Efficient metallurgical and processing development requires ability to predict the solidification microstructure of products. Models were developed to calculate possible growth morphologies and phases in order to select the most efficient one for a given solidification condition. It is then possible to predict solidification microstructure selection maps (SMSM) indicating the microstucture as a function of the composition and the growth rate. A summary of the methodology of solidification microstructure prediction is given. Results obtained in a large velocity range for Al-Fe and Al-Cu systems are reviewed with experimental comparison and emphasis on the role played by metastable phase equilibria in the modelling. Finally, an outline of the actual work extending this approach to multicomponent alloys is given. Provisional calculated solidification microstructure maps for the Al-Cu-Si system are presented.

View full abstract

A numerical analysis is presented of the solidification of compositions close to Ni₃Al. A two-sublattice thermodynamic model for the γ and γ' phases is combined with disorder trapping theory and dendrite growth modelling based on the marginal stability hypothesis. It is thereby possible to predict the composition and degree of order in the growing solid as a function of liquid composition and undercooling. Preliminary results are presented on solidification of the liquid of composition Ni-2at%Al.

View full abstract

Undercooling of liquids is a common occurrence in solidification. The level of undercooling influences both the microstructural development by controlling phase selection during nucleation and the morphological evolution during the growth phase of solidification. The development of large liquid undercooling is linked to kinetic control of solidification processes including suppression of heterogeneous nucleation during slow cooling and constrained growth during rapid quenching. The deepest undercoolings have been measured in droplet samples and approach 0.3-0.4·T_m. At high undercooling, solidification can yield metastable product structures, whose constituents are the result of kinetic competition which usually is determined during the nucleation phase and controlled by heterogeneous nucleation. Recently, theoretical, experimental and simulation advances have led to a better understanding of nucleation catalysis reactions and the influence of thermal history during processing at high and low undercoolings. Detailed analysis of the kinetic competition process during solidification along with novel experiments have probed the limits of nucleation theory in both size and time scales and have stimulated new developments in theoretical modeling and computer simulation studies.

View full abstract

An investigation on the static undercooling behavior of Ni and Ni-Co alloys using an alumina crucible under both air and Ar atmosphere was carried out. The degree of undercooling in both Ni and Ni-Co alloys decreased with increasing oxygen content. Impurities of spinels NiAl₂O₄ and CoAl₂O₄ were detected in the low undercooled specimens by XMA and X-ray diffraction methods. Both spinels have low value of planar disregistry which means they have a good potency to be heterogeneous uncleation sites. These inclusions are created as results of reactions between an alumina crucible, soluble oxygen, and molten metal. Furthermore, an equation which gave the fraction solid was derived considering the effect of the crucible. The equation is a function of undercooling, cooling rate, thermal properties of crucible and melt, and time. This equation gives a relationship between undercooling and solidification time by substituting '1' in place of the fraction solid. The calculated solidification time is in good agreement with the experimental data. In low undercooling, the maximum recalescence temperature was almost constant and a refining of casting structure by oxygen was achieved.

View full abstract

The flow in the melt will affect to the morphology of solidification but has rarely been studied quantitatively, because of difficulties in how to bring the controlled flow of melt to the region near the tip of dendrite. In order to realize the flow of melt, the undercooled melt in a cylindrical circular loop cell is partially heated to drive the melt and is forced to flow by the buoyancy. The dendritic growth in this forced melt flow is studied for dendrites growing horizontally in the undercooled melt of pure succinonitrile. The flow affects more significantly to the growth rate of dendrite in lower undercooling than in higher undercooling in the melt. The growth rate of the dendrite tip increases with the increase of forced melt flow when the flow is coming to the tip. When the velocity of the forced flow becomes small, the growth rate of the tip of dendrite approaches one predicted by the theory, where the dendritic growth is studied by the diffusion of heat alone under the marginal stability assumption.

View full abstract

Analytical solutions for melt flow with Oseen viscous flow approximation are obtained for the Navier-Stokes equations in the region near the tip of a dendrite of which shape is approximated to be a paraboloid of revolution. Temperature is also analyzed in the presence of flow in a melt. A theory of dendritic growth is proposed with flow in the undercooled pure melt. Local equilibrium condition at the interface is applied in the theory. The predicted growth rate of the tip of dendrite as functions of the melt undrcooling with and without forced flow velocity is successfully compared with the experimental results for the pure succinonitrile.

View full abstract

Dendritic growth is the common mode of solidification encountered when metals and alloys freeze under low thermal gradients. The growth of dendrites in pure melts is generally acknowledged to be controlled by the transport of latent heat from the moving crystal-melt interface into its supercooled melt. The Ivantsov formulation solves the equation of heat flow from a paraboloidal dendrite tip for the case of diffusive heat transport. However, this formulation is incomplete, and the physics of an additional selection rule, coupled to the Ivantsov solution, is necessary to predict the dendrite tip velocity and radius of curvature as a unique function of the supercooling. Unfortunately, the experimental evidence is not definitive because dendritic growth can be complicated by buoyancy-induced convection, which is normally unavoidable under terrestrial conditions. Recent experiments performed in the microgravity environment of the space shuttle Columbia (STS-62) show quantitatively that convection alters the tip velocities and radii of curvature of dendrites in both terrestrial and microgravity conditions. In addition, these data can be used to evaluate both how well the Ivantsov diffusion solution and the selection rule (the product of the dendrite tip velocity and the tip radius of curvature squared is a constant) match the dendritic growth data under microgravity conditions.

View full abstract

Solidification characteristics in most rapid solidification processes are controlled by heat extraction by the chill substrate and by melt undercooling, especially near the substrate. A dendrite growth model for this situation is derived in the present study. The model involves combined effects of the undercooled melt and the substrate as effective heat sinks. A dendrite is assumed to grow into an undercooled melt, with heat evolution at the dendrite tip occurring both into the undercooled melt and into the chill substrate through the growing solid. The significance of the proposed dendrite model and the stability of dendrite growth under the thermal conditions considered are discussed, using the Fe-Cr-Ni alloy system as an example.

View full abstract

Copper-manganese alloys of a near-azeotropic composition show a rather different solidification behaviour at only slight variations of the concentration. The reason for this is the fact that during the directional solidification vertically upwards the less dense manganese or the more dense copper is enriched in front of the moving solid-liquid interface depending on a hypo- or hyperazeotropic alloy concentration, respectively. Therefore, in the samples solidifying at low solidification velocities with a planar interface a convective unstable or stable behaviour is found, which leads to different concentration profiles in the solidified alloy. At higher solidification rates cellular and dendritic morphologies are produced in a reproducible way and can be excellently metallographically preparated and visualized. In this paper a survey of typical hyper- and hypoazeotropic non-planar morphologies in radial cross-sections is presented. This morphological structures are the basis for statistical evaluations of he patterns describing the interaction between the interface and the heat and mass transfer in the melt.

View full abstract

The effects of cooling rate on the growth behavior of austenite phase during cooling of an iron-carbon alloy are investigated by means of a numerical simulation . In the cooling process of this alloy, austenite phase nucleates at the interface between δ-ferrite and liquid phases at the peritectic temperature 1768 K and then keeps growing during cooling. The growth mechanisms of austenite phase during cooling are: (1) carbon diffusion from liquid phase through austenite phase into δ-ferrite phase, (2) precipitation from δ-ferrite phase, and (3) crystallization from liquid phase. All these mechanisms induce the growth of austenite phase with increasing cooling rate. The ratio of austenite phase which grows by precipitation and crystallization increases with increasing cooling rate, while that by carbon diffusion decreases. The decrease in the ratio of the diffusional growth is more remarkable for the migration of austenite/liquid interface than for that of δ-ferrite/austenite interface.

View full abstract

The solidification path of an Fe-Cr-Ni ternary alloy was calculated numerically on the basis of thermodynamic analysis. In order to evaluate the effect of back diffusion in the solid, three models are used for the analysis. That is, 1) a equilibrium model with perfect diffusion in the solid, 2) the Scheil model with no diffusion, 3) a limited back diffusion model, and obtained results were compared. In the calculations, the equilibrium tie line at the solid-liquid interface, which varies depending on both temperature and composition, was determined for each step of numerical computation. The solidification path with back diffusion was calculated by a new technique combining thermodynamic calculation and diffusion analysis. In this model, δ-γ transformation, which occurs during solidification of the Fe-Cr-Ni ternary system, was dealt with under the assumption that the transformation is diffusion controlled. The relationship between temperature and the fraction solid of the ternary alloy for different cooling rates obtained from the limited back diffusion model was located between those obtained from the equilibrium and Scheil models, demonstrating the validity of the analysis.

View full abstract

During the last decade, solidification modeling has known a sustained development effort, supported by academic as well as industrial research. The driving force behind this undertaking was the promise of predictive capabilities that will allow process and material improvement. The most significant recent rogress has been incorporation of transformation kinetics, for both the liquid/solid and the solid/solid transformation, in the macro-transport models. The results of these efforts have materialized in a proliferation of publications and commercial software, some of which have penetrated the industry. Numerous claims are made regarding modeling methods accuracy and capabilities. They include prediction of casting defects, of microstructure length scale and composition, and even of mechanical properties. A reality test of these claims is the subject of the present paper.
The methodologies for macro transport-transformation kinetics modeling (MT-TK), and therefore for prediction of microstructural evolution, can be broadly classified as being based on the continuum approach (deterministic), or on the stochastic (probabilistic) approach, or, more recently, on a combined approach. Originally the MT-TK analysis was performed for the liquid/solid transformation. Subsequently, it has been extended to the solid/solid transformation, thus resulting in prediction of room temperature microstructure. Specifically, it has been attempted to predict features like microsegregation, microstructure length scale, fraction of phases, structural transitions, hardness, microhardness, and tensile properties. The success of these efforts is critically reviewed in the paper.

View full abstract

Stochastic models have been developed for the simulation of grain structure formation during the solidification of metallic alloys. Nucleation is assumed to occur randomly in space according to a prescribed distribution of nucleation sites. For dendritic alloys, the hypothesis of a random orientation of the ‹100› crystallographic directions of the new nuclei is also made. A cellular automaton (CA) and an interface-tracking technique are used to follow the growth-impingement of dendritic and eutectic grains, respectively. The influence of the local thermal conditions, namely the thermal gradient and the velocity of the isotherm, and of the nucleation parameters on the resulting grain structures is assessed. In particular, it is shown that the asymmetry of the grains along the thermal gradient is an increasing function of the thermal gradient and nucleation undercooling and a decreasing function of the velocity and grain density. The presence of the outer equiaxed zone and the transition from columnar to equiaxed microstructures can also be explained using such models.

View full abstract

In the as-cast state, wrought aluminium alloys present chemical heterogeneities, the so-called microsegregations, which are due to the partitioning of alloying elements between liquid and solid during solidification. While most of the experimental characterizations of solute distribution reported in the literature are dealing with fully solidified material, this paper presents results related to the build-up of microsegregation during the solidification of an aluminium alloy. The features of the corresponding cumulative distributions are presented and then discussed by comparing them to predictions made with a numerical programme which takes into account both the build-up of the main solutes in the liquid and their diffusion in the solid during the solidification. From this comparison, it is concluded that the build-up of microsegregation during the solidification of aluminium alloys could be described as a phenomenon which occurs in a spherical elementary volume element the diameter of which equals the dendrite arm spacing, and which is not greatly affected by the structural coarsening during solidification. The discrepancies between the calculated and measured cumulative distributions at low solid fraction are associated with a bias of the experimental distributions due to the physical noise of X-ray emission.

View full abstract

A model for the solidification of multicomponent steels is presented and used to simulate the solidification of an austenitic steel. Assuming stationary solid phases, conservation of multiple species is considered simultaneously with the solution of the energy and Navier-Stokes equations, with full coupling of the temperature and concentrations through thermodynamic equilibrium requirements. By including finite-rate microscopic solid solute diffusion in the model, the solidus temperature of multicomponent austenitic steels can be accurately calculated. The extension of the model to incorporate a microscopic model of the peritectic transformation is described. A simulation of the austenitic solidification of a steel containing ten elements in a rectangular cavity cooled from the side shows the formation of macrosegregation, channel segregates, and islands of mush surrounded by the bulk melt. The global severity of macrosegregation of an element is found to be linearly dependent on its partition coefficient, although such scaling is not possible locally.

View full abstract

The continuous forging equipment was constructed at No. 3 Continuous Casting Machine of Mizushima Works in 1990 to prevent the centerline segregation in continuously cast blooms. To clarify the mechanism of prevention of centerline segregation of blooms by this process, a mathematical model was developed by considering the discharge of the solute-enriched liquid between dendrites. The model shows that the segregation ratio in the central portion of the bloom decreases as the solid fraction at the forging point decreases. The solute concentration in the central portion of the bloom during forging calculated from the model is in good agreement with the observed values. The solute concentration in the central portion of the bloom can be predicted and controlled to a desired value by changing the solid fraction at the forging point by use of casting speed as a variable. As forging proceeds, a solute-enriched liquid region is formed in the upstream liquid region close to the forging point. The carbon concentration in the solute-enriched liquid region, where the solute accumulates, becomes constant after several meters from the forging point.

View full abstract

It is well known that center segregation of continuous casting steel is much improved by low superheat casting. However, the low superheat casting is not utilized because of the problem of nozzle clogging. The techniques of electromagetic stirring and reduction technique have been employed to improve the segregation even under the high superheat casting. In these methods, a large amount of equiaxed crystals is favorable to maintain the internal quality. Therefore, low superheat casting method by a new type nozzle with air cooling and electromagnetic stirring was devised and teeming experiment using 500 kg or 3 t molten steel was carried out to investigate the cooling ability of the nozzle.
With increasing the intensity of stirring, the flow rate was reduced by the effect of electromagnetic value and the temperature drop in the nozzle was increased from 22 to 50°C. From the experiment, thermal resistances of each part of nozzle were obtained, and it was estimated that temperature drop of 30°C is gained when the nozzle is applied to billet continuous casting.

View full abstract

Remarkable segregation was observed in the modified super clean CrMoV steel forgings for electric power generation applications. Therefore, in order to make clear the mechanism, effect of such elements as Mn, Ni, Cr and Mo on segregation was studied, using 8 tons sand mold ingots and unidirectionally solidified experimental ingots.
As the result, it has been found that remarkable macrosegregation with opposite inclining angle compared with that seen in conventional ingots was formed by the sinking of heavy segregated liquid. It has also found that remarkable carbon segregation was formed because of small partition coefficient of carbon due to high δ solidification ratio and because of heavy density due to high Mo content.

View full abstract

During solidification, particles (inclusions and equiaxed grains) get pushed or engulfed by the freezing front. The redistribution of particles in the solid is controlled by the forces acting on the particles and the flow in the liquid melt. The present paper describes recent work on determining some of the more important parameters which govern the redistribution of particles during solidification.

View full abstract

An analytical steady state model for predicting the critical velocity of engulfment of an insoluble particle by the advancing melt interface was developed. This model couples the thermal and force fields acting on the particle at the liquid-solid interface. Equations that determine the equilibrium distance for the force and thermal fields as a function of velocity were derived. The solution of these equations gives a unique value for the critical velocity of engulfment, and fo the critical particle-interface distance during pushing. The roles of the thermal gradient, of particle and matrix thermal conductivities, and of the interface force were discussed. The model is validated with experimental results for succinonitrile containing SiC or polystyrene particles. Literature data on the critical velocity for the water-tungsten system also compares favorably with calculated results.

View full abstract

Effect of cooling rate on composition of oxides precipitated during solidification have been investigated using Ti deoxidized steel. The composition and size of oxides in continuously cast steels have been measured and theoretically analyzed.
The oxide mainly consists of Ti₂O₃, Al₂O₃ and MnO. The composition of oxide whose diameter is less than about 10 μm changes with the cooling rate during solidification. The Ti₂O₃ content increases and Al₂O₃ content increases and Al₂O₃ content decreases with the decrease in the cooling rate. On the other hand, the composition of oxide whose diameter is larger than about 10 μm does not change with the cooling rate. When the size of oxide is smaller, the effect of the cooling rate on oxide compositions is remarkable.
As a result of the theoretical analysis of the oxide growth during solidification on the basis of a diffusion growth model, it has been found that the increase in the diameter of oxide which grows during solidification is larger when the size of oxide before solidification and the cooling rate during solidification are smaller. The Ti₂O₃ content increases with the decrease in the diameter of oxide before solidification and in the cooling rate. The theoretically estimated results on the changes in compositions and diameter of oxides during solidification of steel qualitatively agree with the observed results.

View full abstract

An original model for the calculation of the precipitation of non-metallic inclusions during solidification has been developed at IRSID. The microsegregation equations and the equilibrium conditions between liquid steel and oxide, sulphide, nitride, carbide inclusions are combined in a general multiphase equilibrium code.
Sulphide precipitation in several steel grades: plate grades, medium carbon and bearing steels has been analyzed in laboratory samples quenched from a partially solidified state. For the various steel grades investigated, the computed results of sulphides (Mn, Fe)S or (Mn, Fe, Cr)S compositions are in good agreement with the results of the experimental study.
Another example concerns the precipitaion of oxide inclusions in semi-killed high-carbon steels: The calculation represents very precisely the observed heterogeneous population of oxide inclusions formed at different stages of the industrial process.

View full abstract

A high temperature tensile test of high carbon steels was executed in the temperature range of the mushy zone. Results were interpreted from the viewpoint of solidification progress which was estimated by numerical simulation of microsegregation. The minimum temperature among those at which ductility is zero corresponded to a solid fraction around 1 for high purity samples; this is because the brittleness in this temperature range is caused by the remaining liquid. Low purity samples exhibited a different ductility.

View full abstract

The deformation of initial solid shell has important effects on the formation of surface defects of continuously cast slabs of steel. In order to understand the thermal deformation behavior of initial solid shell of steel, droplets of Fe-C alloys and commercial steels were solidified on a chill plate and the bottom shape of solid shell was measured. The solid shell always deforms during solidification and cooling to give a spherical bottom shape. This deformation of solid shell shows a very strong dependence on carbon content. When carbon content is lower than about 0.6 mass% or higher than 2.5 mass%, solid shells show positive deformation, i.e. convex towards the chill. When crbon content lays between 0.6-2.5 mass%, solid shells show negative deformation, i.e. concave towards the chill. Two peaks of positive deformation were found as carbon contents come near to 0 mass% (pure iron) or 0.12 mass%. These two peaks of positive deformation of solid shell correspond to the reported peaks of frequencies of surface defect occurrence such as longitudinal cracks and deep oscillation mark of continuously cast steel slabs. The deformation of pure iron solid shell was accurately pridicted with a deformation model of initial solid shell previously derived by the authors. An explanation for the maximum deformation at 0.12 mass% carbon conent was also made by the model using a numerical heat transfer calculation with a coupling of deformation and solid shell-chill interface heat transfer coefficient.

View full abstract

Mathematical modeling of stress generation in casting processes is a difficult, complex subject that is now receiving increased attention. This paper reviews the basic equations, solution methods and important phenomena associated with casting processes that require special numerical treatment. Stress modeling begins with a coupled, transient heat transfer analysis, including solidification, shrinkage-dependent interfacial heat transfer, and fluid flow effects. Further complicating phenomena inlude phase transformations, temperature, stress and structure-dependent plastic-creep, interaction between the casting and the mold, hydrostatic pressure from the liquid, the effects of fluid flow, and crack formation. Computational issues include numerical methods for handling these phenomena, mesh refinement, and two-dimensional stress state. Example applications are presented for the thermal-mechanical behavior of the solidifying steel shell in the mold region of a continuous slab caster, using a finite-element model, which accounts for many of these phenomena.

View full abstract

Steady-state solidification processes of the crystal pulling method for single crystal production and unidirectional solidification by the zone melting method from the partial molten state for production of highly oriented polycrystalline crystals were reviewed. In the case of the single crystal pulling, large faceted (1.45 cm×1.45 cm×13 mm) YBCO crystals are continuously pulled along either c-axis or a-axis at about 0.1 mm/h of the pulling rate. In the case of the zone melting, results indicated that higher G/R values were preferable for the continuous growth. The sample grown at 1 mm/h possessed a large faceted solid/liquid interface and the volume fraction of 211 phase particles drastically decreases from in the liquid to the 123 phase solids at the interface. The result leads to the idea that the necessary solute for peritectic reaction is provided from 211 particles to 123 interface through a liquid. Based on this idea, we developed a simple solidification model which is in good agreement with the experimental results.

View full abstract

Advances in predictive modelling and in the application of experimental techniques in which conditions of formation are well-difined have enabled considerable progress to be made over the last decade in matching predictions with experimental fact in relation to the formation and characteristics of rapidly solidified microstructure. This progress is reviewed and evaluated with the conclusion that a basic framework for prediction is now available for application by both alloy designers and regular users of rapid solidification processing technologies.

View full abstract

Laser cladding is an effective material processing that produces a surface layer havig good wear and corrosion properties with minimized dilution. In the present work, Co based hardfacing alloy claddings by the blown powder process and Ni-Cr-Al-Y alloy claddings by the preplaced powder process were performed. Formation behavior of the lase clad layers and their properties were investigated. Low dilution in the laser cladding process using powder materials was attributed to low melting efficiency for substrate melt due to inidirect heat transfer via melting clad layer. For wear properties, carbide reinforcing improved wear resistance of the Co based hardfacing alloy clad layer. Superior oxidation resistance was obtained in the Ni-Cr-Al-Y claddings. Chromium carbide reinforcing improved wear resistance of the Ni-Cr-Al-Y alloy cladding with a little sacrifice of oxidation resistance.

View full abstract

The Plasma-Rotating-Electrode-Process (PREP) is based on the pulverization by rotation of meta bars in contact with a Ar/N₂ plasma arc. Due to the high centrifugal forces of the rapidly rotating bars, high nitrogen steel powders are produced with diameters in the range of 0.02 to 0.5 mm. The low diameters and the high centrifugal forces, as well as the high particle velocities, cause the steel droplets to cool down rapidly in the reactor chamber. Mathematical calculations show that cooling rates of up to 10⁵ K/s are attained. It is demonstrated that the cooling rate of all powder particles and the structures produced can be predicted.

View full abstract

In order to elucidate the formation of solidification structure in twin roll process, casting experiments of SUS304 austenitic stainless steel were done by using a laboratory scale twin roll caster. In the cast strips, there exist two kinds of solidification structures; columnar dendrite and equiaxed crystal zones. With increasing contact time between roll and metal, the thicknesses of the strip and the dendritic zone increase, while that of equiaxed zone is kept almost constant. With increasing initial roll gap, only the equiaxed zone is enlarged. Furthermore, effects of the superheat of molten steel and the roll supporting force on the formation of solidification structure have been made clear.
The columnar dendrite zone is formed when the solidifying shells are contacting with rolls. An unsolidified layer remained in the strip solidifies under slow cooling rate after leaving roll nip to form a equiaxed zone. On the basis of the experimental results and theoretical analysis of heat transfer in the strip, it has been deduced that, under the conditions of relatively high superheat of molten metal and small roll supporting force, the equiaxed zone results from the preferential growth of free crystals ahead of the dendritic solidification front, with the abrupt decrease in the heat transfer coefficient on strip surface after the strip goes away from rolls.

View full abstract

The rapid solidification process, which omits the hot rolling stage in the production of thin strip, was used to produce materials which typically have poor workability. Ni base alloys were cast into a thin strip less than 1 mm in thickness by the twin roll process. A cast strip was continuously coiled with low tension control and secondary cooling. Coils of these alloys are being delivered as overlay welding hoops after appropriate post treatment.
For quality improvement, various aspects of twin roll casting have been investigated. Melt level control by side-dams and roll profile control have decreased the strip thickness deviation to within ±7%. Internal cavities and localized shrinkage were often detected in strips of these alloys made with flat rolls, while they are rarely detected in the strips produced with grooved roll.
This difference was studied in a two dimensonal calculation of solidification. It has been considered that these alloys, if cast with flat rolls, have a longer liquid phase in the strip than the strip cast with grooved rolls, when gas or scale is trapped between the roll and the melt during casting, and that the elongated liquid phase forms internal cavities due to the difficulty of melt flow to the final solidification point.

View full abstract

For the development of strip casting technique for direct production of thin strips from molten metal, casting experiments of SUS304 steel is made in a twin roll strip caster. Longitudinal and transverse cracks and small depressions are the main defects appearing on the surface of directly cast strips. Water model experiments are applied to find the conditions for pouring the molten metal uniformly over the whole strip width and reducing the fluctuations of meniscus. By reduction of the fall height of pouring metal, optimization of pouring position and increase in the ratio of pouring rate to molten steel weight on the rolls, uniformity in the thickness of solidified shell is achieved and consequently occurrence of the longitudinal cracks and small depressions are decreased. In addition, the heat transfer coefficient between rolls and cast strip is determined from the measurements such as temperature changes in cooling water, temperature in the interior of rolls and on the surface of casting strips during casting. By using the measured heat transfer coefficient, optimum casting conditions are found for the completion of solidification at the kissing point of rolls to optimize the roll separating force on cast strips and thus for the prevention of transverse cracks.

View full abstract

A new process to manufacture semi-solid alloys was proposed. A teting machine was built and a series of experiments were carried out by using Pb-Sn-aloys, aluminum alloys and cast irons with aim to characterize the process.
In each case, semi-solid alloys with varied levels of fraction solid, from low to high, were successfully manufactured by changing temperature of roll and shoe surfaces, gap between them, roll speed and superheating temperature of supplied molten metal. Through the tests, it has become clear that the proposed process is effective to manufacture various semi-solid alloys.
Mechanical properties of the alloys completely solidified after the semi-solid processing were investigated. The flow stress, hardness, elongation and deformability of the semi-solid processed aloys were measured. As the results, it becomes clear that their alloys have better quality than the conventionally cast alloys and, in some cases, their quality is distinctively improved by the semi-solid processing.

View full abstract

The aim of this paper is to study the influence of solidification conditions on the structure of semi-solid alloys achieved during partial remelting for various holding times and to determine the corresponding rheological behaviour during compression tests. A357 Al-alloys either solidified under pressure, or cast with inoculation, or electromagnetically stirred were used for this investigation. Compression tests were performed both at constant strain rate and with strain rate jumps to characterize the thixotropic behaviour of these alloys. Depending on the holding time in the semi-solid state before compression two main rheological regimes were observed after strain rate jumps: a first one typical of a pseudo-plastic behaviour (only observed for the pressurized material for small holding times) and a second one at long holding time for which the stress evolves with strain after the jump. Extensive study of the microstructure before and after deformation using image analysis coupled with some rheological study of composite materials was performed to allow correlation between microstructure and mechanical bbehaviour.

View full abstract

The literature survey about the thermomechanical behavior of the mushy state first demonstrates the lack of data about the rheological parameters for viscoplastic component. Then it shows discrepancies between experimental and modelling studies about the elastic component. In order to fill the gap underlined in the first point, we develop a new methodology which consists in associating a simple test geometry (needle indentation test), able to prduce the suitable strains and strain rates, with an FEM analysis of the induced flow. The thermal dependance of the power creep law (Norton-Hoff) can than be identified. The results obtained for the Al-4.5%Cu-Mg-Ti alloy are presented and applied to the thermomechanical prediction of air gap in permanent mold casting by comparisons of measurements and calculations.

View full abstract

A new continuous rheocasting process for steels and high temperature alloys has been developed by Inland Steel incorporating a dual chamber casting machine utilizing separate electromagnetic stirrers around each chamber. The caster parameters were varied in a series of experiments to determine the optimum combination to provide a fine degenerate dendrite size, small peripheral dendritic zone, and minimum percent dendricity. The optimized process produces a degenerate dendrite size of about 80 microns with a peripheral dendritic zone about 1.5 mm thick and no dendricity. A series of casts was made of 12 different steel and high temperature alloys to determine the capability of the Inland Steel rheocaster to provide a satisfactory rheocast structure on various alloys. The rheocaster produced an acceptable rheocast structure for all alloys cast. Comparison of the cast structure of the Inland Steel rheocast material to conventional continuous cast material has shown that highly alloyed materials can be cast using rheocasting and provide a fine, homogeneous, and unsegregated cast structure. This is not the case for conventional continuous casting processes available currently.

View full abstract