implies that the common 〈1 0 0〉 axis intersects with the region of the nucleation event. Models are presented for two pseudo/metastable hexagonal unit cells based upon {1 0 1} or {3 0 1} twins that introduce the cyclic twinning structure at the nucleation stage. Formation of these hexagonal unit cells may be facilitated by the presence of alloy elements. Subsequent epitaxial growth of the tetragonal unit cell on this nucleus can account for all three types of morphologies observed in microstructures of Sn-rich solder alloys.     

Controlling of electrical characteristics of Al/p-Si Schottky diode by tris(8-hydroxyquinolinato) aluminum organic film

We study how tris(8-hydroxyquinolinato) aluminum organic semiconductor layer at p-silicon/Al interface can affect electrical transport across this interface. Al/Alq3/p-Si device shows a good rectifying behavior with an ideality factor value of 1.95. The barrier height values obtained from I–V and Norde method were found to be 0.84 and 0.82 eV, respectively. This indicates that the barrier height obtained from Norde method is lower than that of barrier height value obtained from I–V due to the series resistance effect. The modification of the interfacial potential barrier for Al/p-Si diode was achieved using an interlayer of the Alq3 organic semiconductor and this is ascribed to the fact that the Alq3 interlayer increases the effective barrier height, because of the interface dipole induced by passivation of the organic layer. The frequency dispersion in capacitance and conductance can be interpreted in terms of the series resistance and interface state density values. The series resistance of the diode was changed from 9 kΩ to 1 kΩ with increasing frequency. The distribution profile of R_s–V gives a peak at low frequencies in the depletion region and disappears with increasing frequency.     

Interfacial morphology development and solute trapping behavior during rapid solidification of an Al–Li–Cu alloy

An aluminum–lithium–copper alloy was rapidly solidified via the electrospark deposition process. High-resolution scanning electron microscopy (HR-SEM), time-of-flight secondary-ion-mass-spectroscopy (TOF-SIMS) and atom probe tomography (APT) were employed to investigate the distribution of solute within the deposited materials. The TOF-SIMS data revealed evidence that solute trapping of lithium occurred during solidification, while SEM and APT revealed the presence of fine copper-rich cells within the microstructure (～30–60 nm in width). This morphology correlated directly with the microstructural morphology predicted by the Kurz–Giovanola–Trivedi (KGT) model for microstructural development during rapid solidification. The KGT model, which can be used to describe the planar–cellular transition within a microstructure, then predicted a solidification front velocity of ～1 m s^?1 being realized during electrospark deposition solidification. This SFV corroborated the chemical mapping data, and therefore supported the solute trapping hypothesis, as the continuous growth model for solute trapping as developed by Aziz and Kaplan (Acta Metallurgica 1988; 36:2335) predicts significant trapping of lithium at a SFV of 1 m s^?1. Finally APT revealed the presence of Al₃Li phase upon the copper-rich cell walls. It was then determined that the Al₃Li was not formed during solidification, as predicted by a time-dependent nucleation model for phase prediction during rapid solidification, and therefore is the result of a subsequent aging process.     

Solidification paths of multicomponent monotectic aluminum alloys

Solidification paths of three ternary monotectic alloy systems, Al–Bi–Zn, Al–Sn–Cu and Al–Bi–Cu, are studied using thermodynamic calculations, both for the pertinent phase diagrams and also for specific details concerning the solidification of selected alloy compositions. The coupled composition variation in two different liquids is quantitatively given. Various ternary monotectic four-phase reactions are encountered during solidification, as opposed to the simple binary monotectic, L′ → L′′ + solid. These intricacies are reflected in the solidification microstructures, as demonstrated for these three aluminum alloy systems, selected in view of their distinctive features. This examination of solidification paths and microstructure formation may be relevant for advanced solidification processing of multicomponent monotectic alloys.     

On the presence of work-hardened zones around fibers in a short-fiber-reinforced Al metal matrix composite

Dislocation densities are investigated in a short-fiber-reinforced Al–11 wt.% Zn–0.2 wt.% Mg metal matrix composite (MMC) with a special focus on regions near the fiber–matrix interfaces. Clear microstructural evidence is provided for the formation of work-hardened zones (WHZs) around fibers during creep using transmission electron microscopy (TEM). The dislocation densities in the WHZs are higher after creep than after squeeze casting, where the plastic strains associated with the thermal stresses that build up during solidification also result in an increased dislocation density close to fibers. The effect of heating and cooling on the dislocation substructure is also considered. The results are discussed in light of previous findings and provide microstructural evidence for the presence of WHZs as predicted by the Dlouhy model of MMC creep.     

Unidirectional solidification of Zn-rich Zn–Cu peritectic alloys—I. Microstructure selection

A solidification microstructure selection diagram has been determined for Zn-rich Zn–Cu peritectic alloys containing up to 7.37 wt% Cu over the growth velocity range 0.02–4.82 mm/s and at a temperature gradient of 15 K/mm by means of the Bridgman technique. Regular and plate-like cellular growth was observed in a range of composition near the peritectic point at growth velocities above 0.5 mm/s. The minimum growth velocity for the formation of plate-like cellular growth was about 2.64 mm/s in the compositional range from 2.17 to 4.94 wt% Cu. Two transitions, namely, arrayed to nonaligned growth of primary dendrites, and peritectic to non-peritectic reaction, were also characterized and analyzed. The transition velocity from arrayed to nonaligned growth of primary dendrites increased with increasing alloy concentration, which is consistent with the prediction of a modified Hunt model. The transition growth velocity from peritectic to non-peritectic reaction exhibited a maximum near the peritectic composition.