The influence of alloying elements on graphite morphology and cooling curves was studied using Ni-C eutectic alloys. Ninety grams of Ni-C parent alloys, prepared using high purity metals, was melted in an Ar atmosphere at 1500°C and cooled at the rate of 40K/min. Their cooing curves were monitored every 0.1s using a B-type thermocouple located at the center of the samples. The graphite morphologies of the samples were observed using an SEM and optical microscope. <br>  Experimental results showed that spheroidal graphite formed after the crystallization of chunky graphite. This phenomenon completely differed from that of the Fe-C alloys, in which after the spheroidal formation the chunky graphite crystallizes. These results were confirmed by the cooling curves and microstructure.

The aim of this study was to fabricate thin and light castings to realize an energy saving highly efficient society. High melt flow rate in a mold cavity is indispensable for the production of thin castings. However, until now, the flow ability of melt has been studied mainly by investigating flow length, but the flow rate is thought to be more important and affected by permeability, specific heat of sand mold, and atmosphere. Consequently, the influence of permeability, atmosphere, air, CO₂, and He on the flow rate was investigated using a fine Teflon tube with pure water system in this study. <br>  The results show that the flow rate is significantly affected by permeability and gas species, and is the fastest in a He atmosphere, followed by air and CO₂. These results show that the flow rate is a function of the specific density of gases and is not the function of kinetic viscosity.

The influence of boron on the ferrite formation in pearlite matrix spheroidal graphite cast iron has been investigated using the advanced analytical equipments such as an SEM and a TEM. The surface of spheroidal graphite in the bulls eye structure, produced by the addition of quite small amount of boron, was analyzed for discussing the ferritizing mechanism. <br>  The microstructure in the spheroidal graphite was investigated using the SEM and TEM, and we found the triangular precipitates, consisted of Fe-Si-Cu-B size of 150nm, in the surface layer of the graphite. These precipitates are consisted of the nano-crystal and act as the nucleus sites of flake graphite size of 0.5μm. The formation of the flake graphite due to the ferritizing above their eutectoid temperature, is the key mechanism of the ferrite formation.

In order to experimentally elucidate foam stabilizing factors in aluminum foam fabrication, the contribution of solid particles in the liquid to increasing viscosity and stabilizing foam was investigated using both solution and aluminum melt. Regardless of the type liquid, the wettability and shape of solid particle showed opposite effects between thickening and foam stability. Even though a poor wettable particle contributes effectively to enhancing the liquid viscosity, the particle leads to the collapse of the foam cell because it acts like a notch in terms of foam stabilization, because of poor energy balance at the interface between solid and liquid. For this reason, it is important to take into account not only the relative viscosity but also the physical properties of solid particle for the fabrication of fine aluminum foams.

In order to examine eutectic solidification in Al-Si alloy, the role of the purity of Al-Si alloy and the eutectic modification agent, strontium or sodium, on eutectic solidification sequences was investigated by a technique of quenching during eutectic reaction. <br>  If Al-Si alloy contains more than a few ppm of phosphorous (commercial purity alloys), eutectic cells in Al-Si alloy melt grow on centrally-located polygonal silicon which nucleates on AlP particles. However, nucleation of eutectic cells tends to be difficult within the melt with sodium addition due to low phosphorous content, even though the sample is made of identical commercial purity start materials. In the case of samples containing little phosphorous, eutectic phases tend to nucleate on the mold surface and grow in the opposite direction of the heat flow. <br>  Addition of strontium or sodium results in the morphology of eutectic silicon, which acts as the leading phase of eutectic coupled growth and promotes transition from plate to fiber (coral) shape with decreased anisotropy. This causes eutectic aluminum to grow according to the crystal orientation of the eutectic silicon in unmodified melt, but grow freely with subgrain formations in modified melt. <br>  These results reveal that eutectic solidification of Al-Si alloy is controlled by the nucleation frequency of eutectic cells depending on phosphorous content and decrease in the anisotropy of eutectic silicon due to modification.

TiH2 powder has been used to fabricate aluminum foams its a blowing agent for more than two decades. The aim of this paper is to understand the detailed decomposition behavior of TiH2 powder and to control the phenomenon by a heat treatment for the fabrication of fine foams by the melt route. TiH2 powders whose qualities were different, were characterized using TG-DTA and XRD. As heat treatment factors. temperature and time were applied. We have found differences in the decomposition behavior between the as-received Tilt, powders. Regardless of quality of TiH2, powder. increasing temperature and time of the heat treatment in air elevates the decomposition temperature and decreases the amount of released hydrogen during reheating due to the oxide barrier layer formed on the powder surface. The influence of the heat treatment temperature on the decomposition modification is more significant than that of the little. To control the decomposition phenomenon of TiH2 powder, the heat treatment condition has to be optimized taking into account the quality and the purity of the powder. [doi: 10.2320/matertrans.MBW200829]

TiH2 powder has been used to fabricate aluminum foams its a blowing agent for more than two decades. The aim of this paper is to understand the detailed decomposition behavior of TiH2 powder and to control the phenomenon by a heat treatment for the fabrication of fine foams by the melt route. TiH2 powders whose qualities were different, were characterized using TG-DTA and XRD. As heat treatment factors. temperature and time were applied. We have found differences in the decomposition behavior between the as-received Tilt, powders. Regardless of quality of TiH2, powder. increasing temperature and time of the heat treatment in air elevates the decomposition temperature and decreases the amount of released hydrogen during reheating due to the oxide barrier layer formed on the powder surface. The influence of the heat treatment temperature on the decomposition modification is more significant than that of the little. To control the decomposition phenomenon of TiH2 powder, the heat treatment condition has to be optimized taking into account the quality and the purity of the powder. [doi: 10.2320/matertrans.MBW200829]

The influence of mass transfer and chemical reactions on wetting has been studied using metal systems and aqueous solution systems to determine the equilibrium wetting. The influence of mass transfer on wetting has been studied using Al/Si systems and aqueous solution/NaCl systems and the influence of chemical reactions is discussed using Al/Ni systems and aqueous solution/B2O3 systems. The reaction products in the Al/Ni system change with temperature to NiAl3, Ni2Al3 and NiAl, and the reaction product for the aqueous solution/B2O3 system is B(OH)(3). also discuss the influence of relative humidity on wetting in the aqueous solution systems. The results show that the contact angle is highest under equilibrium humidity conditions. (C) 2008 Published by Elsevier B.V.

The influence of mass transfer and chemical reactions on wetting has been studied using metal systems and aqueous solution systems to determine the equilibrium wetting. The influence of mass transfer on wetting has been studied using Al/Si systems and aqueous solution/NaCl systems and the influence of chemical reactions is discussed using Al/Ni systems and aqueous solution/B2O3 systems. The reaction products in the Al/Ni system change with temperature to NiAl3, Ni2Al3 and NiAl, and the reaction product for the aqueous solution/B2O3 system is B(OH)(3). also discuss the influence of relative humidity on wetting in the aqueous solution systems. The results show that the contact angle is highest under equilibrium humidity conditions. (C) 2008 Published by Elsevier B.V.

Inoculation of cast iron has been mainly studied for the nucleus materials and the development of new inoculants. The effect is widely recognised
nevertheless, the mechanism was not fully understood. Inoculation leads to a decrease in chill depth, depressed the formation of D type graphite to form A type graphite and increase in the eutectic cell number. The transition from A type to the D type graphite is also accelerated by Ti addition
therefore we used Ti added samples for the study of D type graphite formation. The increase in the eutectic cell number produces a decrease in the cell size, namely the radius
nevertheless, the solidification time of the castings does not change by inoculation. Therefore, the solidification rate of the cell, namely the linear solidification rate, should decrease with inoculation and the decrease produces a depression in the chill and D type graphite due to the smaller undercooling. Therefore, we have to consider the three-dimensional solidification rate for discussing the cooling curve. © 2008 W. S. Maney &amp
Son Ltd.

Flake graphite cast iron (FC) samples containing 2%Si, in which the S contents changed from 0.006 to 0.076%, were heated at 1123 to 1323 K and then cooled to determine the pearlite morphological change, namely lamellar or granular. For the spheroidal graphite cast iron (FCD) samples, the P contents changed from 0 to 0.2% and those of Sn were from 0 to 0.06% with 0.5% Cu using the 2.5% Si and 0.3% Mn iron.
The pearlite morphology of the FC samples cooled from 1123 K was granular. The pearlite structure of the Sn-FCD samples was also granular, when cooled from 973 and 1023 K, and that of P-FCD was granular only in the sample cooled from 1023 K, otherwise the morphology was lamellar.
The pearlite morphology of FC and FCD samples depends on the solubility of C in the austenite which can be varied by the heating temperature. This mechanism is identical to that of the lamellar to rod transition in eutectic alloys.

The effects of hydrogen levels on microstructure of porosity and mechanical properties of magnesium alloy castings were investigated. The hydrogen content of AZ91 alloy melt under Ar+HFC-134a mixed gas protection was 103 cm(3) kg(-1), which was less than that of 139 cm(3) kg(-1) under flux protection, therefore the effect of gas protection was better. The porosity ratio in microstructure of castings melted under gas protection was 1.8%, which was also less than that of 2.6% under flux protection. The hydrogen content of melt under flux protection and degassed with Ar gas was 70 cm(3) kg(-1), and the corresponding porosity ratio in microstructure of castings was only 0.6%. The density of the samples was increased with decreasing hydrogen content. The tensile strength of AZ91 alloy casting sample under flux protection and degassed with Ar gas was 21% higher than that of undegassed one, and the elongation was also 50% higher.

Inoculation of cast iron has been mainly studied for the nucleus materials and the development of new inoculants. The effect is widely recognised; nevertheless, the mechanism was not fully understood. Inoculation leads to a decrease in chill depth, depressed the formation of D type graphite to form A type graphite and increase in the eutectic cell number. The transition from A type to the D type graphite is also accelerated by Ti addition; therefore we used Ti added samples for the study of D type graphite formation. The increase in the eutectic cell number produces a decrease in the cell size, namely the radius; nevertheless, the solidification time of the castings does not change by inoculation. Therefore, the solidification rate of the cell, namely the linear solidification rate, should decrease with inoculation and the decrease produces a depression in the chill and D type graphite due to the smaller undercooling. Therefore, we have to consider the three-dimensional solidification rate for discussing the cooling curve.

Flake graphite cast iron (FC) samples containing 2%Si, in which the S contents changed from 0.006 to 0.076%, were heated at 1123 to 1323 K and then cooled to determine the pearlite morphological change, namely lamellar or granular. For the spheroidal graphite cast iron (FCD) samples, the P contents changed from 0 to 0.2% and those of Sn were from 0 to 0.06% with 0.5% Cu using the 2.5% Si and 0.3% Mn iron.
The pearlite morphology of the FC samples cooled from 1123 K was granular. The pearlite structure of the Sn-FCD samples was also granular, when cooled from 973 and 1023 K, and that of P-FCD was granular only in the sample cooled from 1023 K, otherwise the morphology was lamellar.
The pearlite morphology of FC and FCD samples depends on the solubility of C in the austenite which can be varied by the heating temperature. This mechanism is identical to that of the lamellar to rod transition in eutectic alloys.

The effects of hydrogen levels on microstructure of porosity and mechanical properties of magnesium alloy castings were investigated. The hydrogen content of AZ91 alloy melt under Ar+HFC-134a mixed gas protection was 103 cm(3) kg(-1), which was less than that of 139 cm(3) kg(-1) under flux protection, therefore the effect of gas protection was better. The porosity ratio in microstructure of castings melted under gas protection was 1.8%, which was also less than that of 2.6% under flux protection. The hydrogen content of melt under flux protection and degassed with Ar gas was 70 cm(3) kg(-1), and the corresponding porosity ratio in microstructure of castings was only 0.6%. The density of the samples was increased with decreasing hydrogen content. The tensile strength of AZ91 alloy casting sample under flux protection and degassed with Ar gas was 21% higher than that of undegassed one, and the elongation was also 50% higher.

The formation mechanism of chunky graphite has been reviewed and studied. The study consisted of a unidirectional solidification method, a small droplet method and a furnace cooling method. Four kinds of iron samples were prepared, namely, the pure Fe-C, Fe-C-S, Fe-C-Ce and Fe-C-Si-Ce alloys, and three kinds of nickel samples, namely the Ni-C, Ni-C-S and Ni-C-Mg alloys. The results of the unidirectional solidification of the Ni-C alloys showed that spheroidal graphite is not observed in the continuous solidified region, in which only flake-like graphite is observed, while spheroidal graphite is usually observed in the quenched liquid region. The existence of spheroidal graphite in the solidified phase is recognized only in the discontinuous growth mode of the Ni-C-Mg alloy solidified at 150 mm/h. This means that the spheroidal graphite is directly crystallized from the melt and entrapped by the flake-like chunky graphite that is formed by the continuous growth mode. In the small droplet method, a small piece of the Fe-C or Fe-C-Ce sample was melted on a pure graphite plate then cooled at a different cooling rate in a He-3%H-2 atmosphere. The graphite in the Fe-C-Ce alloy is usually spherical. Nevertheless, the graphite morphology of the final solidified area changed from spherical to chunky and chunky to ledeburite with an increase in the cooling rate. This means that the chunky graphite is formed in the residual liquid region by the solidification into Fe-graphite system. The sample was cooled in a furnace, and the graphite morphology changes from spherical to chunky and chunky to ledeburite with the decrease in the Si content. These phenomena can be confirmed by the cooling curves of these samples.

The formation mechanism of chunky graphite has been reviewed and studied. The study consisted of a unidirectional solidification method, a small droplet method and a furnace cooling method. Four kinds of iron samples were prepared, namely, the pure Fe-C, Fe-C-S, Fe-C-Ce and Fe-C-Si-Ce alloys, and three kinds of nickel samples, namely the Ni-C, Ni-C-S and Ni-C-Mg alloys. The results of the unidirectional solidification of the Ni-C alloys showed that spheroidal graphite is not observed in the continuous solidified region, in which only flake-like graphite is observed, while spheroidal graphite is usually observed in the quenched liquid region. The existence of spheroidal graphite in the solidified phase is recognized only in the discontinuous growth mode of the Ni-C-Mg alloy solidified at 150 mm/h. This means that the spheroidal graphite is directly crystallized from the melt and entrapped by the flake-like chunky graphite that is formed by the continuous growth mode. In the small droplet method, a small piece of the Fe-C or Fe-C-Ce sample was melted on a pure graphite plate then cooled at a different cooling rate in a He-3%H-2 atmosphere. The graphite in the Fe-C-Ce alloy is usually spherical. Nevertheless, the graphite morphology of the final solidified area changed from spherical to chunky and chunky to ledeburite with an increase in the cooling rate. This means that the chunky graphite is formed in the residual liquid region by the solidification into Fe-graphite system. The sample was cooled in a furnace, and the graphite morphology changes from spherical to chunky and chunky to ledeburite with the decrease in the Si content. These phenomena can be confirmed by the cooling curves of these samples.

A new technique of introducing mechanical vibration during isothermal holding period of hypoeutectic Al-Si alloy to prepare semi-solid slurry has been studied. The convection of melt caused by vibration had notable effects on the formation of near-spherical primary Al particles. When the vibrating frequency increased from 11.7 to 33.3 Hz, the average diameter of non-dendritic primary particles decreased from 125 to 90 gm, and the average shape coefficient increased from 0.40 to 0.55. (C) 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

A new technique of introducing mechanical vibration during isothermal holding period of hypoeutectic Al-Si alloy to prepare semi-solid slurry has been studied. The convection of melt caused by vibration had notable effects on the formation of near-spherical primary Al particles. When the vibrating frequency increased from 11.7 to 33.3 Hz, the average diameter of non-dendritic primary particles decreased from 125 to 90 gm, and the average shape coefficient increased from 0.40 to 0.55. (C) 2007 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

The influence of the interfacial energy between the Fe-C melt and basal plane of graphite, YGr/L, and the cooling rate on the formation of spheroidal graphite was investigated as a function of the sulfur content in cast iron using the sessile drop method in a He-3% H-2 atmosphere. The result showed that the gamma(Gr/L) decreased from 2.62 to 1.48 J/m(2) with an increase in the sulfur. The corresponding surface energy of the graphite, gamma(Gr/V), also varied from 1.17 to 0.65 J/m(2). It is found that the spheroidal graphite is formed when the gamma(Gr/V) is greater than 2.5 J/m(2), which is the critical interfacial energy of the spheroidal graphite formation. In addition, the cooling rate is also an important factor for producing the spheroidal graphite in the Fe-C alloy. The present study clearly shows that the critical cooling rate is a function of the sulfur content, namely gamma(Gr/V), in the Fe-C melt. (c) 2007 Elsevier B.V. All rights reserved.

The influence of the interfacial energy between the Fe-C melt and basal plane of graphite, YGr/L, and the cooling rate on the formation of spheroidal graphite was investigated as a function of the sulfur content in cast iron using the sessile drop method in a He-3% H-2 atmosphere. The result showed that the gamma(Gr/L) decreased from 2.62 to 1.48 J/m(2) with an increase in the sulfur. The corresponding surface energy of the graphite, gamma(Gr/V), also varied from 1.17 to 0.65 J/m(2). It is found that the spheroidal graphite is formed when the gamma(Gr/V) is greater than 2.5 J/m(2), which is the critical interfacial energy of the spheroidal graphite formation. In addition, the cooling rate is also an important factor for producing the spheroidal graphite in the Fe-C alloy. The present study clearly shows that the critical cooling rate is a function of the sulfur content, namely gamma(Gr/V), in the Fe-C melt. (c) 2007 Elsevier B.V. All rights reserved.

The shrinkage and chilling tendency of spheroidal graphite (abbreviated SG) cast iron is much greater than that of the flake graphite cast iron in spite of its higher amount of C and Si contents. Why? The main reason should be the difference in their graphitization during the eutectic solidification. In this paper, we discuss the difference in the solidification mechanism of both cast irons for solving these problems using unidirectional solidification and the cooling curves of the spheroidal graphite cast iron. The eutectic solidification rate of the SG cast iron is controlled by the diffusion of carbon through the austenite shell, and the final thickness is 1.4 times the radius of the SG, therefore, the reduction of the SG size, namely, the increase in the number, is the main solution of these problems.

The shrinkage and chilling tendency of spheroidal graphite (abbreviated SG) cast iron is much greater than that of the flake graphite cast iron in spite of its higher amount of C and Si contents. Why? The main reason should be the difference in their graphitization during the eutectic solidification. In this paper, we discuss the difference in the solidification mechanism of both cast irons for solving these problems using unidirectional solidification and the cooling curves of the spheroidal graphite cast iron. The eutectic solidification rate of the SG cast iron is controlled by the diffusion of carbon through the austenite shell, and the final thickness is 1.4 times the radius of the SG, therefore, the reduction of the SG size, namely, the increase in the number, is the main solution of these problems.

The significant features of closed cell aluminum foams are low density, high specific strength, high energy absorption, etc. These advantages are affected by the size, shape, and uniformity of cells in the foam. In order to improve the cellular structure of metal foams, foam stability of liquid metals to prevent rapture and drainage events has to be understood and controlled. A clear understanding of factors is therefore expected to improve quality and process control more easily. <br>  In this study, the influence of the temperature for adding foaming agent viscosity of the melt on foamability, and the growing mechanism of hydrogen bubbles in aluminum melt were investigated. Aluminum alloy foams were fabricated using the powder decomposition in the melt process and the unidirectional solidification method. As starting materials, Al-25mass%Cu-5mass%Si and Al-12.6mass%Si alloys were used. It was found that the temperature of foaming agent addition affected the number of cells and cell wall thickness. This effect was greater than those of viscosity and alloy components. Namely, the drainage of the liquid in foam film is the main factor for fabricating fine aluminum foam effectively.

The growth process of a new chromium-free multi-elements complex coating (MECC) on AZ91D magnesium alloy has been studied. The morphology and phase composition of the coatings have been determined with SEM, EDX, XRD and AFM. In addition, the thickness change of MECC layer and weight change of the specimen have been measured. On the basis of these experimental results, the coating formation is divided into initial stage, metaphase stage and late stage. At the initial stage, the compact coating is an amorphous material composed of O, Mg, Al and P. Meanwhile, coating growth is fast and mass-gain in unit area is negative owing to severe corrosion. At the metaphase stage, the coating layer consists of amorphous material of O, Mg, Al, P, Mn, and crystalline Mn5.65P3, (Mg0.66Al0.34) (Al0.83Mg0.17)(2)O-4, and the mass-gain in unit area decreases and coating quality degenerates. At the late stage, with the quantity increase of crystalline Mn5.65P3, (Mg0.66Al0.34) (Al0.83Mg0.17)(2)O-4, the coating surface becomes smooth and compact, and the coating grows into steady condition. (c) 2006 Elsevier B.V. All rights reserved.

The influence of wettability between molten metal and mold materials on fluidity has been studied using glass and Teflon fine tubes/aqua-solutions system. For the discussion of work of adhesion and external pressure at the tip of flow on fluidity, flow rates were measured in laminar flow conditions. Flow was induced by the suction of solution with reduced pressure. The flow rate was calculated from images of a high speed video recorder. The inside diameter of the tubes was changed from 1.8mm to 4.0mm. The contact angle (wettability) was changed by the combination of tube materials and addition of a surfactant into pure water. <br>  The flow rate in the wetting system (water/glass tube system) was higher than that of the non-wetting system (water/Teflon tube one) due to the external pressure caused by interfacial energy, namely the contact angles and surface tension. These results indicate that the influence of the work of adhesion is considerably weaker than that of external pressure.

Aluminum foams are very attractive materials for automobile and aerospace applications, where the weight reduction and the improvement in safety devices as energy absorbing materials are requested. These advantages should be affected by the size of the foam cell, namely, the smaller the better. Therefore, we shall discuss how to make finer foams. The most popular fabrication process of aluminum foams is particle decomposition in the melt process using TiH2 powder addition. In this study, by using an aqueous solution as a visible model, dynamic changes in the shape of the meniscus and the force which acts on a particle were examined as a function of the contact angle during the transfer process. We propose a model which can estimate the force necessary to entrap the particle in the liquid. Furthermore, the optimum stirring conditions for entrapping the particle and for uniformly dispersing it in the liquid were also investigated using the visible model. During the fabrication of the aluminum foams, as the starting materials, pure aluminum and Al-12.6Si alloys were used. The viscosity of these alloys increased to the objective value by the addition of Ca and aluminum powder, then TiH2 powder was added to the melts using the pre-optimized stirring conditions. It was found that the number of cells per unit area and the cell wall thickness were determined by the viscosity of the molten alloys.

The growth process of a new chromium-free multi-elements complex coating (MECC) on AZ91D magnesium alloy has been studied. The morphology and phase composition of the coatings have been determined with SEM, EDX, XRD and AFM. In addition, the thickness change of MECC layer and weight change of the specimen have been measured. On the basis of these experimental results, the coating formation is divided into initial stage, metaphase stage and late stage. At the initial stage, the compact coating is an amorphous material composed of O, Mg, Al and P. Meanwhile, coating growth is fast and mass-gain in unit area is negative owing to severe corrosion. At the metaphase stage, the coating layer consists of amorphous material of O, Mg, Al, P, Mn, and crystalline Mn5.65P3, (Mg0.66Al0.34) (Al0.83Mg0.17)(2)O-4, and the mass-gain in unit area decreases and coating quality degenerates. At the late stage, with the quantity increase of crystalline Mn5.65P3, (Mg0.66Al0.34) (Al0.83Mg0.17)(2)O-4, the coating surface becomes smooth and compact, and the coating grows into steady condition. (c) 2006 Elsevier B.V. All rights reserved.

Aluminum foams are very attractive materials for automobile and aerospace applications, where the weight reduction and the improvement in safety devices as energy absorbing materials are requested. These advantages should be affected by the size of the foam cell, namely, the smaller the better. Therefore, we shall discuss how to make finer foams. The most popular fabrication process of aluminum foams is particle decomposition in the melt process using TiH 2 powder addition. In this study, by using an aqueous solution as a visible model, dynamic changes in the shape of the meniscus and the force which acts on a particle were examined as a function of the contact angle during the transfer process. We propose a model which can estimate the force necessary to entrap the particle in the liquid. Furthermore, the optimum stirring conditions for entrapping the particle and for uniformly dispersing it in the liquid were also investigated using the visible model. During the fabrication of the aluminum foams, as the starting materials, pure aluminum and Al-12.6Si alloys were used. The viscosity of these alloys increased to the objective value by the addition of Ca and aluminum powder, then TiH2 powder was added to the melts using the pre-optimized stirring conditions. It was found that the number of cells per unit area and the cell wall thickness were determined by the viscosity of the molten alloys.

A chromium-free conversion coating for AZ91D magnesium alloy has been obtained by using a phosphate-penrianganate solution, which has been developed out without acid pickling. Examinations have been carried out on the conversion coatings for morphology, composition, adhesion force, and corrosion resistance. Results show that the conversion coatings are relatively uniform and continuous, with thickness from 7 mu m to 10 mu m. They demonstrate good adhesion to matrix and have some non-penetrating tiny holes on the surface. The main elements of the conversion coating of AZ91D alloy are Mg, O, P, K, Al and Mn by EPMA analysis. Results of corrosion resistance test indicate that the corrosion resistance of the conversion coating by phosphate-permanganate solution is matched to that of the conversion coating by traditional chromate solution, but for the corrosion resistance after painting, the former is better than the later. (c) 2005 Elsevier B.V. All rights reserved.

A chromium-free conversion coating for AZ91D magnesium alloy has been obtained by using a phosphate-penrianganate solution, which has been developed out without acid pickling. Examinations have been carried out on the conversion coatings for morphology, composition, adhesion force, and corrosion resistance. Results show that the conversion coatings are relatively uniform and continuous, with thickness from 7 mu m to 10 mu m. They demonstrate good adhesion to matrix and have some non-penetrating tiny holes on the surface. The main elements of the conversion coating of AZ91D alloy are Mg, O, P, K, Al and Mn by EPMA analysis. Results of corrosion resistance test indicate that the corrosion resistance of the conversion coating by phosphate-permanganate solution is matched to that of the conversion coating by traditional chromate solution, but for the corrosion resistance after painting, the former is better than the later. (c) 2005 Elsevier B.V. All rights reserved.

The present research was conducted to verify the effects of sulfur and oxygen on graphite morphologies using a Fe-C-120 massppm S mother alloy. In order to control the sulfur content in melt, three kinds of crucibles, namely, alumina, calcia, and calcia crucible with CaO-CaF₂ powders were used. To control oxygen partial pressure, air, argon, and Ar-3%H₂ atmosphere were employed. <br>  After melting in the alumina crucible, only flake graphite was formed in the specimen containing 78～98 massppm S. In the case of calcia crucible, aggregated and chunky graphite were observed in the specimen containing 11～20 massppm S. The specimen melted in a calcia crucible with CaO-CaF₂ powders, showed spheroidal graphite with chunky graphite whose sulfur content was 1 to 11 massppm. <br>  We confirmed that the spheroidal graphite is formed due to the low activity of sulfur and oxygen below critical levels. The critical levels changed with the cooling rate of samples. In addition, the cooling curve of the specimen melted in calcia crucible with CaO-CaF₂ powders did not show constant eutectic temperature, which is generally observed in that of flake graphite cast iron. This result is attributed to the solidification mode of spheroidal graphite, which differs from that of the flake graphite cast iron.

The influence of sulfur content on the matrix structure of cast iron was studied using spheroidal graphite and flake graphite cast iron samples whose chemical composition was 3.6mass%C and 2.0mass%Si. The sulfur content of the flake graphite cast iron samples was changed from 0.006mass% to 0.076mass% with FeS addition. The matrix structure was observed for the as cast and heat treated samples. The heat treatments were annealing (furnace cooled) and normalizing (air cooled), cooled from the different heating temperature, as 850°C, 950°C and 1050°C. <br>  The formation of pearlite structure was accelerated with sulfur in flake graphite cast iron samples. The formation of pearlite in spheroidal graphite cast iron was nearly identical with that of the low sulfur flake samples. When the heating temperature was 850°C, the pearlite morphology was granular and the pearlite formed from the higher temperature and the as cast sample, the morphology was lamellar. The difference should be affected by the solubility of carbon in austenite.

We used an improved sessile drop method, which can measure the equilibrium contact angle of molten Fe-C-S alloys on graphite basal planes using an equilibrated method. We adopted this method for the measurement of the interfacial energy, gamma(s/l), between the basal plane of graphite and the Fe-C melts at 1573 K in a purified He-3%H-2 atmosphere. For the calculation of gamma(s/l), we have to use the true graphite surface energy, gamma(s/v) and the surface tension of the iron melt, gamma(l/v), as the reference. However, we could not get a reliable gamma(s/l) value from the literature, because the gamma(s/v), value could be changed by the contamination of the vapor from the melt. For this reason, we measured the gamma(s/v) value using a nonequilibrated method based on the interfacial energy balance at the trijunction using gamma(l/v) and the contact angle theta and phi, where theta is the contact angle between the liquid metal and graphite substrate observed during the measurement by an optical microscope, and phi is the hidden angle observed at cross section of the sample.
We found that the gamma(s/v), and gamma(s/l) values decrease when the sulfur contents are greater than 10 mass ppm sulfur. From 10 to 120 ppm sulfur, the decreasing rate has a constant value on a logarithmic scale of sulfur activities. The morphology of the graphite changes from spheroidal to flake as gamma(s/l) decreases due to increase in the sulfur content. (C) 2005 Springer Science + Business Media, Inc.

The surface roughness factors, such as the Wenzel roughness factor and so on, are very interrelated to each other. Therefore, it makes a precise discussion difficult on how the surface roughness affects the wettability. We already reported the effect of the surface roughness on the wettability at a constant Wenzel roughness factor using two kinds of models, the hemisphere close packing model and the hemiround rod close lining model. Nevertheless, the pitch is proportional to the height in these models. Therefore, we could not independently discuss the influence of roughness height and roughness pitch on the wettability.
We developed our new models which can independently describe the influence of the surface roughness height and the roughness pitch on the wettability. We simulated loose packing sphere models by periodically placing small ball bearings and the loose lining round rod models by winding fine wires. The wettability was measured by the sessile drop method for the non-wetting system using paraffin coated samples and aqueous solutions. These results show that there is a critical pitch which determines the maximum contact angle in both systems. These results can be explained by the ratio of the solid/liquid/vapor and liquid/vapor line length at the three phase line. (C) 2005 Springer Science + Business Media, Inc.

We used an improved sessile drop method, which can measure the equilibrium contact angle of molten Fe-C-S alloys on graphite basal planes using an equilibrated method. We adopted this method for the measurement of the interfacial energy, gamma(s/l), between the basal plane of graphite and the Fe-C melts at 1573 K in a purified He-3%H-2 atmosphere. For the calculation of gamma(s/l), we have to use the true graphite surface energy, gamma(s/v) and the surface tension of the iron melt, gamma(l/v), as the reference. However, we could not get a reliable gamma(s/l) value from the literature, because the gamma(s/v), value could be changed by the contamination of the vapor from the melt. For this reason, we measured the gamma(s/v) value using a nonequilibrated method based on the interfacial energy balance at the trijunction using gamma(l/v) and the contact angle theta and phi, where theta is the contact angle between the liquid metal and graphite substrate observed during the measurement by an optical microscope, and phi is the hidden angle observed at cross section of the sample.
We found that the gamma(s/v), and gamma(s/l) values decrease when the sulfur contents are greater than 10 mass ppm sulfur. From 10 to 120 ppm sulfur, the decreasing rate has a constant value on a logarithmic scale of sulfur activities. The morphology of the graphite changes from spheroidal to flake as gamma(s/l) decreases due to increase in the sulfur content. (C) 2005 Springer Science + Business Media, Inc.

The surface roughness factors, such as the Wenzel roughness factor and so on, are very interrelated to each other. Therefore, it makes a precise discussion difficult on how the surface roughness affects the wettability. We already reported the effect of the surface roughness on the wettability at a constant Wenzel roughness factor using two kinds of models, the hemisphere close packing model and the hemiround rod close lining model. Nevertheless, the pitch is proportional to the height in these models. Therefore, we could not independently discuss the influence of roughness height and roughness pitch on the wettability.
We developed our new models which can independently describe the influence of the surface roughness height and the roughness pitch on the wettability. We simulated loose packing sphere models by periodically placing small ball bearings and the loose lining round rod models by winding fine wires. The wettability was measured by the sessile drop method for the non-wetting system using paraffin coated samples and aqueous solutions. These results show that there is a critical pitch which determines the maximum contact angle in both systems. These results can be explained by the ratio of the solid/liquid/vapor and liquid/vapor line length at the three phase line. (C) 2005 Springer Science + Business Media, Inc.

&nbsp;&nbsp;To produce partially composed metal matrix composites, namely discontinuous MMCs, a spontaneous infiltration process was examined. The preforms of the interpenetrated composites with SiC particles were infiltrated without pressure using cast iron melt. The preforms were fabricated using furan resin and waterglass as the binder containing CuO, Cu₂O and Fe₂O₃ powder for improving their wettability between the melt and SiC preform. As the infiltration process is controlled by a wetting phenomenon, how to improve this infiltration technique based on reactive wetting was investigated.<br>&nbsp;&nbsp;The preforms were sintered in a Tamman tube at 1173 K before infiltration. Cast iron sample weighing 40 g was placed on the preform and then infiltrated in air at 1673 K for two hours. The furan resin combined with CuO and Cu₂O resulted in complete infiltration without pressure in air at 1673 K. The waterglass combined one was found to be inferior to that of the furan.

&nbsp;&nbsp;Flake graphite cast iron is a high damping metal. Cast iron is used for engine block and brake materials in the automobile industry due to its low cost and high abrasive characteristics. High drivability characteristics of automobiles such as silence and effective braking ability are becoming more and more important recently. Since the anti-brake squeal materials are therefore needed for the brake materials, we studied the influence of graphite morphology and matrix on the damping capacity of hypereutectic cast iron.<br>&nbsp;&nbsp;The results charified that the damping capacity of the cast iron has good correlation with the graphite shape factor K_o, length of periphery of a graphite divided by the square root of the area, and kind of matrix. The damping capacity increases with the K_o value. This means that longer and thinner graphite promotes high damping capacity. Moreover, if the matrix consists of small amounts of martensite and bainite, up to 10%, with pearlite, the damping capacity increases with the fraction of the martensite and bainite.

The effect of viscosity and cooling conditions on the foamability of the molten Al-Si-Mg alloy was investigated using the unidirectional solidification method. The mechanism of pore formation in the Al-Si-Mg alloy has been observed and analyzed theoretically. The results obtained are given as follows: 1. The optimum torque of the melt for foaming is approximately 0.29Nm and is about 1.7 times that of the original melt. 2. The existence of alumina particles in the melt is necessary for nucleation of the bubbles and the increase in viscosity. 3. With the increase in the solidification time, the time for bubble growth increases during the unidirectional solidification. 4. The pore size and the total volume of the porosity increase with the diffusion of hydrogen and is controlled by the local solidification time during cooling of the hydrogen saturated melt. 5. Properly controlled viscosity and solidification of the melt are good for promoting uniformity of the pore structure of the aluminum foams. © 2002 Published by Elsevier Science B.V.

The effect of viscosity and cooling conditions on the foamability of the molten Al-Si-Mg alloy was investigated using the unidirectional solidification method. The mechanism of pore formation in the Al-Si-Mg alloy has been observed and analyzed theoretically. The results obtained are given as follows:
1. The optimum torque of the melt for foaming is approximately 0.29 N m and is about 1.7 times that of the original melt.
2. The existence of alumina particles in the melt is necessary for nucleation of the bubbles and the increase in viscosity.
3. With the increase in the solidification time, the time for bubble growth increases during the unidirectional solidification.
4. The pore size and the total volume of the porosity increase with the diffusion of hydrogen and is controlled by the local solidification time during cooling of the hydrogen saturated melt.
5. Properly controlled viscosity and solidification of the melt are good for promoting uniformity of the pore structure of the aluminum foams. (C) 2002 Published by Elsevier Science B.V.

The effect of viscosity and cooling conditions on the foamability of the molten Al-Si-Mg alloy was investigated using the unidirectional solidification method. The mechanism of pore formation in the Al-Si-Mg alloy has been observed and analyzed theoretically. The results obtained are given as follows:
1. The optimum torque of the melt for foaming is approximately 0.29 N m and is about 1.7 times that of the original melt.
2. The existence of alumina particles in the melt is necessary for nucleation of the bubbles and the increase in viscosity.
3. With the increase in the solidification time, the time for bubble growth increases during the unidirectional solidification.
4. The pore size and the total volume of the porosity increase with the diffusion of hydrogen and is controlled by the local solidification time during cooling of the hydrogen saturated melt.
5. Properly controlled viscosity and solidification of the melt are good for promoting uniformity of the pore structure of the aluminum foams. (C) 2002 Published by Elsevier Science B.V.

&nbsp;&nbsp;The effects of wetting on the transfer of a particle into molten metal were investigated using three theoretical models. The effects of particle diameter and contact angle between the particle and molten metal on the force for particle transfer were also calculated. Using these theoretical models, for any arbitrary value of the three interfacial energies (solid/vapor, solid/liquid and liquid/vapor), contact angle, particle radius, ratio of dipped length, particle and liquid density, the actual force required for the particle transfer can be obtained. Furthermore, these equations were verified using water solutions and paraffin coated steel balls with a 1mm radius in experiments where the contact angle was varied from 0&deg; to 105&deg; by changing the concentration of the surfactant (sodium lauryl sulfate). The surface tension of these aquasolutions were measured by the maximum bubble pressure method.

&nbsp;&nbsp;The effects of wetting on the transfer of a particle into molten metal were investigated using three theoretical models. The effects of particle diameter and contact angle between the particle and molten metal on the force for particle transfer were also calculated. Using these theoretical models, for any arbitrary value of the three interfacial energies (solid/vapor, solid/liquid and liquid/vapor), contact angle, particle radius, ratio of dipped length, particle and liquid density, the actual force required for the particle transfer can be obtained. Furthermore, these equations were verified using water solutions and paraffin coated steel balls with a 1mm radius in experiments where the contact angle was varied from 0&deg; to 105&deg; by changing the concentration of the surfactant (sodium lauryl sulfate). The surface tension of these aquasolutions were measured by the maximum bubble pressure method.

&nbsp;&nbsp;The effects of wetting on the transfer of a particle into molten metal were investigated using three theoretical models. The effects of particle diameter and contact angle between the particle and molten metal on the force for particle transfer were also calculated. Using these theoretical models, for any arbitrary value of the three interfacial energies (solid/vapor, solid/liquid and liquid/vapor), contact angle, particle radius, ratio of dipped length, particle and liquid density, the actual force required for the particle transfer can be obtained. Furthermore, these equations were verified using water solutions and paraffin coated steel balls with a 1mm radius in experiments where the contact angle was varied from 0&deg; to 105&deg; by changing the concentration of the surfactant (sodium lauryl sulfate). The surface tension of these aquasolutions were measured by the maximum bubble pressure method.

Thermal analysis is used to establish the relationship between solidification history and the microstructure of SiC particulate reinforced Al-Si alloy-matrix composites. The results show that cooling curves are influenced by the presence of SiC particles and by strontium modification. The eutectic growth temperature of SiCP/359 composites modified with Sr lies in the range of 840 to 843 K, i.e., about 5 to 7 K higher than that of Sr-modified unreinforced 359. For the same composite, the eutectic undercooling is higher with Sr modification than without. The eutectic solidification time of the composites is shorter than that of the unreinforced base alloy because of the presence of the ceramic particles. Strontium modification has the tendency to extend the eutectic solidification time. Microstructure analysis reveals that Sr modification has a refining effect on eutectic silicon for the composites, and SiC particles in the composite melt serve as the substrates for eutectic Si phase nucleation. (C) 2002 Kluwer Academic Publishers.

Thermal analysis is used to establish the relationship between solidification history and the microstructure of SiC particulate reinforced Al-Si alloy-matrix composites. The results show that cooling curves are influenced by the presence of SiC particles and by strontium modification. The eutectic growth temperature of SiCP/359 composites modified with Sr lies in the range of 840 to 843 K, i.e., about 5 to 7 K higher than that of Sr-modified unreinforced 359. For the same composite, the eutectic undercooling is higher with Sr modification than without. The eutectic solidification time of the composites is shorter than that of the unreinforced base alloy because of the presence of the ceramic particles. Strontium modification has the tendency to extend the eutectic solidification time. Microstructure analysis reveals that Sr modification has a refining effect on eutectic silicon for the composites, and SiC particles in the composite melt serve as the substrates for eutectic Si phase nucleation. (C) 2002 Kluwer Academic Publishers.

Thermal analysis is used to establish the relationship between solidification history and the microstructure of SiC particulate reinforced Al-Si alloy-matrix composites. The results show that cooling curves are influenced by the presence of SiC particles and by strontium modification. The eutectic growth temperature of SiCP/359 composites modified with Sr lies in the range of 840 to 843 K, i.e., about 5 to 7 K higher than that of Sr-modified unreinforced 359. For the same composite, the eutectic undercooling is higher with Sr modification than without. The eutectic solidification time of the composites is shorter than that of the unreinforced base alloy because of the presence of the ceramic particles. Strontium modification has the tendency to extend the eutectic solidification time. Microstructure analysis reveals that Sr modification has a refining effect on eutectic silicon for the composites, and SiC particles in the composite melt serve as the substrates for eutectic Si phase nucleation. (C) 2002 Kluwer Academic Publishers.

Thermal analysis is used to establish the relationship between solidification history and the microstructure of SiC particulate reinforced Al-Si alloy-matrix composites. The results show that cooling curves are influenced by the presence of SiC particles and by strontium modification. The eutectic growth temperature of SiCP/359 composites modified with Sr lies in the range of 840 to 843 K, i.e., about 5 to 7 K higher than that of Sr-modified unreinforced 359. For the same composite, the eutectic undercooling is higher with Sr modification than without. The eutectic solidification time of the composites is shorter than that of the unreinforced base alloy because of the presence of the ceramic particles. Strontium modification has the tendency to extend the eutectic solidification time. Microstructure analysis reveals that Sr modification has a refining effect on eutectic silicon for the composites, and SiC particles in the composite melt serve as the substrates for eutectic Si phase nucleation. (C) 2002 Kluwer Academic Publishers.

&nbsp;&nbsp;The measurement method of macroscopic graphite morphology in flake graphite cast iron was studied using a laser roughness meter on the fractured surface. The fracture of flake graphite cast iron mainly goes through the graphite matrix interface and the inside of the graphite. Assuming that the fracture surface contains considerable information on the graphite morphology, we therefore measured the surface roughness using a laser roughness meter.<br>&nbsp;&nbsp;The experimental results show that the graphite morphology, the ratio of the A-type and D-type graphite areas in the macroscopic range, can be accurately estimated using the laser roughness patterns of the fractured surface in flake graphite cast iron.

&nbsp;&nbsp;To study the effects of sulfur content on the nodule count of spheroidal graphite cast iron, the sulfur content of the base melt was varied from 0.0004 mass % to 0.1mass %. The melt was then treated by Mg-ferrosilicon using the sandwich process to produce spheroidal graphite cast iron. The post-inoculated melt with ferrosilicon was poured into the silica sand molds with 2 mm thickness cavities to make thin castings specimens. The graphite nodule count and chilled structures in the thin sections were studied. The relationship between the nodule count and chill area was clarified and consequently the optimum sulfur content to avoid chill structure was obtained.

&nbsp;&nbsp;The measurement method of macroscopic graphite morphology in flake graphite cast iron was studied using a laser roughness meter on the fractured surface. The fracture of flake graphite cast iron mainly goes through the graphite matrix interface and the inside of the graphite. Assuming that the fracture surface contains considerable information on the graphite morphology, we therefore measured the surface roughness using a laser roughness meter.<br>&nbsp;&nbsp;The experimental results show that the graphite morphology, the ratio of the A-type and D-type graphite areas in the macroscopic range, can be accurately estimated using the laser roughness patterns of the fractured surface in flake graphite cast iron.

&nbsp;&nbsp;To study the effects of sulfur content on the nodule count of spheroidal graphite cast iron, the sulfur content of the base melt was varied from 0.0004 mass % to 0.1mass %. The melt was then treated by Mg-ferrosilicon using the sandwich process to produce spheroidal graphite cast iron. The post-inoculated melt with ferrosilicon was poured into the silica sand molds with 2 mm thickness cavities to make thin castings specimens. The graphite nodule count and chilled structures in the thin sections were studied. The relationship between the nodule count and chill area was clarified and consequently the optimum sulfur content to avoid chill structure was obtained.

&nbsp;&nbsp;The measurement method of macroscopic graphite morphology in flake graphite cast iron was studied using a laser roughness meter on the fractured surface. The fracture of flake graphite cast iron mainly goes through the graphite matrix interface and the inside of the graphite. Assuming that the fracture surface contains considerable information on the graphite morphology, we therefore measured the surface roughness using a laser roughness meter.<br>&nbsp;&nbsp;The experimental results show that the graphite morphology, the ratio of the A-type and D-type graphite areas in the macroscopic range, can be accurately estimated using the laser roughness patterns of the fractured surface in flake graphite cast iron.

This paper describes a new measuring apparatus for the maximum bubble pressure method to confirm the pressure change pattern during bubble formation for wetting and nonwetting systems. Generally speaking, a sawtooth pressure pattern has been widely accepted by many researchers in the maximum bubble pressure method. Nevertheless, when the accumulated gas volume is very small, the pressure change pattern shows a chopping wave pattern as theoretically estimated for both systems. These phenomena were confirmed by the observation of the bubbles during the measurement in an aqueous system, and also confirmed by the identical pressure pattern in a mercury system.

The mechanical properties of flake graphite cast iron were studied using normal cast and unidirectionally solidified (UDS) test pieces. Their chemical composition was eutectic, namely CE = 4.3%, differing only in sulfur content. To eliminate the influence of the discontinuity of graphite at the eutectic cell boundary we used UDS samples. Fully annealed tensile test pieces were used to exclude the influence of the matrix. The experimental results are as follows: The tensile strength and elongation of normal and unidirectionally solidified cast iron are increased by the sulfur addition as the graphite morphology is modified. The tensile strength is improved by UDS. One of the reasons must be based on the continuity of the matrix. Nevertheless, the tensile strength of the UDS samples is not as strong as that of spheroidal graphite cast iron because the fracture occurs at the graphite-terminated area in the aligned graphite samples.

The mechanical properties of flake graphite cast iron were studied using normal cast and unidirectionally solidified (UDS) test pieces. Their chemical composition was eutectic, namely CE = 4.3%, differing only in sulfur content. To eliminate the influence of the discontinuity of graphite at the eutectic cell boundary we used UDS samples. Fully annealed tensile test pieces were used to exclude the influence of the matrix. The experimental results are as follows: The tensile strength and elongation of normal and unidirectionally solidified cast iron are increased by the sulfur addition as the graphite morphology is modified. The tensile strength is improved by UDS. One of the reasons must be based on the continuity of the matrix. Nevertheless, the tensile strength of the UDS samples is not as strong as that of spheroidal graphite cast iron because the fracture occurs at the graphite-terminated area in the aligned graphite samples.

The influence of growth direction on the monotectic structure of the Cu-Pb alloy is studied. In order to examine the influence under a ly environment, both the upward (opposite to the direction of gravity) and downward (the direction of gravity) unidirectional solidifications (UDS) are carried out. In the case of the upward UDS, a banded structure, which consists of Pb-rich and Cu-rich layers, is observed. The L-2 droplets pile up in front of the solid/liquid interface. On the other hand, in the downward UDS, the irregularly shaped L-2 phase uniformly disperses in the specimen and no banded structure is found. The gravity macrosegregation of the L-2 liquid is observed at the bottom of the molten alloy in the downward solidified specimen. This is caused by the difference in the density between the L-1 and L-2 phases. Furthermore, a mechanism for the formation of a banded structure is suggested. This mechanism suggests that the coalesced L-2 phase covers the solid/liquid interface by producing a Pb-rich layer that permits an increase in the undercooling of the L-1/L-2 interface compared to the monotectic temperature. As nucleation of the CL-Cu phase occurs on the Pb-rich layer, the coexisting three phases are then restored. The temperature at the growth front is also returned to the monotectic temperature. The repetition mentioned above will result in the banded structure found in the upward UDS. :(C) 2001 Elsevier Science B.V. All rights reserved.

The influence of growth direction on the monotectic structure of the Cu-Pb alloy is studied. In order to examine the influence under a ly environment, both the upward (opposite to the direction of gravity) and downward (the direction of gravity) unidirectional solidifications (UDS) are carried out. In the case of the upward UDS, a banded structure, which consists of Pb-rich and Cu-rich layers, is observed. The L-2 droplets pile up in front of the solid/liquid interface. On the other hand, in the downward UDS, the irregularly shaped L-2 phase uniformly disperses in the specimen and no banded structure is found. The gravity macrosegregation of the L-2 liquid is observed at the bottom of the molten alloy in the downward solidified specimen. This is caused by the difference in the density between the L-1 and L-2 phases. Furthermore, a mechanism for the formation of a banded structure is suggested. This mechanism suggests that the coalesced L-2 phase covers the solid/liquid interface by producing a Pb-rich layer that permits an increase in the undercooling of the L-1/L-2 interface compared to the monotectic temperature. As nucleation of the CL-Cu phase occurs on the Pb-rich layer, the coexisting three phases are then restored. The temperature at the growth front is also returned to the monotectic temperature. The repetition mentioned above will result in the banded structure found in the upward UDS. :(C) 2001 Elsevier Science B.V. All rights reserved.

In order to solve the problem of particles settling and agglomeration in front of solidifying interface in unidirectional solidification (UDS) experiments, a zone-melted process has been utilized in this study. The experimental results show that, the melting zone could be kept in 30-40 mm width and the zone melted UDS experiments are realized with Al2O3 particle reinforced aluminum-matrix composites. But particle settling still occurs in the liquid, and becomes severe as the particle volume fraction decreases. However, when the volume fraction of the particles is more than 20-22 vol.%, no further settling occurs under a solidification rate of 8-16 mm/h. Investigation on the interaction of particles and solid/liquid interface reveals that the Al2O3 particles are rejected into liquid and pushed by the growing solid phase in Al2O3(P)/Al and Al2O3(P)/Al-0.23wt.%Ce composites. Some particles are mechanically entrapped between cells, and distributed along the crystal grain boundaries. (C) 2001 Kluwer Academic Publishers.

In order to solve the problem of particles settling and agglomeration in front of solidifying interface in unidirectional solidification (UDS) experiments, a zone-melted process has been utilized in this study. The experimental results show that, the melting zone could be kept in 30-40 mm width and the zone melted UDS experiments are realized with Al2O3 particle reinforced aluminum-matrix composites. But particle settling still occurs in the liquid, and becomes severe as the particle volume fraction decreases. However, when the volume fraction of the particles is more than 20-22 vol.%, no further settling occurs under a solidification rate of 8-16 mm/h. Investigation on the interaction of particles and solid/liquid interface reveals that the Al2O3 particles are rejected into liquid and pushed by the growing solid phase in Al2O3(P)/Al and Al2O3(P)/Al-0.23wt.%Ce composites. Some particles are mechanically entrapped between cells, and distributed along the crystal grain boundaries. (C) 2001 Kluwer Academic Publishers.

In order to solve the problem of particles settling and agglomeration in front of solidifying interface in unidirectional solidification (UDS) experiments, a zone-melted process has been utilized in this study. The experimental results show that, the melting zone could be kept in 30-40 mm width and the zone melted UDS experiments are realized with Al2O3 particle reinforced aluminum-matrix composites. But particle settling still occurs in the liquid, and becomes severe as the particle volume fraction decreases. However, when the volume fraction of the particles is more than 20-22 vol.%, no further settling occurs under a solidification rate of 8-16 mm/h. Investigation on the interaction of particles and solid/liquid interface reveals that the Al2O3 particles are rejected into liquid and pushed by the growing solid phase in Al2O3(P)/Al and Al2O3(P)/Al-0.23wt.%Ce composites. Some particles are mechanically entrapped between cells, and distributed along the crystal grain boundaries. (C) 2001 Kluwer Academic Publishers.

In order to solve the problem of particles settling and agglomeration in front of solidifying interface in unidirectional solidification (UDS) experiments, a zone-melted process has been utilized in this study. The experimental results show that, the melting zone could be kept in 30-40 mm width and the zone melted UDS experiments are realized with Al2O3 particle reinforced aluminum-matrix composites. But particle settling still occurs in the liquid, and becomes severe as the particle volume fraction decreases. However, when the volume fraction of the particles is more than 20-22 vol.%, no further settling occurs under a solidification rate of 8-16 mm/h. Investigation on the interaction of particles and solid/liquid interface reveals that the Al2O3 particles are rejected into liquid and pushed by the growing solid phase in Al2O3(P)/Al and Al2O3(P)/Al-0.23wt.%Ce composites. Some particles are mechanically entrapped between cells, and distributed along the crystal grain boundaries. (C) 2001 Kluwer Academic Publishers.

The mechanical properties of flake graphite cast iron were studied using normal cast and unidirectionally solidified (UDS) test pieces. Their chemical composition was cutectic, namely CE = 4.3%, differing only in sulfur content. To eliminate the influence of the discontinuity of graphite at the cutectic cell boundary, we used UDS samples. Fully annealed tensile test pieces were used to exclude the influence of the matrix. The experimental results are as follows: The tensile strength and elongation of normal and unidirectionally solidified cast iron are increased by the sulfur addition as the graphite morphology is modified. The tensile strength is improved by UDS. One of the reasons must be based on the continuity of the matrix. Neverthless, the tensile strength of the UDS samples is not as strong as that of spheroidal graphite cast iron because the fracture occurs at the graphite-terminated area in the aligned graphite samples.

In recent years, there is a high demand for the development of light-weight metals foams. In this paper, the factors which affect the foaming in a foamed aluminum casting process were investigated. The powdered titanium hydride as a foaming agent was added to the molten aluminum alloy in the amount of 0.5 to 2.5% to adjust a proper amount of the foaming agent, and foaming condition was controlled in the temperature range, from 893 to 933 K, to determine the optimum foaming process to obtain a homogeneous aluminum foam. It was found that a properly controlled holding temperature and titanium hydride content of the melt were good for the acquirement of foamed aluminum which contains uniform cell structure of high porosity. The optimum foamed aluminum with air bubbles of 2 to 6 mm diameter and the uniform distribution with 86% porosity was obtained by the addition of 1.0% titanium hydride at 913 K.

In recent years, there is a high demand for the development of light-weight metals foams. In this paper, the factors which affect the foaming in a foamed aluminum casting process were investigated. The powdered titanium hydride as a foaming agent was added to the molten aluminum alloy in the amount of 0.5 to 2.5% to adjust a proper amount of the foaming agent, and foaming condition was controlled in the temperature range, from 893 to 933 K, to determine the optimum foaming process to obtain a homogeneous aluminum foam. It was found that a properly controlled holding temperature and titanium hydride content of the melt were good for the acquirement of foamed aluminum which contains uniform cell structure of high porosity. The optimum foamed aluminum with air bubbles of 2 to 6 mm diameter and the uniform distribution with 86% porosity was obtained by the addition of 1.0% titanium hydride at 913 K.

In recent years, there is a high demand for the development of light-weight metals foams. In this paper, the factors which affect the foaming in a foamed aluminum casting process were investigated. The powdered titanium hydride as a foaming agent was added to the molten aluminum alloy in the amount of 0.5 to 2.5% to adjust a proper amount of the foaming agent, and foaming condition was controlled in the temperature range, from 893 to 933 K, to determine the optimum foaming process to obtain a homogeneous aluminum foam. It was found that a properly controlled holding temperature and titanium hydride content of the melt were good for the acquirement of foamed aluminum which contains uniform cell structure of high porosity. The optimum foamed aluminum with air bubbles of 2 to 6 mm diameter and the uniform distribution with 86% porosity was obtained by the addition of 1.0% titanium hydride at 913 K. (C) 2000 Elsevier Science B.V. All rights reserved.

For the production of partially composed metal matrix composites, namely discontinuous MMCs, using a casting process, the interpenetrated phase composites with SiC particles were fabricated by pressureless infiltration using molten Al-Si alloys. This process is based on a wetting phenomenon, therefore, we discussed the technique based on the reactive wetting.
The SiC particles coated by diluted water glass were sintered in a mullite Tammann tube and simultaneously infiltrated with molten Al-Si alloys in air at 1173 similar to 1573K. As for the improvement of the wettability between the SiC and the liquid aluminum alloy, Fe3O4 powder was mixed with the water glass due to the thermit reaction between Fe3O4 and molten aluminum alloy. In some experiments, pure aluminum powder was placed on the sintered SiCp to increase the interfacial area and the chemical reaction between the sintered SiCp and the melt. By combining these conditions, we succeeded in the complete infiltration without pressure at 1173K.

In 1968 J. N. Reding et al. suggested processes for fabricating composites. According to their patent, decompression atmosphere, which occurs during the reaction between magnesium and oxygen and/or nitrogen in the closed system, causes the spontaneous infiltration of melt into narrow spaces such as ceramic preforms. However, they did not report the proof of this phenomenon or the industrial manufacturing method. Therefore, this research measured the pressure change among SiC particles surrounding molten aluminum. As a result, the pressure reduction during the Mg reaction was identified. Moreover, the existence of Mg,N,, which improves the wettability between Al alloy melt and SiC particles on the interface, was observed on the surface of particles. Applying these two phenomena, a new process for the manufacturing of cast discontinuously reinforced aluminum composites (DRA) was developed. This new method is expected to produce DRA without additional pressure.

The influence of gravity on the microstructure of unidirectionally solidified Cu-Pb monotectic, Ag-Si and Fe-C eutectic alloys is studied. In order to examine the influence of gravity in a Ig environment, both the upward (opposite to direction of gravity) and downward (direction of gravity) unidirectional solidifications are carried out. The results are as follows: (1) For the upward solidified Ag-Si eutectic alloy, the eutectic Si continuously grows. On the other hand, in the downward solidified Ag-Si alloy, the Si grows discontinuously and disperses into the a-Ag matrix. (2) No morphological difference in the eutectic growth is observed between the upward and downward solidified Fe-C eutectic alloys. (3) In the case of the upward solidified Cu-Pb monotectic alloy, a banded structure, which consists of Pb-rich layers and Cu-rich layers, is observed. However, such a banded structure is not observed in the downward specimen. (4) The banded structure is not found in the upward solidified Cu-Pb-0.05 mass%Al and Cu-Pb-0.1 mass%Al alloys. In the Cu-Pb-0.05 mass%Al alloy, a fibrous L-2 phase grows continuously parallel to the growth direction at low growth rate. Only 500 mass ppm addition of aluminum induces the morphological change of the monotectic L-2 phase.

The effect of the foam structure on the compression behavior of an A-Si-Mg alloy foam, which was produced by a liquid metal method was experimentally investigated. The compressive stress-strain behavior is discussed due to the deformation of a cell within the foam. It is found that the foam with an 81.3 % porosity has both a higher energy absorption capacity and a higher efficiency. The compressive strength decrease as the porosity increases. Experimental result showed that the low compressive strength is attributed to cells imperfection due to the higher porosity.

&nbsp;&nbsp;The burn-on of iron castings, namely chemical penetration, has posed as one of the main casting defects for a long time. However, no solution has yet been clarified for this intricate phenomena. In this paper, we discuss how the burn-on is affected with the SiO₂ content of sands from the standpoint of wettability using the sessile drop method. The apparent contact angle between the sand grains and melt decreases with the formation of FeO slag at the interface, and stop the advancing by the dissolution of SiO₂into the slag due to the increase in the melting temperature. The reason why chemical penetration defects can be prevented by the addition of coal dust into mold, is not only due to the decrease the FeO slag formation but also due to the ease of saturation of the slag with SiO₂ to increase the melting temperature.

&nbsp;&nbsp;In recent years, there is a high demand for the development of light weight metals in industry. One of the materials is foamed aluminum alloys. In this research, we studied the foaming mechanism for the development of materials using a solidification process.<br>&nbsp;&nbsp;The viscosity in the melt was increased by the oxidation of the melt to form fine Al₂O₃ particles, and then 1 mass % titanium hydride powder was added to the melt and unidirectionally solidified to make the aluminum alloy foam. Hydrogen bubbles nucleated on the Al₂O₃ particles at the solid/liquid interface and entrapped in the solid. This shows that the Al₂O₃ particles not only affect the increase in viscosity but also serve as nucleating agents for the bubbles.

For the production of partially composed metal matrix composites, namely discontinuous MMCs, using a casting process, the interpenetrated phase composites with SiC particles were fabricated by pressureless infiltration using molten Al-Si alloys. This process is based on a wetting phenomenon, therefore, we discussed the technique based on the reactive wetting.
The SiC particles coated by diluted water glass were sintered in a mullite Tammann tube and simultaneously infiltrated with molten Al-Si alloys in air at 1173 similar to 1573K. As for the improvement of the wettability between the SiC and the liquid aluminum alloy, Fe3O4 powder was mixed with the water glass due to the thermit reaction between Fe3O4 and molten aluminum alloy. In some experiments, pure aluminum powder was placed on the sintered SiCp to increase the interfacial area and the chemical reaction between the sintered SiCp and the melt. By combining these conditions, we succeeded in the complete infiltration without pressure at 1173K.

The influence of gravity on the microstructure of unidirectionally solidified Cu-Pb monotectic, Ag-Si and Fe-C eutectic alloys is studied. In order to examine the influence of gravity in a Ig environment, both the upward (opposite to direction of gravity) and downward (direction of gravity) unidirectional solidifications are carried out. The results are as follows: (1) For the upward solidified Ag-Si eutectic alloy, the eutectic Si continuously grows. On the other hand, in the downward solidified Ag-Si alloy, the Si grows discontinuously and disperses into the a-Ag matrix. (2) No morphological difference in the eutectic growth is observed between the upward and downward solidified Fe-C eutectic alloys. (3) In the case of the upward solidified Cu-Pb monotectic alloy, a banded structure, which consists of Pb-rich layers and Cu-rich layers, is observed. However, such a banded structure is not observed in the downward specimen. (4) The banded structure is not found in the upward solidified Cu-Pb-0.05 mass%Al and Cu-Pb-0.1 mass%Al alloys. In the Cu-Pb-0.05 mass%Al alloy, a fibrous L-2 phase grows continuously parallel to the growth direction at low growth rate. Only 500 mass ppm addition of aluminum induces the morphological change of the monotectic L-2 phase.

For the production of partially composed metal matrix composites, namely discontinuous MMCs, using a casting process, the interpenetrated phase composites with SiC particles were fabricated by pressureless infiltration using molten Al-Si alloys. This process is based on a wetting phenomenon, therefore, we discussed the technique based on the reactive wetting.
The SiC particles coated by diluted water glass were sintered in a mullite Tammann tube and simultaneously infiltrated with molten Al-Si alloys in air at 1173 similar to 1573K. As for the improvement of the wettability between the SiC and the liquid aluminum alloy, Fe3O4 powder was mixed with the water glass due to the thermit reaction between Fe3O4 and molten aluminum alloy. In some experiments, pure aluminum powder was placed on the sintered SiCp to increase the interfacial area and the chemical reaction between the sintered SiCp and the melt. By combining these conditions, we succeeded in the complete infiltration without pressure at 1173K.

In 1968 J. N. Reding et al. suggested processes for fabricating composites. According to their patent, decompression atmosphere, which occurs during the reaction between magnesium and oxygen and/or nitrogen in the closed system, causes the spontaneous infiltration of melt into narrow spaces such as ceramic preforms. However, they did not report the proof of this phenomenon or the industrial manufacturing method. Therefore, this research measured the pressure change among SiC particles surrounding molten aluminum. As a result, the pressure reduction during the Mg reaction was identified. Moreover, the existence of Mg,N,, which improves the wettability between Al alloy melt and SiC particles on the interface, was observed on the surface of particles. Applying these two phenomena, a new process for the manufacturing of cast discontinuously reinforced aluminum composites (DRA) was developed. This new method is expected to produce DRA without additional pressure.

The influence of gravity on the microstructure of unidirectionally solidified Cu-Pb monotectic, Ag-Si and Fe-C eutectic alloys is studied. In order to examine the influence of gravity in a Ig environment, both the upward (opposite to direction of gravity) and downward (direction of gravity) unidirectional solidifications are carried out. The results are as follows: (1) For the upward solidified Ag-Si eutectic alloy, the eutectic Si continuously grows. On the other hand, in the downward solidified Ag-Si alloy, the Si grows discontinuously and disperses into the a-Ag matrix. (2) No morphological difference in the eutectic growth is observed between the upward and downward solidified Fe-C eutectic alloys. (3) In the case of the upward solidified Cu-Pb monotectic alloy, a banded structure, which consists of Pb-rich layers and Cu-rich layers, is observed. However, such a banded structure is not observed in the downward specimen. (4) The banded structure is not found in the upward solidified Cu-Pb-0.05 mass%Al and Cu-Pb-0.1 mass%Al alloys. In the Cu-Pb-0.05 mass%Al alloy, a fibrous L-2 phase grows continuously parallel to the growth direction at low growth rate. Only 500 mass ppm addition of aluminum induces the morphological change of the monotectic L-2 phase.

&nbsp;&nbsp;The burn-on of iron castings, namely chemical penetration, has posed as one of the main casting defects for a long time. However, no solution has yet been clarified for this intricate phenomena. In this paper, we discuss how the burn-on is affected with the SiO₂ content of sands from the standpoint of wettability using the sessile drop method. The apparent contact angle between the sand grains and melt decreases with the formation of FeO slag at the interface, and stop the advancing by the dissolution of SiO₂into the slag due to the increase in the melting temperature. The reason why chemical penetration defects can be prevented by the addition of coal dust into mold, is not only due to the decrease the FeO slag formation but also due to the ease of saturation of the slag with SiO₂ to increase the melting temperature.

&nbsp;&nbsp;In recent years, there is a high demand for the development of light weight metals in industry. One of the materials is foamed aluminum alloys. In this research, we studied the foaming mechanism for the development of materials using a solidification process.<br>&nbsp;&nbsp;The viscosity in the melt was increased by the oxidation of the melt to form fine Al₂O₃ particles, and then 1 mass % titanium hydride powder was added to the melt and unidirectionally solidified to make the aluminum alloy foam. Hydrogen bubbles nucleated on the Al₂O₃ particles at the solid/liquid interface and entrapped in the solid. This shows that the Al₂O₃ particles not only affect the increase in viscosity but also serve as nucleating agents for the bubbles.

The effect of the foam structure on the compression behavior an A-Si-Mg alloy foam, which was produced by a liquid metal method was experimentally investigated. The compressive stress-strain behavior is discussed due to the deformation of a cell within the foam. It is found that the foam with an 81.3% porosity has both a higher energy absorption capacity and a higher efficiency. The compressive strength decrease as the porosity increases. Experimental result showed that the low compressive strength is attributed to cells imperfection due to the higher porosity.

For the production of partially composed metal matrix composites, namely discontinuous MMCs, using a casting process, the interpenetrated phase composites with SiC particles were fabricated by pressureless infiltration using molten Al-Si alloys. This process is based on a wetting phenomenon, therefore, we discussed the technique based on the reactive wetting.
The SiC particles coated by diluted water glass were sintered in a mullite Tammann tube and simultaneously infiltrated with molten Al-Si alloys in air at 1173 similar to 1573K. As for the improvement of the wettability between the SiC and the liquid aluminum alloy, Fe3O4 powder was mixed with the water glass due to the thermit reaction between Fe3O4 and molten aluminum alloy. In some experiments, pure aluminum powder was placed on the sintered SiCp to increase the interfacial area and the chemical reaction between the sintered SiCp and the melt. By combining these conditions, we succeeded in the complete infiltration without pressure at 1173K.

The influence of gravity on the microstructure of unidirectionally solidified Cu-Pb monotectic, Ag-Si and Fe-C eutectic alloys is studied. In order to examine the influence of gravity in a Ig environment, both the upward (opposite to direction of gravity) and downward (direction of gravity) unidirectional solidifications are carried out. The results are as follows: (1) For the upward solidified Ag-Si eutectic alloy, the eutectic Si continuously grows. On the other hand, in the downward solidified Ag-Si alloy, the Si grows discontinuously and disperses into the a-Ag matrix. (2) No morphological difference in the eutectic growth is observed between the upward and downward solidified Fe-C eutectic alloys. (3) In the case of the upward solidified Cu-Pb monotectic alloy, a banded structure, which consists of Pb-rich layers and Cu-rich layers, is observed. However, such a banded structure is not observed in the downward specimen. (4) The banded structure is not found in the upward solidified Cu-Pb-0.05 mass%Al and Cu-Pb-0.1 mass%Al alloys. In the Cu-Pb-0.05 mass%Al alloy, a fibrous L-2 phase grows continuously parallel to the growth direction at low growth rate. Only 500 mass ppm addition of aluminum induces the morphological change of the monotectic L-2 phase.

The effect of the foam structure on the compression behavior an A-Si-Mg alloy foam, which was produced by a liquid metal method was experimentally investigated. The compressive stress-strain behavior is discussed due to the deformation of a cell within the foam. It is found that the foam with an 81.3% porosity has both a higher energy absorption capacity and a higher efficiency. The compressive strength decrease as the porosity increases. Experimental result showed that the low compressive strength is attributed to cells imperfection due to the higher porosity.

Competitive phase selection of undercooled melts between equilibrium ferrite and metastable austenite has been investigated as a function of undercooling. Stainless steel type 316 was undercooled up to 250 K using an electromagnetic levitation method. The microstructure showed different morphologies depending on the undercooling due to different solid phase transformation mechanisms. However, metastable austenite was not formed during the solidification for the undercooling up to 250 K due to the favorable nucleation kinetics of ferrite. The control of the phase selection has also been attempted using an external nucleation seed. Undercooled melts were touched by Fe-50 at.% Ni powders in the levitation coil, whose lattice constant is almost the same as that of metastable austenite. The microstructure showed a dramatic change in the solidification mode from equilibrium ferrite to metastable austenite during the first stage of the solidification. (C) 1999 Acta Metallurgica Inc. Published by Elsevier Science Ltd. All rights reserved.

Wetting phenomenon has been discussed on the basis of the energy balance among solids, liquids and vapors. In these discussions, many researchers have dealt with the surface energy value as an intrinsic parameter. But the extrinsic one, which is changed by the vapor of the liquid, should be used. In this paper we discussed the change in the solid surface energy of solid-Si in the Si/Al-Si systems for equilibrium wetting and nonequilibrium wetting using the sessile drop method. The final objective of this research is the solution of the modification mechanism of Si in Al-Si alloys. The experimental results are as follows;
1. The surface free energy of Si decreases much due to the Al vapor, but the decrease is less at the (111) face than at the other faces.
2. The decreasing rate in Si/Al-Si-Sr systems is much higher than in the Si/Al-Si systems.
3. These phenomena can be recognized by the change in the contact angles with time for the Si/Al-Si and Al-Si-Sr equilibrium systems.

Competitive phase selection of undercooled melts between equilibrium ferrite and metastable austenite has been investigated as a function of undercooling. Stainless steel type 316 was undercooled up to 250 K using an electromagnetic levitation method. The microstructure showed different morphologies depending on the undercooling due to different solid phase transformation mechanisms. However, metastable austenite was not formed during the solidification for the undercooling up to 250 K due to the favorable nucleation kinetics of ferrite. The control of the phase selection has also been attempted using an external nucleation seed. Undercooled melts were touched by Fe-50 at.% Ni powders in the levitation coil, whose lattice constant is almost the same as that of metastable austenite. The microstructure showed a dramatic change in the solidification mode from equilibrium ferrite to metastable austenite during the first stage of the solidification. (C) 1999 Acta Metallurgica Inc. Published by Elsevier Science Ltd. All rights reserved.

Competitive phase selection of undercooled melts between equilibrium ferrite and metastable austenite has been investigated as a function of undercooling. Stainless steel type 316 was undercooled up to 250 K using an electromagnetic levitation method. The microstructure showed different morphologies depending on the undercooling due to different solid phase transformation mechanisms. However, metastable austenite was not formed during the solidification for the undercooling up to 250 K due to the favorable nucleation kinetics of ferrite. The control of the phase selection has also been attempted using an external nucleation seed. Undercooled melts were touched by Fe-50 at.% Ni powders in the levitation coil, whose lattice constant is almost the same as that of metastable austenite. The microstructure showed a dramatic change in the solidification mode from equilibrium ferrite to metastable austenite during the first stage of the solidification. (C) 1999 Acta Metallurgica Inc. Published by Elsevier Science Ltd. All rights reserved.

Wetting phenomenon has been discussed on the basis of the energy balance among solids, liquids and vapors. In these discussions, many researchers have dealt with the surface energy value as an intrinsic parameter. But the extrinsic one, which is changed by the vapor of the liquid, should be used. In this paper we discussed the change in the solid surface energy of solid-Si in the Si/Al-Si systems for equilibrium wetting and nonequilibrium wetting using the sessile drop method. The final objective of this research is the solution of the modification mechanism of Si in Al-Si alloys. The experimental results are as follows;
1. The surface free energy of Si decreases much due to the Al vapor, but the decrease is less at the (111) face than at the other faces.
2. The decreasing rate in Si/Al-Si-Sr systems is much higher than in the Si/Al-Si systems.
3. These phenomena can be recognized by the change in the contact angles with time for the Si/Al-Si and Al-Si-Sr equilibrium systems.

Competitive phase selection of undercooled melts between equilibrium ferrite and metastable austenite has been investigated as a function of undercooling. Stainless steel type 316 was undercooled up to 250 K using an electromagnetic levitation method. The microstructure showed different morphologies depending on the undercooling due to different solid phase transformation mechanisms. However, metastable austenite was not formed during the solidification for the undercooling up to 250 K due to the favorable nucleation kinetics of ferrite. The control of the phase selection has also been attempted using an external nucleation seed. Undercooled melts were touched by Fe-50 at.% Ni powders in the levitation coil, whose lattice constant is almost the same as that of metastable austenite. The microstructure showed a dramatic change in the solidification mode from equilibrium ferrite to metastable austenite during the first stage of the solidification. (C) 1999 Acta Metallurgica Inc. Published by Elsevier Science Ltd. All rights reserved.

This paper describes a new measuring apparatus for the maximum bubble pressure method to cofirm the pressure change pattern during bubble formation for a wetting system. Generally speaking, a sawtooth pressure pattern has been widely accepted by many researchers. Nevertheless, when the accumulated gas volume is very small, the pressure change pattern has a chopping wave pattern.

In the development of Al-Si composites dispersed with ceramic particles by gravity casting, a uniform particles distribution may be obtained if the particle interfacial energy can be decreased and the particle contact angle and the solid/liquid interface is less than 90°. The relationship between the growth morphology of microstructures and particle distribution was examined for Al-Si/Al2O3 composites by directional solidification technique. A uniform Al2O3 particles distribution was realized in Al-Si-Sr-Ca/Al2O3, wherein the Al2O3 particles were not pushed by the growing solid front, but engulfed into the growing front.

&nbsp;&nbsp;It is well-known that the strength of flake graphite cast iron is controlled by the morphology of the graphite. Studies on the strength of flake graphite cast iron have mainly been discussed in relation to the shape of graphite as a notch effect so far. However, the observation of the graphite shape with a microscope to evaluate only by the notch defect raises certain doubts. We therefore investigated the effect of the graphite tip radius on the mechanical properies of cast iron (Experiment-1). In Experiment-2, using pure Fe-C-Si cast iron (3.4 mass % Cand 1.8 mass % Si) as the parent material, we evaluated the influence of minor elements such as P and S on the microstructure, and tensile properties of as-cast and ferritized samples.

In order to study the influence of gravity on the eutectic growth of Ag-Si and Al-Si alloys, both the downward (direction of gravity) and upward (opposite to direction of gravity) unidirectional solidifications were carried out. In the case of the downward U.D.S., the molten metal is confined between the crucible and solid. This inevitably produces the shrinkage cavity at the solid/liquid interface during solidification. In this study, the zone-melting technique was used to avoid this cavity formation. The results of this study were as follows:
(1) In the case of the downward U.D.S. of the Ag-Si alloy, the eutectic silicon phase grew discontinuously and dispersed in the alpha-Ag phase with a diameter of several micrometers. This is caused by the large difference in specific density between silver and the silicon phase. On the other hand, in the case of the upward U.D.S., the silicon phase grew continuously and its morphology showed the typical faceted/non-faceted eutectic.
(2) The eutectic silicon phase of the Al-Si alloy grew continuously regardless of growth direction. There was no morphological difference between the upward and downward U.D.S. specimens.

For the discussion of the formation of spherical graphite in Fe-C alloys, the solid/liquid interfacial morphology of Fe-C alloys was studied using a self-sealing UDS compared with Al-Si alloys. The influence of sulfur on the Fe-C alloys and strontium on the Al-Si alloys was investigated in order to discuss the change in the morphology of the eutectic graphite and silicon. The experimental results are as follows:
If the low sulfur Fe-C alloy contains cerium, the alloy solidifies as spheroidal graphite in the usual solidification. Nevertheless, the structure of the UDS sample is flake-like and identical with that of the 0.003 mass % S sample.
The graphite of the Fe-C-0,003 mass % S sample at the sn interface is covered with the gamma-Fe, but the graphite morphology is flake-like. The solidification rate becomes more than 4 mm/h, and the melt solidifies as undercooled graphite by the branching of the graphite.
These phenomena are identical with the effect of strontium in Al-Si samples.

In the development of Al-Si composites dispersed with ceramic particles by gravity casting, a uniform particles distribution may be obtained if the particle interfacial energy can be decreased and the particle contact angle and the solid/liquid interface is less than 90°. The relationship between the growth morphology of microstructures and particle distribution was examined for Al-Si/Al2O3 composites by directional solidification technique. A uniform Al2O3 particles distribution was realized in Al-Si-Sr-Ca/Al2O3, wherein the Al2O3 particles were not pushed by the growing solid front, but engulfed into the growing front.

&nbsp;&nbsp;It is well-known that the strength of flake graphite cast iron is controlled by the morphology of the graphite. Studies on the strength of flake graphite cast iron have mainly been discussed in relation to the shape of graphite as a notch effect so far. However, the observation of the graphite shape with a microscope to evaluate only by the notch defect raises certain doubts. We therefore investigated the effect of the graphite tip radius on the mechanical properies of cast iron (Experiment-1). In Experiment-2, using pure Fe-C-Si cast iron (3.4 mass % Cand 1.8 mass % Si) as the parent material, we evaluated the influence of minor elements such as P and S on the microstructure, and tensile properties of as-cast and ferritized samples.

In order to study the influence of gravity on the eutectic growth of Ag-Si and Al-Si alloys, both the downward (direction of gravity) and upward (opposite to direction of gravity) unidirectional solidifications were carried out. In the case of the downward U.D.S., the molten metal is confined between the crucible and solid. This inevitably produces the shrinkage cavity at the solid/liquid interface during solidification. In this study, the zone-melting technique was used to avoid this cavity formation. The results of this study were as follows: (1) In the case of the downward U.D.S. of the Ag-Si alloy, the eutectic silicon phase grew discontinuously and dispersed in the α-Ag phase with a diameter of several micrometers. This is caused by the large difference in specific density between silver and the silicon phase. On the other hand, in the case of the upward U.D.S., the silicon phase grew continuously and its morphology showed the typical faceted/non-faceted eutectic. (2) The eutectic silicon phase of the Al-Si alloy grew continuously regardless of growth direction. There was no morphological difference between the upward and downward U.D.S. specimens.

For the discussion of the formation of spherical graphite in Fe-C alloys, the solid/liquid interfacial morphology of Fe-C alloys was studied using a self-sealing UDS compared with Al-Si alloys. The influence of sulfur on the Fe-C alloys and strontium on the Al-Si alloys was investigated in order to discuss the change in the morphology of the eutectic graphite and silicon. The experimental results are as follows:
If the low sulfur Fe-C alloy contains cerium, the alloy solidifies as spheroidal graphite in the usual solidification. Nevertheless, the structure of the UDS sample is flake-like and identical with that of the 0.003 mass % S sample.
The graphite of the Fe-C-0,003 mass % S sample at the sn interface is covered with the gamma-Fe, but the graphite morphology is flake-like. The solidification rate becomes more than 4 mm/h, and the melt solidifies as undercooled graphite by the branching of the graphite.
These phenomena are identical with the effect of strontium in Al-Si samples.

This paper describes a new measuring apparatus for the maximum bubble pressure method to confirm the pressure change pattern during bubble formation for a wetting system. Generally speaking, a sawtooth pressure pattern has been widely accepted by many researchers. Nevertheless, when the accumulated gas volume is very small, the pressure change pattern has a chopping wave pattern.

Through the analysis of forces acting on the particles at the solid/liquid interface based on the interfacial energies, we propose a new theoretical particle transfer model. This model can predict the behavior of the particles at the interface. The main concept of this model is proved by the mutual wetting behavior among the solid, liquid and particle phases. If the contact angle at a solid/liquid interface and a particle is &lt; 90 degrees, the particle can be engulfed into the solid, and if the contact angle is &gt; 90 degrees, the particle would be pushed. This contact angle is measured by unidirectional solidification (UDS) experiments. But if solidification experiments were carried out for particle dispersed composites in a conventional UDS, the particles completely descended before freezing, therefore, a zone-UDS was used in the experiment. In Al-Sr, Al-Ca and AI-Ce matrix composites, Al2O3 particles are pushed by the solid because the contact angles are &gt; 90 degrees. On the other hand, in Al-12.6wt.%Si-Sr-Al2O3 composites alloyed with Ca, the Al2O3 particles are engulfed into the solid because the contact angle is &lt; 90 degrees. (C) 1998 Elsevier Science S.A. All rights reserved.

The effects of surface roughness on wettability have been studied by many researchers using surface roughness factors such as the Wenzel roughness factor and so on. However, these Factors depend on each other which makes a precise discussion difficult. We adopted two kinds of model, a hemispherical close-packed model and a hemiround-rods close-packed model. Using these models, the changes in height roughness, namely the radius, can be varied without changing the Wenzel roughness factors. This paper analyzed the influence of surface height roughness on wettability using the sessile drop method with nonwetting paraffin!water systems. For the hemispherical close-packed models, the effect of height roughness on wetting can be explained by the change in the radius of curvature, R, of the liquid in trapped air pockets at the solid/liquid interface. When the height roughness, h, is more than 125 mu m, R is constant at 50 mu m, but when h is less than 125 mu m, R is conjectured to be a function of the h, theta and gamma(LV) values. From these discussions, we define the wettable flat surface and the atomistic flat surface. In the case of the hemiround-rods close-packed models, we find that the contact angles, measured from the direction parallel to the rods, resemble the advancing and receding angles of contact angle hysteresis. (C) 1998 Acta Metallurgica Inc.

The effects of surface roughness on wettability have been studied by many researchers using surface roughness factors such as the Wenzel roughness factor and so on. However, these Factors depend on each other which makes a precise discussion difficult. We adopted two kinds of model, a hemispherical close-packed model and a hemiround-rods close-packed model. Using these models, the changes in height roughness, namely the radius, can be varied without changing the Wenzel roughness factors. This paper analyzed the influence of surface height roughness on wettability using the sessile drop method with nonwetting paraffin!water systems. For the hemispherical close-packed models, the effect of height roughness on wetting can be explained by the change in the radius of curvature, R, of the liquid in trapped air pockets at the solid/liquid interface. When the height roughness, h, is more than 125 mu m, R is constant at 50 mu m, but when h is less than 125 mu m, R is conjectured to be a function of the h, theta and gamma(LV) values. From these discussions, we define the wettable flat surface and the atomistic flat surface. In the case of the hemiround-rods close-packed models, we find that the contact angles, measured from the direction parallel to the rods, resemble the advancing and receding angles of contact angle hysteresis. (C) 1998 Acta Metallurgica Inc.

&nbsp;&nbsp;Using a thermal analysis system consisting of three cups (First : Inoculated, Second : Base melt, Third : Tellurium added), the relationship between the thermal analysis curve and chill depth was investigated. The content of the alloying elements, such as C, Si and Cr, varies the graphite and cementite eutectic temperature and the difference between them, &Delta;T_E. It is, therefore, impossible to predict the chill depth correctly only from the cooling curve, or by ignoring &Delta;T_E. However, use of the &Delta;T₁/&Delta;T_E value, which is the dimensionless chilling tendency of melt, enables correct prediction of the chill depth. Regardless of the variation of contents and elements, the relation between the chill depth (D, mm) and &Delta;T₁/&Delta;T_E is as follow : D=-27 (&Delta;T₁/&Delta;T_E) + 26. It can be concluded that thermal analysis system consisting of three cups serves as an efficient method for the rapid prediction of chilling tendency of cast iron melts.

&nbsp;&nbsp;The ferritizing process of high purity cast iron was investigated. Specimens used for the experiments were pure Fe-C-Si, Fe-C-Si-P, Fe-C-Si-S, Fe-C-Si-P-S alloys and commercial grade cast iron. The specimens used for the investigation were packed in a container with electrode graphite and fully annealed during two stages in air or argon atmospheres. The first stage was fixed at 1173K for 3 hours while the second stage, the cooling stage from 1173K to 973K was performed at various times such as 5, 10, 20, 40, 80 hours. The results obtained were as follows : 1) The ferritization of pure alloy is difficult to compare with that of commercial grade cast iron. 2) The pure alloys are ferritized easily in argon atmosphere annealing. 3) The titanium in the commercial grade cast iron combines with the nitrogen in iron to form TiN. As it decreases the effect of nitrogen on pearlitization, the ferritization is accelerated. 4) Due to the low Ti content of pure alloys, nitriding occurs when heat treatment is performed, thus preventing ferritization. 5) The ferritizing speed of the specimens decreases in the following order C. G. , 7P, 7P-4S, B, 4S.

&nbsp;&nbsp;The ferritizing process of high purity cast iron was investigated. Specimens used for the experiments were pure Fe-C-Si, Fe-C-Si-P, Fe-C-Si-S, Fe-C-Si-P-S alloys and commercial grade cast iron. The specimens used for the investigation were packed in a container with electrode graphite and fully annealed during two stages in air or argon atmospheres. The first stage was fixed at 1173K for 3 hours while the second stage, the cooling stage from 1173K to 973K was performed at various times such as 5, 10, 20, 40, 80 hours. The results obtained were as follows : 1) The ferritization of pure alloy is difficult to compare with that of commercial grade cast iron. 2) The pure alloys are ferritized easily in argon atmosphere annealing. 3) The titanium in the commercial grade cast iron combines with the nitrogen in iron to form TiN. As it decreases the effect of nitrogen on pearlitization, the ferritization is accelerated. 4) Due to the low Ti content of pure alloys, nitriding occurs when heat treatment is performed, thus preventing ferritization. 5) The ferritizing speed of the specimens decreases in the following order C. G. , 7P, 7P-4S, B, 4S.

&nbsp;&nbsp;Using a thermal analysis system consisting of three cups (First : Inoculated, Second : Base melt, Third : Tellurium added), the relationship between the thermal analysis curve and chill depth was investigated. The content of the alloying elements, such as C, Si and Cr, varies the graphite and cementite eutectic temperature and the difference between them, &Delta;T_E. It is, therefore, impossible to predict the chill depth correctly only from the cooling curve, or by ignoring &Delta;T_E. However, use of the &Delta;T₁/&Delta;T_E value, which is the dimensionless chilling tendency of melt, enables correct prediction of the chill depth. Regardless of the variation of contents and elements, the relation between the chill depth (D, mm) and &Delta;T₁/&Delta;T_E is as follow : D=-27 (&Delta;T₁/&Delta;T_E) + 26. It can be concluded that thermal analysis system consisting of three cups serves as an efficient method for the rapid prediction of chilling tendency of cast iron melts.

&nbsp;&nbsp;Using a thermal analysis system consisting of three cups (First : Inoculated, Second : Base melt, Third : Tellurium added), the relationship between the thermal analysis curve and chill depth was investigated. The content of the alloying elements, such as C, Si and Cr, varies the graphite and cementite eutectic temperature and the difference between them, &Delta;T_E. It is, therefore, impossible to predict the chill depth correctly only from the cooling curve, or by ignoring &Delta;T_E. However, use of the &Delta;T₁/&Delta;T_E value, which is the dimensionless chilling tendency of melt, enables correct prediction of the chill depth. Regardless of the variation of contents and elements, the relation between the chill depth (D, mm) and &Delta;T₁/&Delta;T_E is as follow : D=-27 (&Delta;T₁/&Delta;T_E) + 26. It can be concluded that thermal analysis system consisting of three cups serves as an efficient method for the rapid prediction of chilling tendency of cast iron melts.

&nbsp;&nbsp;The ferritizing process of high purity cast iron was investigated. Specimens used for the experiments were pure Fe-C-Si, Fe-C-Si-P, Fe-C-Si-S, Fe-C-Si-P-S alloys and commercial grade cast iron. The specimens used for the investigation were packed in a container with electrode graphite and fully annealed during two stages in air or argon atmospheres. The first stage was fixed at 1173K for 3 hours while the second stage, the cooling stage from 1173K to 973K was performed at various times such as 5, 10, 20, 40, 80 hours. The results obtained were as follows : 1) The ferritization of pure alloy is difficult to compare with that of commercial grade cast iron. 2) The pure alloys are ferritized easily in argon atmosphere annealing. 3) The titanium in the commercial grade cast iron combines with the nitrogen in iron to form TiN. As it decreases the effect of nitrogen on pearlitization, the ferritization is accelerated. 4) Due to the low Ti content of pure alloys, nitriding occurs when heat treatment is performed, thus preventing ferritization. 5) The ferritizing speed of the specimens decreases in the following order C. G. , 7P, 7P-4S, B, 4S.

Utilising a combination of infiltration and combustion synthesis, a new method for fabricating a high melting point metallic compound (Ni3Al) with only a few pores and little change in the shape, al a low temperature and at atmospheric pressure, has been established. The main procedure for this method is the dropping of molten aluminium or Al-Ni alloys (Ni content up to 18 mass%) onto a sintered nickel sample with 34-36 vol.% pores at 1073 K. Immediately after a molten aluminium or an Al-Ni alloy is dropped onto a sintered nickel sample, the molten metal infiltrates into the sintered material accompanied by an exothermic reaction. When a molten Al-Ni alloy (with a nickel content of over 10 mass%) is dropped, a Ni3Al material is fabricated with only a few pores and little change in the shape. The addition of nickel to the drop delays the onset of the interfacial reaction though the total heat produced is not very different. The delay in the onset of the interfacial reaction allows the infiltration to finish or proceed to some extent before a certain critical point in the interfacial reaction. As a result, a uniform Ni3Al compound is produced. The experimental data clearly relate to the adiabatic combustion temperatures, which indicates that the heat loss is relatively low. As a result, a high melting point metallic compound can be produced at a low temperature and at atmospheric pressure. (C) 1997 Elsevier Science S.A.

Utilising a combination of infiltration and combustion synthesis, a new method for fabricating a high melting point metallic compound (Ni3Al) with only a few pores and little change in the shape, al a low temperature and at atmospheric pressure, has been established. The main procedure for this method is the dropping of molten aluminium or Al-Ni alloys (Ni content up to 18 mass%) onto a sintered nickel sample with 34-36 vol.% pores at 1073 K. Immediately after a molten aluminium or an Al-Ni alloy is dropped onto a sintered nickel sample, the molten metal infiltrates into the sintered material accompanied by an exothermic reaction. When a molten Al-Ni alloy (with a nickel content of over 10 mass%) is dropped, a Ni3Al material is fabricated with only a few pores and little change in the shape. The addition of nickel to the drop delays the onset of the interfacial reaction though the total heat produced is not very different. The delay in the onset of the interfacial reaction allows the infiltration to finish or proceed to some extent before a certain critical point in the interfacial reaction. As a result, a uniform Ni3Al compound is produced. The experimental data clearly relate to the adiabatic combustion temperatures, which indicates that the heat loss is relatively low. As a result, a high melting point metallic compound can be produced at a low temperature and at atmospheric pressure. (C) 1997 Elsevier Science S.A.

&nbsp;&nbsp;Unidirectional solidification experiments have been carried out in Al/Al₂O₃ composites by using the zone-melting process. The experimental results show as follows. When the volume fraction of Al₂O₃ particles is 5&sim;20vol %, particles can be distributed uniformly in the solidified samples. But particle settling still occurs in the liquid, and becomes severer as the particle volume fraction decreases. However, when the volume fraction of the particles is more than 20&sim;22 vol %, no further settling occurs under a solidification velocity of 8&sim;16 mm/h. The solidification rate has a very small effect on the settling height with the same particle contents. In order to investegate the interaction of solid/liquid interface and particles, a main solidification experiment was carried out using the samples made in the above mentioned experiment. The Al₂O₃ particles are pushed by the growing solid phase under a solidification velocity of 2mm/h. The particles are mechanically entrapped between cells, and distributed along the crystal grain boundaries.

&nbsp;&nbsp;Unidirectional solidification experiments have been carried out in Al/Al₂O₃ composites by using the zone-melting process. The experimental results show as follows. When the volume fraction of Al₂O₃ particles is 5&sim;20vol %, particles can be distributed uniformly in the solidified samples. But particle settling still occurs in the liquid, and becomes severer as the particle volume fraction decreases. However, when the volume fraction of the particles is more than 20&sim;22 vol %, no further settling occurs under a solidification velocity of 8&sim;16 mm/h. The solidification rate has a very small effect on the settling height with the same particle contents. In order to investegate the interaction of solid/liquid interface and particles, a main solidification experiment was carried out using the samples made in the above mentioned experiment. The Al₂O₃ particles are pushed by the growing solid phase under a solidification velocity of 2mm/h. The particles are mechanically entrapped between cells, and distributed along the crystal grain boundaries.

The microstructure of flake graphite cast iron is the most important factor in determining the mechanical properties. The microstructure consists of two principal parts : the graphite and the matrix. Therefore, the present paper discusses the influence of the matrix and graphite morphology on mechanical properties. Graphite morphology was varied based on inoculation and holding of the melt and the addition of Ti. The influence of matrix on the tensile properties of cast iron were investigated using kinds of heat treatments. The experimental results were as follows : (1) The influence of graphite morphology on tensile strength of cast iron can be explained by the Rule of Mixture. (2) In the case of the difference of elongation between A-type graphite samples and D-type graphite samples is more than 0.4%, we can use the Rule of Mixtures. But when the difference is less than 0.2%, can not be used the Rule.

&nbsp;&nbsp;Unidirectional solidification experiments have been carried out in Al/Al₂O₃ composites by using the zone-melting process. The experimental results show as follows. When the volume fraction of Al₂O₃ particles is 5&sim;20vol %, particles can be distributed uniformly in the solidified samples. But particle settling still occurs in the liquid, and becomes severer as the particle volume fraction decreases. However, when the volume fraction of the particles is more than 20&sim;22 vol %, no further settling occurs under a solidification velocity of 8&sim;16 mm/h. The solidification rate has a very small effect on the settling height with the same particle contents. In order to investegate the interaction of solid/liquid interface and particles, a main solidification experiment was carried out using the samples made in the above mentioned experiment. The Al₂O₃ particles are pushed by the growing solid phase under a solidification velocity of 2mm/h. The particles are mechanically entrapped between cells, and distributed along the crystal grain boundaries.

Undercooling and solidification phenomena of Cu and Cu-O and Cu-S alloys were investigated. The ''one-way nucleation'' of Cu/Cu2S eutectic was also examined. The undercooling of Cu-O and Cu-S alloys increased with increasing oxygen or sulfur content. This may be caused by oxidation or sulfurizing of effective nucleants in molten copper, or by the surfactant roll of oxygen and sulfur. Regarding the ''one-way nucleation'', the Cu phase nucleates the Cu2S phase at small undercooling, but the Cu2S phase does not nucleate Cu.

Undercooling and solidification phenomena of Cu and Cu-O and Cu-S alloys were investigated. The ''one-way nucleation'' of Cu/Cu2S eutectic was also examined. The undercooling of Cu-O and Cu-S alloys increased with increasing oxygen or sulfur content. This may be caused by oxidation or sulfurizing of effective nucleants in molten copper, or by the surfactant roll of oxygen and sulfur. Regarding the ''one-way nucleation'', the Cu phase nucleates the Cu2S phase at small undercooling, but the Cu2S phase does not nucleate Cu.

The change in the contact angle between two kinds of MgO [sintered sample and single crystal (100) sample] and molten aluminium over time was measured at 1373 K. The atmosphere was a purified He-3vol%H-2 where the formation of the alumina is prevented, and the pressure of the atmosphere was 1.2 arm (1.2 x 10(5) Pa) to minimise the vaporisation of aluminium. Using a logarithmic time scale, three distinct wetting phases (I, II, III) were easily observed in both systems, as is typical in reactive systems. Phase IV (equilibrium-wetting phase), however, was not observed in the present systems even when the experiment was conducted over a long time. Instead, when investigating the change in interfacial area (interfacial diameter) with the change in contact angle, four detailed stages were observed in phase III (interfacial-reaction-wetting phase). These stages in phase III were caused by a decrease in the volume of the molten aluminium due to the interfacial reaction and the vaporisation of the drop. The progress through the stages represented a transition in contact angle from an advancing contact angle to a receding contact angle. In systems where the decrease in volume is too great, the equilibrium contact angle cannot be obtained directly. Here, a method is proposed for the estimation of the equilibrium contact angle in such a system. Copyright (C) 1996 Acta Metallurgica Inc.

A series of melting and solidification experiments were carried out using high-purity SiO2 and Al2O3 powders as the starting materials. The effect of melting temperature on the morphology of the primary mullite was investigated. The following conclusions were made:
1) The morphology of the primary mullite went through five stages as the melting temperature increased; massive, fibrous, faceted dendritic, feathery and fine feathery structures. These phenomena can be explained based on the number of effective nuclei in the melt.
2) The fibrous mullite grew from mullite precursors formed in the melt. Under highly supercooled conditions, the fibrous morphology becomes unstable and begins to branch and form a faceted dendritic crystal.
3) When the number of nuclei is extremely small, the mullite nucleated on the rhombic dodecahedron Mo originating from the molybdenum crucible and crystallizes out into a feathered structure.

A series of melting and solidification experiments were carried out using high-purity SiO2 and Al2O3 powders as the starting materials. The effect of melting temperature on the morphology of the primary mullite was investigated. The following conclusions were made:
1) The morphology of the primary mullite went through five stages as the melting temperature increased; massive, fibrous, faceted dendritic, feathery and fine feathery structures. These phenomena can be explained based on the number of effective nuclei in the melt.
2) The fibrous mullite grew from mullite precursors formed in the melt. Under highly supercooled conditions, the fibrous morphology becomes unstable and begins to branch and form a faceted dendritic crystal.
3) When the number of nuclei is extremely small, the mullite nucleated on the rhombic dodecahedron Mo originating from the molybdenum crucible and crystallizes out into a feathered structure.

The morphological transition of rods to lamellae with increasing growth rate in a Ni-W eutectic alloy was investigated. The eutectic alloy was directionally solidified in a Bridgman-type apparatus at the eight steps of growth rates from 1.0 to 16 mm/h.
The change in morphology with increasing growth rate appeared in the following two different ways of lamella formation:
(1) Under the conditon of the planar solid/liquid interface, the eutectic grain was uniformly comprised of rods or lamellae (blades). In this case, the deviation of the gamma-phase [100] direction from the growth direction controlled the morphology.
(2) When cellular interface was formed at relatively higher growth rates, both rods and radial lamellae coexist simultaneously in a cell.
We confirmed that, as a common factor in the above two ways, the driving force of lamellae formation should be influenced by the angle between the gamma-phase [100] direction and the local growth direction defined as being perpendicular to the solid/liquid interface.

The morphological transition of rods to lamellae with increasing growth rate in a Ni-W eutectic alloy was investigated. The eutectic alloy was directionally solidified in a Bridgman-type apparatus at the eight steps of growth rates from 1.0 to 16 mm/h.
The change in morphology with increasing growth rate appeared in the following two different ways of lamella formation:
(1) Under the conditon of the planar solid/liquid interface, the eutectic grain was uniformly comprised of rods or lamellae (blades). In this case, the deviation of the gamma-phase [100] direction from the growth direction controlled the morphology.
(2) When cellular interface was formed at relatively higher growth rates, both rods and radial lamellae coexist simultaneously in a cell.
We confirmed that, as a common factor in the above two ways, the driving force of lamellae formation should be influenced by the angle between the gamma-phase [100] direction and the local growth direction defined as being perpendicular to the solid/liquid interface.

The effect of gravity on the contact angle was calculated using a precise drop shape model. The drop shape, whose surface curvatures change under the influence of gravity, was calculated following Laplace's equation. The total free energy of the three interfacial free energies and the potential energy of the drop were calculated
the contact angle where the total free energy is minimal at each drop weight was defined as the equilibrium contact angle. These calculations indicate that gravity has no effect on the equilibrium contact angle. © 1995 Taylor &amp
Francis Group, LLC.