Search

Rapid solidification of Al–Cu droplets of eutectic composition was carried out using Impulse Atomization (a type of drop tube). Two distinct morphologies were observed: an irregular undulated eutectic assumed to form during recalescence, followed by a regular lamellar eutectic. The volume fraction of each morphology was measured and used to deduce the nucleation undercooling based on the hypercooling limit. A model of the eutectic solidification was developed assuming that the kinetics of the undulated and regular regions is the same and follows scaling laws established experimentally. The simulated solid fraction forming during recalescence matches the experimental undulated eutectic fraction. Furthermore, the heat balance confirms the adiabatic nature of the solidification during recalescence. Good agreement is found between the model and experimental measurements of lamellar spacing for the regular eutectic. However, the predicted spacing of the undulated eutectic is much lower than what is observed experimentally. This difference as well as the nature of this morphology is attributed to coarsening during the remaining of solidification of the very fine eutectic formed during recalescence.

High-speed camera (HSC) is an advanced technology, which can be applied to observe the solid–liquid interface directly during solidification. It has never been reported that the temperature evolutions of solidification that was monitored by using cameras and their images. In this study, a new application of HSC technology was developed for the temperature evolution investigation. With this technology, the effects of the measured position, the thermocouple size and thermocouple sensitivity (delay) can be demonstrated in temperature measurement. It is important that the higher accuracy of temperature profiles can be obtained by the HSC technology. This method can be expected to analyze the temperature profiles of small melt-pool and subsequent rapid solidification process in additive manufacturing. In the end, we discussed the relationship of the recalescence rate and the solidification interface.

A population dynamics-cellular automaton (PD-CA) model is developed to describe the microstructure formation in an inoculated Al alloys. The model involves the dynamics behaviors of the inoculated particles and the nucleation, initial spherical growth of nuclei as well as the subsequent dendritic growth. The model was validated by using the experimental results for the Al-Cu alloys inoculated by Al–Ti–C refiner first, and then used to simulate the detailed solidification process in an inoculated Al–Cu alloy. The results indicate that the TiC is not stable in Al melt. The heterogeneous nucleation process consists of two stages: a very short initial stage dominated by the cooling rate and the later stage dominated by the number of the active TiC. It stops at the very moment the recalescence occurs. The average grains size d of the aluminum alloys inoculated by the Al–Ti–C refiner can be calculated by $$d(\upmu{\text{m}}) = \frac{{a \cdot \exp (t/60 \cdot \ln (C_{ 0} /w))}}{{(\nu_{\text{cool}} \cdot t)^{1/3} w^{1/3} }} + \frac{b \cdot \ln t}{{Q \cdot v_{\text{cool}}^{1/2} }}$$ where Q is the growth restriction factor, C0 (%) is the initial solutes composition, w (%) is the additive amount of Al–Ti–C refiner. t (min) is the holding temperature time since the Al–Ti–C refiner is added into the melt. a and b are the constants.

The solidification of spheroidal graphite cast iron (SGI) plays an important role in production of sound quality castings. However, owing to the complex solidification behaviour of SGI, prepouring melt quality evaluation technique such as thermal analysis is used to predict quality of liquid metal by analysing the cooling curve pattern. In this study, an attempt has been made to investigate the role of speed of recalescence (SOR) in nucleation of graphite during solidification of SGI. Cooling curve analysis of liquid samples was carried out at different inoculation degrees. Recalescence and SOR data were obtained from thermal analysis, whereas nodule counts were obtained from microstructural examination of respective sample. It was observed that recalescence and nodule counts (nucleation sites) have a statistically significant inverse linear relationship; with the increase in recalescence, the nodule count was found to get reduced. However, the multiple values of nodule count at same recalescence degree were to be understood, which was clearly justified by SOR studies. After analysing the SOR vs nodule count relation, almost each SOR value corresponds to different nodule counts and the relation was statistically more significant. This infers that for an in-depth nodule count estimation, SOR analysis will be more helpful as compared to ∆Tr; hence, it is envisaged that speed of recalescence is true indicator of graphite nucleation as compared to recalescence in SGI.

The electromagnetic stirring (EMS) effects on peritectic solidification kinetics of undercooled liquid Fe-Ti alloys have been investigated by electrostatic levitation(ESL) and electromagnetic levitation (EML) methods assisted with in-situ diagnostic techniques. The high-sensitivity pyrometer and high-speed camera were employed to monitor the complete solidification process for levitated liquid Fe59Ti41 alloy in undercooling ΔT range of 0 K to 213 K. Theoretical calculations showed that there existed EMS inside electromagnetically levitated alloy melts, and the internal fluid flow dynamics depended on levitation height and melt undercooling. As ΔT rised, the primary dendrite growth velocity V increased according to a power function. Meanwhile, the peritectic recalescence degree ΔTpr and the peritectic recalescence rate Rpr were enhanced gradually, whereas the peritectic recalescence time tpr and the peritectic solidification time tps were shortened linearly. The comparison between ESL and EML experiments revealed that the EMS resulted in four respects of influences including (1) dendrite growth effect, (2) concentration field effect, (3) peritectic reaction effect and (4) microstructure evolution effect. In contrast with ESL, the V of Fe50Ti50 alloy measured by EML was slightly larger at small undercoolings, indicating the EMS affected dendrite growth processes. The solute concentration $$ C_{\text{L}}^{*} $$ around primary Fe2Ti dendrites for electrostatically levitated liquid Fe59Ti41 alloy deviated far away from original composition, while the EMS homogenized concentration field and the $$ C_{\text{L}}^{*} $$ variation was weak under EML condition. Both tpr and tps in the absence of EMS were longer that those in the presence of EMS, and it was demonstrated that the EMS accelerated peritectic reaction. Except for microstructure refinement, the EMS modulated the microstructure type and also changed the faceted-growth mode of intermetallic compound phases.

International audience; A melt maintained for hours in a press pour unit allowed the following changes over time from spheroidal graphite to compacted graphite iron by casting thermal cups at regular time intervals. This provided extensive experimental information for checking the possibility of simulating solidification of compacted graphite irons by means of a microstructure modelling approach. During solidification, compacted graphite develops very much as lamellar graphite but with much less branching. On this basis, a simulation of the thermal analysis records was developed which considers solidification proceeding in a pseudo binary Fe-C system. The simulated curves were compared with the experimental ones obtained from three representative alloys that cover the whole microstructure change during the holding of the melt. The most relevant result is that the parameter describing branching capability of graphite is the most important for reproducing the minimum eutectic temperature and the recalescence which are so characteristic of the solidification of compacted graphite cast irons.

In this paper, the 24 karat (99.99 wt.% pure) and 22 karat gold (Au-5.8wt.%Cu-2.5wt.%Ag) were grain refined with Au-6wt.%Ti master alloy and the mechanism of grain refinement was studied. The Au-6wt.%Ti master alloy containing Au2Ti and Au4Ti intermetallic particles was added into 24 karat and 22 karat gold at various addition levels of Ti (0.1, 0.2 and 0.3 wt.%) for grain refinement. The intermetallic particles undergo peritectic reaction with liquid Au in sequence to nucleate α-Au solid. It is observed that both the 24 karat gold and 22 karat gold showed efficient grain refinement at the lowest addition level of Ti (0.1 wt.%). However, 22 karat showed better grain-refining efficiency than 24 karat at all addition level of Ti studied. The mechanism of grain refinement of gold in this study matches with the “Peritectic theory” and “Solute paradigm” proposed for the grain refinement of aluminium. It is also observed that increasing the master alloy addition level to 0.2 and 0.3 wt.% Ti coarsening in the grain size in both 24 karat and 22 karat, but significant in the case of 24 karat gold due to recalescence effect. The grain-refined 24 karat and 22 karat gold at 0.1 wt.% Ti showed improvement in the hardness in comparison to the un-grain refined one.

One of the greatest environmental goals for the aluminum alloys industry is generating higher quality products by introducing cleaner input materials while maintaining low production costs. A typical dilemma for the master alloy producers is the cleanness level of the master alloy since insoluble inclusions could serve as inoculants during the solidification process. In this work, commercial Ti5B1 master alloy is used for grain refinement of Al7Si4Cu aluminum alloy and compared with the cleaned master alloy that contained a lower amount of residual refractory oxides and salts. Metallography analysis was used for grain size measurement while Computer Aided Cooling Curve Analysis was used for assessment of the undercooling and heat release values. In all instances, specimens treated with the cleaned master alloy showed smaller grains in the final structure and lower undercooling values. The difference in released heat between liquidus and recalescence temperatures was about 25% in specimens where added 0.66 wt% of cleaned master alloys compared to specimens where the commercial master alloys were added. Using cleaner Ti5B1 master alloy with a higher number of TiAl3 and TiB2 particles improves its grain refinement efficiency and transmits fewer impurities in produced parts. Producing cleaner master alloy would be beneficial from economic and environmental aspects by increasing its value and service time of produced parts besides simplifying the recycling process at the end of parts life-cycle. This is the peer reviewed version of the paper: Mitrašinović, A., Tomić, M., 2022. Functional and Environmental Advantage of Cleaning Ti5B1 Master Alloy, International Journal of Precision Engineering and Manufacturing-Green Technology, 9, 783–793 [https://doi.org/10.1007/s40684-021-00339-2] Peer-reviewed manuscript: [https://hdl.handle.net/21.15107/rcub_dais_11733]

This work deals with the mathematical modeling of the transient freezing process of a supercooled water droplet in a cold air stream. The aim is to develop a simple yet accurate lumped-differential model for the energy balance for a freely suspended water droplet undergoing solidification, that allows for cost effective computations of the temperatures and freezing front evolution along the whole process. The complete freezing process was described by four distinct stages, namely, supercooling, recalescence, solidification, and cooling. At each stage, the Coupled Integral Equations Approach (CIEA) is employed, which reduces the partial differential equation for the temperature distribution within the spherical droplet into coupled ordinary differential equations for dimensionless boundary temperatures and the moving interface position. The resulting lumped-differential model is expected to offer improved accuracy with respect to the classical lumped system analysis, since boundary conditions are accounted for in the averaging process through Hermite approximations for integrals. The results of the CIEA were verified using a recently advanced accurate hybrid numerical-analytical solution through the Generalized Integral Transform Technique (GITT), for the full partial differential formulation, and comparisons with numerical and experimental results from the literature. After verification and validation of the proposed model, a parametric analysis is implemented, for different conditions of airflow velocity and droplet radius, which lead to variations in the Biot numbers that allow to inspect for their influence on the accuracy of the improved lumped-differential formulation.

Quenching is an efficient manufacturing technique to improve the strength of steel after hot rolling. The benefit of this application is to enhance the mechanical properties of steel products while reducing strengthening alloying elements, e.g., C, Mn, V, Nb, and N. Quenching and self-tempering (QST) especially for H-beams is a unique material strengthening process that adopts intensive surface cooling and self-tempering. A methodological difficulty in estimating the quenching effect has been a long-standing concern in the QST application. The purpose of this study was therefore to specify quenching parameters, quantify quenching, analyze the effect, and verify the credibility of the results. Transient quenching was simulated in ANSYS to analyze heat transfer and phase transformations due to quenching. An individual concept, e.g., heat exchange, cumulative quenching infiltration, or recalescence phenomena, was merged and interpreted newly for the quenching simulation. Computational results based on theoretical approaches were well consistent with empirical studies.

Rapid solidification of a FeSi stoichiometric intermetallic compound was studied using the glass fluxing method. The recalescence process was in situ observed for the first time by the infrared high-speed high-resolution cameras. The dendrite envelope was found to be non-isothermal during the recalescence process. The growth velocity increased first, then decreased and finally held nearly constant. The average dendrite growth velocity for the recalescence process increased monotonically with undercooling and was described well by the dendrite growth model for a stoichiometric intermetallic compound. At low undercooling, the microstructure transition from coarse dendrites to refined grains was consistent with the dendrite fragmentation model and the chemical superheating model. At high undercooling, dendrite deformation triggers stress accumulation upon rapid solidification, thus providing the driving force for recrystallization. However, there were no evidences for annealing twins accompanied by recrystallization as well as random textures due to recrystallization nucleation. From the local misorientation map, the grain refinement mechanism was suggested to be stress-induced dendrite fragmentation. This study is helpful for not only understanding the intrinsic mechanisms of microstructure transitions in theory but also controlling microstructures and performance of intermetallic compounds in practical applications.

Slightly hypoeutectic ductile irons (4.10–4.2%CE) have a specific response to inoculation, which slightly decreased the temperature of the start of austenite formation (TAL), quickened up the start of eutectic freezing (nucleation) and increased the lowest and the highest (recalescence) eutectic temperatures. Temperature of the end of solidification increasing, with more negative level of the minimum value of the first derivative of cooling curve on the end of solidification, illustrates the efficiency of the applied inoculation to control the latter part of solidification. Increased graphitization factor, illustrating the graphite formation in the second part of eutectic reaction, sustains also the positive effect of inoculation. Inoculated irons show lower cooling rate for all of specific intervals than un-inoculated irons, with increasing tendency from beginning to the end of solidification, as specific values and also as difference between un- and inoculated irons. Inoculation potential, as ratio between the level of one specific parameter in inoculated and un-inoculated irons, has specific position, for each thermal analysis parameter, increasing from the beginning to the end of solidification. This treatment is useful not only to improve some metallurgical events at the beginning of solidification (carbide to graphite transition, graphite morphology control, eutectic cells count) but also it must be capitalized to control the last part of solidification, when the integrity of castings is decided.

The new generation of acid-lined medium-frequency coreless induction furnaces revolutionized the iron foundries, resulting higher overheating and lower S and Al content iron melt, characterized by higher eutectic undercooling solidification and consequently high chill [carbides] and unfavourable undercooled graphite morphologies sensitiveness. Having established that Al, Zr, Ti, Ca, Ba and La appear to have a key role in graphite nucleation in grey cast iron, thermal [cooling curves] investigations are undertaken to explore their effects on solidification parameters. The experimental heats are obtained in acid-lined coreless induction melting. Some important active elements are added to FeSi-based alloy, usually used as conventional inoculant in cast iron foundry industry, resulting complex inoculating systems [Al, Ca–La; Al, Ca–La, Zr; Al, Ca–La, Ba; Al, Ca–La, Zr, Ba; Al, Ca–La, Zr, Ti and Al, Ca–La, Zr, Ba, Ti]. In-mould [ceramic cup, cooling modulus of approximately 0.75 cm] inoculation is applied directly in the Quik-cup™ cooling curve analysis equipment, for the same alloy addition rate. The La effects are evaluated in low S and Al, high overheated electric grey iron solidification pattern with and without any other oxide-sulphides forming elements contribution. Study is focused on the representative temperatures during eutectic reaction and at the end of solidification, and it results in an undercooling referring to equilibrium eutectic temperature in the stable and metastable solidification system. As primary evaluation for general application, La, Ca, Al–FeSi alloy have a high efficiency in low S and Al and higher carbon equivalent grey cast irons, electrically melting, without any other active elements contribution. For specific applications, more complex La-bearing alloys are recommended, such as for higher dendritic austenite amount promotion [LaBaZrTi–FeSi alloy] or for lower eutectic recalescence [LaBaZr–FeSi alloy].

Simultaneous thermal and contraction/expansion analysis of hypoeutectic grey (lamellar graphite) and ductile (nodular graphite) cast irons solidification is recorded, using an equipment with a special designed ceramic cup, incorporating a thermocouple and contraction/expansion measuring device. Both cooling and contraction/expansion curves and their specific parameter values are displayed in a real time. The two compared cast irons occupy different and opposite positions during solidification process evolution: lower level of temperatures at the beginning and especially at the end of solidification, but higher level during the eutectic reaction for ductile cast irons. In the tested conditions, solidification of hypoeutectic cast irons with lamellar graphite is characterized by a greater undercooling during the eutectic reaction (lower ΔT1 and ΔT2, referring to the metastable eutectic temperature), with higher value for eutectic recalescence (ΔTr) and the maximum recalescence rate. Iron castings, including nodular graphite (ductile irons), show an end of solidification at higher undercooling (ΔT3), with less negative pick level of the first derivative of cooling curve. More graphitic initial expansion [edi(gr)], which favours the shrinkage formation, characterized nodular graphite iron castings. There is a good relationship between some parameters on the cooling curves and corresponding events on the contraction/expansion curves, such as solidification undercooling degrees, comparing to metastable eutectic temperature, and graphitic expansion [edi(gr)] for both tested cast irons: higher the level of ΔTr, and of ΔT1, ΔT2, ΔT3 (less negative) and GRF1 (graphitic factor), higher the [edi(gr)] level is.

Recalescence and segregation are two characteristic phenomena for the equiaxed dendritic solidification of alloys. The present work developed a two-dimensional dendritic model with cellular automaton (CA) method to investigate influence mechanisms of thermal conditions on recalescence and segregation behaviors of Fe-0.82wt pct C alloy. The released latent heat reduces the undercooling around the equiaxed dendrite, and thus eases its growth velocity. The predicted steady growth velocity agrees well with the analytical results as the melt undercooling is 9 K. Additionally, the present CA model can ensure the growth consistence of equiaxed dendrites in the undercooled melt. With improving the convective heat transfer coefficients applied around the domain boundaries, the temperature recalescence in the domain center becomes more significant, and the corresponding solid fractions enhance. It is because that the stronger cooling promotes the solidification, resulting in more latent heat released. Accordingly, the heat dissipation can be neutralized. Moreover, a deeper undercooling for the temperature recalescence is needed under a stronger cooling condition. With the increase of both the cooling rate and the convection coefficient, secondary arms of the equiaxed dendrite become more developed. At the lower cooling rate range, the segregation ratio in the domain enlarges with the improvement of the cooling rate. However, it gets weaker under the condition with super cooling intensity due to the expansion of the low concentration region. As the convection coefficient is enhanced, the solute segregation in the domain gets less pronounced.

The increasing demand for pre-pouring melt quality evaluation by thermal analysis has led to some important breakthroughs in understanding the solidification of spheroidal graphite cast iron (SGI). The two important parameters which characterize the solidification cooling curve of SGI are temperature of eutectic undercooling (TEU) temperature of eutectic recalescence (TER). In this study, the response of preconditioning treatment (0.1% Al,Zr,Ca–FeSi) on TEU and TER was investigated in several samples from different heats. It was found that preconditioning increases both TEU and TER. The average eutectic graphitization ability after preconditioning was found to increase from 60 to 75. The average undercooling (∆T) and recalescence (∆Tr) in preconditioned samples were 12 °C and 2 °C, whereas in samples without preconditioning average ∆T and ∆Tr are 19 °C and 7 °C, respectively. Preconditioning improves graphitization ability thereby increasing the graphite nodule count. The actual nodule count of SG iron samples obtained by image analysis was found to be in agreement with that of nodule counts predicted by thermal analysis.

Sessile droplet freezing on textured micropillar surfaces was observed visually and thermographically. Hydrophobic surfaces with three different topographies were selected, the contact angles of water on which were 141°, 102°, and 138°. Droplet surface temperature distribution measurements from lateral view enabled quantitative evaluation of the freezing stages. According to the experiment, it was found that the surface hydrophobicity decreases with the increasing pillar distance. Only a slight increase in hydrophobicity was found with the decreasing pillar diameter. The liquid water droplet experienced five successive stages to become fully frozen. It was also confirmed that the droplet’s opaque appearance is caused by the sudden ice shell formation at the ice incipience. Besides, the solidification frontier (SF) movement inside the droplet was analyzed. The freezing rate was higher at the SF edge. Moreover, the textured surfaces with higher hydrophobicity were prone to further postpone the droplet’s freezing onset. However, surface topography showed minor impact on the droplet surface recalescence stage duration. Additionally, the droplet internal solidification stage lasted longer on samples with higher hydrophobicity.

The Ni-16 at. pct Zr hypoperitectic alloy melt was substantially undercooled using an electromagnetic levitator and a drop tube. The undercooling-induced competitive growth between the primary Ni7Zr2 and peritectic Ni5Zr phases was revealed by observing of the recalescence process in situ and confirmed by analyzing the solidified microstructures, X-ray diffraction pattern as well as dendritic growth velocity. When the liquid undercooling is less than a critical value of 106 K, the primary Ni7Zr2 phase initially precipitates from the parent liquid, which is subsequently followed by the nucleation and growth of the peritectic Ni5Zr phase around it. The solidified microstructure consists of the Ni7Zr2 phase, the Ni5Zr phase, and inter-dendritic eutectics. The orientation relationship and interface characteristics of the Ni7Zr2 and Ni5Zr phases were investigated by electron backscatter diffraction (EBSD) and transmission electron microscopy (TEM). The EBSD results clearly demonstrate that the Ni7Zr2 and Ni5Zr phases have the parallel relationship of {111}Ni7Zr2 // {111}Ni5Zr. TEM analysis reveals that a large-scale flat interface exists between the Ni7Zr2 and Ni5Zr phases, indicating good lattice matching of the two phases along the phase boundary. Once the critical undercooling is exceeded, the peritectic Ni5Zr phase preferentially nucleates and grows from the undercooled melt by completely suppressing the formation of the primary Ni7Zr2 phase. The EBSD analysis shows that the peritectic Ni5Zr phase is highly orientated and its growth mode is almost parallel to the 〈110〉 directions. When containerlessly solidified during free fall, typical peritectic microstructures form in large droplets, while only peritectic phase appears in the small droplets. This result further confirms the strong competition between the primary and peritectic phases in the Ni–Zr hypoperitectic alloy induced by large undercoolings.

The main objective of the present work is to introduce a specific experimental instrument and technique for simultaneously evaluating cooling curves and expansion/contraction of cast metals during solidification, adapted for commercial foundry use. The recorded data are processed using specialized software, which conveniently displays both cooling and contraction/expansion curves and their specific parameter values. Experiments compared hypoeutectic gray (GI) and ductile (DI) irons with white (WI) irons. Three important moments were found on the expansion/contraction-cooling curves: the start of eutectic freezing, the point of maximum expansion and the end of solidification. All of the tested irons have similar values for initial expansion up to the start of eutectic freezing (0.44%) due to the ferrostatic pressure, silica sand mold expansion, mold movement, etc. The maximum expansion, reached between the temperature of eutectic recalescence and the end of solidification, depends on the carbides/graphite ratio and graphite morphology: WI-0.465%, GI-0.552%, DI-1.032%, as averages. Graphitic expansion, absent for WI, increased to 0.109% (GI) and up to 0.596% (DI). For both GI and DI, the expansion at the end of solidification is only 6% lower compared to the maximum level, while for WI it decreased more than 50%. Specifically, the acceleration rate of the graphitic expansion up to the maximum level (Kgr1) is different compared with its deceleration to the end of solidification (Kgr2), and also are different for irons being tested, GI versus DI. Nodular graphite led to the (Kgr1) factor being 2.5 times higher, compared to GI, whereas only a slight difference was observed between GI and DI for the Kgr2 factor. Higher graphitic expansion in DI led to higher shrinkage sensitivity, compared to GI, as measured in furan resin mold test castings.

We present the results of development of the technology of thermal hardening and thermal straightening of shaped rolled St3 steel (structural angles) based on the mathematical and physical simulation of the process of heat transfer with accelerated water cooling of the metal. For this purpose, we used the DEFORM 3D + HEAT TREATMENT software environment of the finite-element method and carried out laboratory experiments in three furnaces for different temperatures and holding times varying within broad ranges in order to simulate the modes of accelerated water cooling of shaped rolled products. The characteristics of mathematical simulation of the technology of thermal treatment of equal-section rolled structural angles 63 × 63 × 4 mm in size made of St3sp steel are presented with an aim to evaluate the structure and mechanical properties of the rolled metal. It is confirmed that the diffusion phase transformation is accompanied by the heat release (the so-called phenomenon of recalescence). We also present the temperature distributions over the sections and areas of the rolled steel showing that the less massive areas of the profile (edges of the structural angles) are cooled much more intensely than the massive sections (the top of the profile angle). However, the areas accumulating more heat transfer their heat to less massive and hardened areas. As a result, we observe the self-tempering of the martensitic areas accompanied by the formation of tempering pearlitic structure quite efficient for the purposes of operation. The laboratory investigations were carried out on three shaped profiles (equal-leg structural angles 40 × 40 × 4 and 75 × 75 × 8 mm in sizes and channel No. 5). The efficiency of thermal treatment was proved for shaped profiles made of unalloyed ordinary St3sp steel hardened within the range of tempering temperatures 200–650°C, with attainment of the hardening coefficient within the range 1.4–2.6 (according to the temporal rupture resistance). This guarantees the possibility of the production of angled rolled products of elevated strength classes within the range 440–1130 MPa instead of the original hot-rolled strength class of 320 MPa (by the yield strength). The obtained results of modeling of the technology of thermal treatment of structural angles made of St3sp steel can be extended (after the required experiments) to other types of steel (e.g., 09G2, 09G2S, 10KhSND, etc.) with the exception of expensive microalloying with vanadium, niobium, titanium, and other elements.

The effect of Zr as a grain refiner on the solidification behavior, micro- and macrostructure of a new Al–Zn–Mg–Cu aluminum super-high strength alloy containing high Zn content was studied. The addition of 2 mass% Zr reduced the grain size from 1500 to 190 μm. Moreover, the dendritic structure of the alloy altered from a coarse, elongated and non-uniform morphology to a rosette-like shape and more uniform one. The parameters of liquidus region of cooling curve obtained from thermal analysis were in a good correlation with grain size results. The maximum of first derivative in the liquidus region was introduced beside recalescence undercooling which could predict the grain refinement level even after disappearing of recalescence in the cooling curve. Furthermore, the addition of 1 mass% Zr enhanced fraction of solid in dendrite coherency point from 21 to 31% and lessened the amounts of porosity from 2.3 to 1.4%.

The effect of cooling rate on the microstructure and solidification parameters of Mg–3Al–3Nd alloy was investigated by thermal analysis. The different cooling rates were chosen in the present work, which were attained by a stepped sand mold, ranging from 0.3 to 3.5 °C s−1. The solidification parameters were determined from the cooling curve as well as its first and second derivative curves. The results show that the microstructure and solidification parameters of Mg–3Al–3Nd alloy are influenced by changing the cooling rate. The nucleation temperature, undercooling temperature, the range of solidification temperature all increase with increasing the cooling rate. The precipitated temperature of Al2Nd and Al11Nd3 phases, recalescence undercooling temperature and total solidification time all decrease with increasing the cooling rate. The grain size of Mg–3Al–3Nd alloy is refined, and the intermetallic phases are distributed more uniformly with increasing the cooling rate. Furthermore, the volume fraction of intermetallic phases and the number density of Al2Nd particle also increase with increasing the cooling rate.

By employing the glass fluxing technique in combination with cyclical superheating, the microstructural evolution of the undercooled Fe82B17Si1 alloy in the obtained undercooling range was studied. With increase in undercooling, a transition of cooling curves was detected from one recalescence to two recalescences, followed by one recalescence. The two types of cooling curves were fitted by the break equation and the Johnson–Mehl–Avrami–Kolmogorov model. Based on the cooling curves at different undercoolings, the recalescence rate was calculated by the multi-logistic growth model and the Boettinger–Coriel–Trivedi model. Both the recalescence features and the interface growth kinetics of the eutectic Fe82B17Si1 alloy were explored. The fitting results that were obtained using TEM (SAED), SEM and XRD were consistent with the changing rule of microstructures. Finally, the relationship between the microstructure and hardness was also investigated.

Grain refinement phenomena during the microstructural evolution upon nonequilibrium solidification of deeply undercooled Ni-20 at. pct Cu melts were systematically investigated. The dendrite growth in the bulk undercooled melts was captured by a high-speed camera. The first kind of grain refinement occurring in the low undercooling regimes was explained by a current grain refinement model. Besides, for the dendrite melting mechanism, the stress originating from the solidification contraction and thermal strain in the FMZ during rapid solidification could be a main mechanism causing the second kind of grain refinement above the critical undercooling. This internal stress led to the distortion and breakup of the primary dendrites and was semiquantitatively described by a corrected stress accumulation model. It was found that the stress-induced recrystallization could make the primary microstructures refine substantially after recalescence. A new method, i.e., rapidly quenching the deeply undercooled alloy melts before recalescence, was developed in the present work to produce crystalline alloys, which were still in the cold-worked state and, thus, had the driven force for recrystallization.

The aim of this research was evaluation of the solidification parameters of A356 alloy, e.g., dendrite arm spacing (DAS), correlation between cooling rate (CR) and DAS, hot tearing, and microstructural analysis at different coating thicknesses and strip sizes during the lost foam casting process (LFC). To achieve this goal, the DAS was measured at six coating thicknesses and six different strip sizes. In addition, thermal characteristics, such as the CR, temperatures of start and finish points of solidification, recalescence undercooling, and hot tearing susceptibility (HCSC), at five coating thicknesses were recognized from the cooling curves and their first derivative and the solid fraction curves, which have been plotted through the thermal analysis technique. The pouring temperature and strip size were fixed at 1063 K (790 °C) and 12 mm, respectively. Besides, to derive a numerical equation to predict the CR by measuring the DAS in this alloy, a microstructural evaluation was carried out on samples cast through 12-mm strip size. The results showed that both coating thickness and strip size had similar influences on the DAS, in which, by retaining one parameter at a constant value and simultaneous enhancement in the other parameter, the DAS increased significantly. Furthermore, at thinner coating layer, the higher amount of the CR was observed, which caused reduction in the temperatures of both the start and finish points of solidification. Also, increasing the CR caused a nonlinear increase in both the recalescence undercooling and the HCSC.

The solidification behavior of AZ Magnesium alloys in various cooling conditions was investigated using a computer-aided cooling curve thermal analysis method. In each case, the cooling curve and its first and second derivative curves have been plotted using accurate thermal analysis equipment and solidification characteristics were recognized from these curves. The cooling rates used in the present study range from 0.22 to 8.13 °C s−1. The results of thermal analysis show that the solidification parameters of AZ alloys such as nucleation temperature (T N,α), nucleation undercooling (∆T N,α), recalescence undercooling (∆T R,α), range of solidification temperature (∆T S) and total solidification time (t f) are influenced by variation of cooling rate. Also, the effect of Al content on these parameters was studied. Microstructural evaluation was carried out to determine the correlation between the cooling rate and secondary dendrite arm spacing.

The cooling curves for Fe-21 at. pct B hypereutectic alloy were measured, which are correlated with the solidification behaviors. At small undercoolings, the cooling curve has two recalescences, corresponding to L → Fe2B and L → Fe2B + Fe, respectively. At moderate undercoolings, also two recalescences can be observed from the cooling curve, corresponding to L → Fe3B and L + Fe3B → Fe2B + Fe. At large undercoolings, the cooling curve has only one recalescence, corresponding to L → Fe3B + Fe. From TEM analysis, the matrix phase is α-Fe in cases of solidification at small and moderate undercoolings, but is Fe3B in case of large undercooling. HRTEM analysis shows that α-Fe stores deformation energy by irregular atom region, but intermetallic phase (Fe2B or Fe3B) stores deformation energy by increasing grain boundaries and stacking faults, which can explain why α-Fe has good deformation ability and small hardness, but the Fe2B or Fe3B phase on the contrary.

The heterogeneous undercooling degree of interstitial-free steel (IF steel) on various substrates was measured through observing the recalescence phenomenon by sessile drop method at the oxygen partial pressure of 10−23 atm, where there was no obvious interaction between the steel and the substrate. The wetting behavior of liquid IF steel against different substrates was also studied during the sessile drop test. The results suggested that the undercooling degree increased with the increase of cooling rate, and the effective contact surface area between the liquid steel and the solid substrate could not fully explain the undercooling behavior of the designed cases as there is no interaction layer. Furthermore, a method to calculate the critical contact angle for IF steel against its substrate was developed in this study, which could be adapted to explain the undercooling variance of IF steel with the change of its substrate. The undercooling degree increased with the increase of the critical contact angle, and the critical contact angle was also improved with the increase of lattice disregistry.

The effects of different Sn contents on the microstructure and tensile properties of an in situ prepared Al–15% Mg2Si composite were investigated. Adding 5 wt% Sn not only reduced the average size of Mg2Si primary particles from ∼55 µm to ∼10 µm, but also increased the ultimate tensile strength and elongation values. Mg2Sn intermetallic was identified in Al–15% Mg2Si–5% Sn composite and the lowest disregistry was also found for ${(001)_{{\rm{M}}{{\rm{g}}_2}{\rm{Sn}}}}\left\| {{{(001)}_{{\rm{M}}{{\rm{g}}_2}{\rm{Si}}}}} \right.$ , indicating that Mg2Sn can be a potent heterogeneous nucleation site for Mg2Si particles. Computer-aided cooling curve analysis revealed that Sn addition changes the solidification performance of the material by increasing both solidification time and recalescence undercooling and also reducing the growth temperature to some extent. The solidification behavior of the composite was also explained in terms of the presence of oxide bifilms in the liquid Al.

The Ni–38 wt% Si alloy whose eutectic products are two stoichiometric intermetallic compounds (i.e., NiSi and NiSi2) was undercooled by the melt fluxing technique. After in situ observations of the recalescence processes using a high-speed camera and by electron back-scattering diffraction analysis of the solidification microstructures, the crystal growth velocities, phase selection, and microstructure evolutions were studied. Due to a growth-controlled mechanism, the primary phase changes from the NiSi to the NiSi2 phase at a critical undercooling ΔT ≈ 48 K. Even in the absence of the driving force of chemical superheating, the transition from regular eutectics to anomalous eutectics happens. The reason is that the single-phase dendrite of NiSi2 phase solidifies firstly and then the NiSi phase grows epitaxially to form an uncoupled eutectic-dendrite at high undercooling. The present work provides further experimental evidences for the dual origins of anomalous eutectics (e.g., uncoupled eutectic-dendrite growth during the recalescence stage and coupled lamellar eutectic growth at low undercooling during the post-recalescence stage) and is helpful for understanding of non-equilibrium phenomena in undercooled melts.

Computer-aided cooling curve analysis is a reliable method to characterize the solidification behavior of an alloy. In this study, the effect of Al–5Ti–1B grain refiner on the solidification path, microstructure and macrostructure of a new Al–Zn–Mg–Cu super high-strength aluminum alloy containing high amounts of zinc was investigated using thermal analysis technique. The grain size measurement showed that Al–5Ti–1B reduces the grain size from 1402 to 405 μm. Solidification parameters in the liquidus region were in a good accordance with microstructural results. The addition of 1 mass% of Al–5Ti–1B grain refiner decreased ΔT N from 9.1 to 7.7 °C. It also diminished recalescence undercooling from 1.42 to 0.32 °C. The grain refinement also altered dendritic structure of the alloy from a coarse, elongated and non-uniform to a rosette and more uniform shape. Moreover, the grain refiner resulted in a more uniform distribution of eutectic structure between dendrite arms. Furthermore, the grain refinement enhanced fraction of solid at dendrite coherency point from 21 % for unrefined alloy to 25 % for the alloy containing 1 mass% Al–5Ti–1B. In the same trend, the addition of 1 mass% Al–5Ti–1B reduced the amounts of porosity from 2.3 to 1.8 %.

During spray atomization process, the heat transfer and solidification of droplets play very important roles for the deposition quality. Due to the difficulties of experimental approach, a numerical model is developed, which integrates liquid undercooling, nucleation recalescence and post-recalescence growth to present the full solidification process of Fe-6.5%Si (mass fraction) droplet. The droplet velocity, temperature, cooling rate as well as solid fraction profiles are simulated for droplets with different sizes to demonstrate the critical role of the size effect during the solidification process of droplets. The relationship between the simulated cooling rate and the experimentally obtained secondary dendrite arm spacing is in excellent agreement with the well-established formula. The pre-constant and exponent values lie in the range of various rapid solidified Fe-based alloys reported, which indicates the validity of the numerical model.

Ultrasonic treatment (UST) of molten metals produces a fine grain microstructure. Several mechanisms of grain refining by UST have been suggested; however, experimental verification has not yet established the actual mechanism. In this study, UST was applied to Al-4 wt pct Si molten alloy (1) above the liquidus temperature, (2) during undercooling before recalescence, (3) during recalescence, and (4) after recalescence. After UST treatment, the average grain sizes of the solidified samples were measured, and the effects of UST were evaluated. In the case of (1), the temperatures of the crucible, ultrasonic horn, and the atmosphere of the molten alloy were also confirmed above the liquidus temperature, and a finer grain structure was obtained. This result reveals that UST promotes the nonequilibrium nucleation mechanism. UST during (2) and (3) also caused finer grain microstructures, whereas (4) did not. UST during undercooling hastened nucleation. However, the previously reported mechanism of the breaking dendrite was not activated by UST after recalescence.

Although a significant amount of work has already been devoted to the prediction of as-cast grain size in inoculated aluminum alloys, most models treated the grains either spherical or directly dendritic. The influence of morphology transition from spherical to dendritic and subsequent dendritic grain growth has not been investigated and discussed in detail, especially in non-isothermal melt solidification condition (without recalescence). In this work, we propose a new grain size prediction model in which the morphology transition and dendritic growth are considered. A comparison has been made between forced spherical growth model and the one taking into account the morphology transition and dendritic growth. It is demonstrated that spherical growth is an acceptable approximation for well grain refined aluminum alloys in both isothermal and non-isothermal melt solidification. The difference in the predicted grain sizes based on the two different grain growth models is small. However, for not well grain refined aluminum alloys, e.g. low inoculation addition level or/and at low cooling rate condition, the dendritic grain growth model has a more significant influence on the predicted grain size and it is important to take it into consideration.

The recalescence events of phase-separated Co-rich phases in undercooled molten CuCo droplets electromagnetically levitated under various static magnetic fields were observed directly using a high-speed camera, and also the surface velocities on the levitated droplets were measured by tracing the trajectories of the phase-separated Co-rich phases as tracer particles. In addition, numerical simulations of melt convection in a spherical electromagnetically levitated CuCo droplet exposed to a static magnetic field were performed assuming laminar flow. We observed the emergence of many intermittent bright spots due to recalescence on the entire surface of the levitated droplet, and the frequency of the bright spots decreased markedly as the static magnetic field increased, with no bright spots observed at fields larger than 1.5 T. Also, the Reynolds numbers were evaluated from the measured and calculated velocities in the droplet for various static magnetic fields and compared with the critical Reynolds number of approximately 600, at which the laminar–turbulent transition of a magnetohydrodynamic (MHD) flow in an electromagnetically levitated droplet occurs, as proposed by Hyers et al. The above results clearly revealed that the marked change in the phase separation structures in undercooled molten CuCo droplets at approximately 1.5 T is due to a convective transition from turbulent flow to laminar flow in the levitated droplets, as speculated in our previous work.

Light alloys like aluminum and its alloys have excellent physical and mechanical properties for a number of applications. The use of aluminum alloys can significantly decrease the mass of automobiles without decreasing structural strength. Therefore, the reason of this work was to determine the optimal cooling rate values, to achieve good mechanical properties for protection of this aluminum cast alloy from losing their work stability, and to make it more resistant to action in hard working conditions. The carried out investigations have allow to found that changes in the cooling rate do not cause changes in the phase composition, revealing the Al2Cu and Al5FeSi phase especially, but only changes the morphology of α + β eutectic as well as the particle size and secondary dendrite arm spacing. As a result, the number of fine crystals in per unit volume increases, leading to a fine grain structure, which influences the recalescence temperature. The purpose of this research work is to investigate the thermo-derivative interdependencies occurred in analyzed aluminum cast alloys using Universal Metallurgical Simulator and Analyzer. For the investigation, the cast AlSi9Cu aluminum alloy was used. As a result of this research, the cooling rate influence on the structure and mechanical properties changes was investigated. The cooling rate was set in a variable range of 0.16–1.25 °C s−1, where the cooling rate of 0.16 °C s−1 corresponds to freely cooling, without any forced air flow.

Drops of liquid are placed on substrates having temperatures below the melting point of the liquids (supercooling). All well-known stages of cooling and freezing (supercooling–recalescence–main solidification–ice cooling) of water drops on both hydrophilic and hydrophobic surfaces were observed using a k-type thermocouple and an IR camera. Duration of the main solidification and the ice cooling stages were found to be dependent on the drop size. However, no conclusive relationship can be made regarding the duration of the supercooling stage with the drop size and the surface hydrophobicity. Using an IR camera, we observed that the solidification initiation of the main drop, which corresponds to the supercooling stage duration, is triggered by the erratic wave-like recalescence front of the surrounding condensed microdrops. The nucleation always initiates from the trijunction and propagates into the drop volume with different structural morphologies determined by temperatures and surface hydrophobicity of the substrate. The recalescence front speed of the main drop varies from 50 to 150 mm/s with supercooling temperatures.

The influence of grain refinement—with the addition of 6 mass% Al–5Ti–B and 10 mass% Al–3Ti–B master alloys—on the temperature-based parameters of a commercial ADC12 die-casting alloy, modified with 0.04 mass% Sr, was investigated using computer-aided cooling curve thermal analysis. Results show that the addition of grain refiners caused the reaction area of the aluminium dendrite to shift up. With the addition of refiner reagents, the nucleation temperature and growth temperature for the evolution of the primary Al dendrite increased, while the recalescence temperature decreased. The dendrite coherency temperature (TDCP), time (tDCP), temperature (TN–TDCP), and time (tN–tDCP) interval for dendrite coherency and solid fraction at coherency point increased with the addition of Al–3Ti–B and Al–5Ti–B grain refiners. Nevertheless, the growth temperature of the eutectic aluminium–silicon phase remained at the modified range (563.5–565 °C) even after the addition of grain refiners. A grain-refined and modified structure was obtained after the addition of refiners, but no obvious mutual effect was found in terms of the coarsening of grain size or silicon demodification in the range of concentrations used in this study.

High magnetic field is a powerful tool to tune the microstructure and improve the properties of materials. In this report, the nucleation behavior of undercooled Co76Sn24 near eutectic alloy under strong homogeneous and gradient magnetic fields have been investigated using glass slag fluxing method in a 12 T superconducting magnet. The mean undercooling of the undercooled melt is not altered by homogeneous magnetic field but depressed by gradient magnetic field. The highest temperature during recalescence is strongly altered by magnetic field, where an enhancement effect is observed under gradient magnetic field and an opposite effect in homogeneous magnetic field. The reason is interpreted by discussion about the magnetic field on the thermodynamics of nucleation and also the purifying effect of the glass slag, the magnetic properties and the magnetic force exerted on the undercooled melt.

We employ the high-speed synchrotron hard X-ray imaging and diffraction techniques to monitor the laser powder bed fusion (LPBF) process of Ti-6Al-4V in situ and in real time. We demonstrate that many scientifically and technologically significant phenomena in LPBF, including melt pool dynamics, powder ejection, rapid solidification, and phase transformation, can be probed with unprecedented spatial and temporal resolutions. In particular, the keyhole pore formation is experimentally revealed with high spatial and temporal resolutions. The solidification rate is quantitatively measured, and the slowly decrease in solidification rate during the relatively steady state could be a manifestation of the recalescence phenomenon. The high-speed diffraction enables a reasonable estimation of the cooling rate and phase transformation rate, and the diffusionless transformation from β to α ’ phase is evident. The data present here will facilitate the understanding of dynamics and kinetics in metal LPBF process, and the experiment platform established will undoubtedly become a new paradigm for future research and development of metal additive manufacturing.

A deterministic multigrain and multiphase model of equiaxed solidification of binary alloys is proposed, implemented, and analyzed. An important feature of the present model is the creation of classes of dendritic and globulitic grains according to their instantaneous sizes during solidification. Globulitic and dendritic grain growth, coarsening of secondary dendrite arms, distribution of nucleation undercoolings, and equiaxed eutectic growth are consistently included in the model equations. Important model assumptions are uniform temperature, negligible liquid convection, and negligible grain movement. Calculated cooling curves, solid fraction evolution, average grain sizes, and eutectic fractions agree well with predictions of previous models for dendritic and globulitic equiaxed grains. Predicted grain sizes decrease with an increase in cooling rate for an Al-2.12 pct Cu alloy and with an increase in Si concentration up to 3 pct for Al-Si alloys, agreeing quantitatively with experimental results. Simulations for an Al-7 pct Si alloy predict that an increase in grain size correlates with an increase in the magnitude of the recalescence observed in cooling curves. These calculations agree well with experimental results when the transition from a globulitic to a dendritic morphology occurs in the model before the minimum temperature of recalescence is reached.

The effects of combined additions of Ca and P on the eutectic Si in a series of high purity Al-5 wt pct Si alloys have been investigated with the entrained droplet technique and complementary sets of conventional castings. Differential scanning calorimetry (DSC) and thermal analysis were used to investigate the eutectic droplet undercooling and the recalescence undercooling, respectively. Optical microscopy, SEM, EPMA, and TEM were employed to characterize the resultant microstructures. It was found that 250 ppm Ca addition to Al-5Si wt pct alloys with higher P contents leads to a significant increase of the eutectic droplet undercooling. For low or moderate cooling rates, the TEM results underline that Ca additions do not promote Si twinning. Thus, a higher twin density cannot be expected in Ca containing Al-Si alloys after, e.g., sand casting. Consequently, a refinement of the eutectic Si from coarse flake-like to fine plate-like structure, rather than a modification of the eutectic Si to a fibrous morphology, was achieved. This strongly indicates that the main purpose of Ca additions is to counteract the coarsening effect of the eutectic Si imposed by higher P concentrations. Significant multiple Si twinning was observed in melt-spun condition; however, this can be attributed to the higher cooling rate. After DSC heating (slow cooling), most of Si twins disappeared. Thus, the well-accepted impurity-induced twinning mechanism may be not valid in the case of Ca addition. The possible refinement mechanisms were discussed in terms of nucleation and growth of eutectic Si. We propose that the pre-eutectic Al2Si2Ca phase and preferential formation of Ca3P2 deactivate impurity particles, most likely AlP, poisoning the nucleation sites for eutectic Si.

A model has been proposed for describing the solidification of undercooled metallic melts, on the basis that the solidification is considered as the sum of a liquid/solid cooling process without latent heat release and a latent heat releasing process without cooling. Application of the model to undercooled solidification of nickel, cobalt, zirconium, and Al–11.3Si–2Cu–0.4Fe alloy samples shows good fits to the temperature curves, and the fitting results of solid fraction curves are consistent with that from the available methods.

A two-dimensional axisymmetric phase-field model of thermo-solutal solidification in freezing-point cells used for calibrating standard platinum resistance thermometers for realization and dissemination of the International Temperature Scale of 1990 is presented. The cell is essentially a graphite crucible containing an ingot of very pure metal (of order 99.9999 %). A graphite tube is inserted along the axis of the ingot to enable immersion of the thermometer in the metal. In this study, the metal is tin (freezing temperature of \(231.928\,^{\circ }\hbox {C}\)). During the freezing of these cells, a steady, reproducible temperature is realized, with a defined temperature that can be used to calibrate thermometers with uncertainties \({

The aim of this work was to investigate the effect of different cooling rates on the microstructure and solidification parameters of 2024 aluminum alloy. Solidification characteristics are recognized from the cooling curve and its first and second derivative curves which have been plotted using thermal analysis technique. In this study, a mold having high cooling rate was designed and used to simulate the direct-chill casting process. The results of thermal analysis show that the characteristic parameters of Al2024 alloy are influenced by cooling rate. The cooling rates used in the present study range from 0.4 to 17.5 °C s−1. Increasing the cooling rate affects the undercooling parameters both in liquidus and eutectic solidification regions. Investigations showed that solidification parameters such as nucleation temperature, recalescence undercooling temperature, and range of solidification temperature are influenced by variation of cooling rates. Microstructural evaluation was carried out to present the correlation between the cooling rate and dendrite arm spacing.

Undercooling and recalescence were studied using the differential scanning calorimetry (DSC) method on real electronic systems. Two solder pastes, Sn62.5Pb36.5Ag1 and Sn96.5Ag3Cu0.5, were used for preparation of electronic joints. Various combinations of these solders and soldering pads with different surface finishes such as Cu, Cu-Ni-Au, Cu-Sn, and Cu-Sn99Cu1 were used. During melting of both pastes, the Sn and Sn99Cu1 surface finishes immediately dissolved in the solder and the Cu surface coating was exposed to the melt. Therefore, practically the same undercooling was found for the Cu, Cu-Sn, and Cu-Sn99Cu1 coatings. The lowest undercooling was found for the Cu-Ni-Au surface finish for both solder pastes. If two separated electronic joints were made on the sample, two separate peaks were found in the DSC signal during solidification. In the sample with only one joint, only one exothermic peak was found. These findings were observed for all paste/surface finish combinations. These data were analyzed, showing that this effect is a consequence of undercooling and recalescence: Latent heat released during solidification of the joint increases the surrounding temperature and influences all the processes taking place.

This paper aims at briefly reviewing phase field models applied to the simulation of heterogeneous nucleation and subsequent growth, with special emphasis on grain refinement by inoculation. The spherical cap and free growth model (e.g. A.L. Greer, et al., Acta Mater. 48, 2823 (2000)) has proven its applicability for different metallic systems, e.g. Al or Mg based alloys, by computing the grain refinement effect achieved by inoculation of the melt with inert seeding particles. However, recent experiments with peritectic Ti-Al-B alloys revealed that the grain refinement by TiB2 is less effective than predicted by the model. Phase field simulations can be applied to validate the approximations of the spherical cap and free growth model, e.g. by computing explicitly the latent heat release associated with different nucleation and growth scenarios. Here, simulation results for point-shaped nucleation, as well as for partially and completely wetted plate-like seed particles will be discussed with respect to recalescence and impact on grain refinement. It will be shown that particularly for large seeding particles (up to 30 μm), the free growth morphology clearly deviates from the assumed spherical cap and the initial growth – until the free growth barrier is reached – significantly contributes to the latent heat release and determines the recalescence temperature.

Obtained results of micro and nano studies reveal that bismuth refines the silicon in which the flake silicon changed to lamellar structure with reduction in twin spacing from 160 to 75 nm. Bismuth segregates towards the inter-dendritic regions and decreases the Al-Si contact angle resulting in suppression of the silicon growth causing refinement of the eutectic structure. Increased recalescence temperature and time confirmed that the refinement effect is attributed to the growth stage.

The role of phosphorus (P) in the heterogeneous nucleation of eutectic silicon (Si) and the evolution of eutectic grains in hypoeutectic aluminum-silicon alloys were investigated. Systematic additions of P in the range of 0.5 to 20 ppm to Al-7 wt pct Si alloys of different purities have shown that the morphology of the eutectic Si changes from a fine plate- to a coarse flake-like structure. The growth of eutectic grains was investigated by interrupting the eutectic reaction by quenching experiments. Moreover, the macroscopic growth mode of the eutectic grains was characterized by electron backscatter diffraction. An increase in P concentration from 2 to 3 ppm resulted in a transition of the macroscopic growth mode of the Al-Si eutectic in high purity alloys from growth with a planar front with a strong dependence of the thermal gradient, to nucleation in the vicinity of the primary Al dendrites and subsequent growth of distinct eutectic grains. It is suggested that AlP particles are the key impurities acting as potential nucleation sites for eutectic Si. This is further substantiated as with increasing P concentration nucleation and growth of the Al-Si occurred at higher temperatures close the equilibrium Al-Si eutectic solidification temperature at 850 K (577 °C). In addition, the recalescence undercooling ΔT R,eu was reduced from 4.5 K (0.5 ppm P) to 1.5 K (20 ppm P) in high purity alloys. This was accompanied by a drastic increase of the nucleation rate of the eutectic grains.

Solidification of undercooled Ni–3.3 wt% B alloy melt was investigated by glass fluxing. If ΔTe < 140 ± 10 K, two recalescences appear, indicating that stable eutectic reaction occurs; if ΔTe ≥ 140 ± 10 K, three recalescences can be observed, indicating that metastable eutectic reaction occurs. Analysis indicates that the phase fractions of the as-solidified structure can be predicted by the recalescence delay times in the cooling curves. High-speed video images show that the solidification interface of primary solidification changes from single dendritic shape to spherical shape with increasing ΔTp; the interface of eutectic solidification changes from many small “dendrites” to a single large one with increasing ΔTe; the interface of residual liquid solidification changes from many small rings to a single large one with increasing ΔTr. The growth velocity of eutectic solidification suggests a coupled growth at small and moderate undercoolings and decoupled growth at large undercooling, whereas that of residual liquid solidification cannot be interpreted by the available models.