The four year project on “Platform Science and Technology for Advanced Magnesium Alloys” started in September 1999 with financial support from the Ministry of Education, Science, Sports and Culture of Japan. The project members consisting mainly of Professors in Japanese Universities are going to make various efforts for the advancement of studies on magnesium alloys. The background and concept of the research project is described, and then the contents and practical activities of the group are briefly explained.

Among the light metal alloys, magnesium is the lightest structural material with a density of 1.74 g/cm³ and many attractive physical and mechanical properties combined with processing advantages. Therefore, it represents a very attractive material for a large amount of applications from main user automotive industry up to other industry fields like sport, robotic and electronic industry. But the usage of magnesium alloys in more complex applications is limited by insufficient properties regarding the ductility, corrosion and creep resistance. Additionally, the high reactivity of magnesium alloys leads to an increased tendency of contamination. In this regard, the paper deals with some general aspects of magnesium alloy development and casting as well as some other production technologies for magnesium alloys with improved properties and for several applications. An emphasis is put on new magnesium-lithium alloy systems with lower density, improved ductility and corrosion resistance. The basic component of metallurgical processing is a magnesium-suitable furnace system which is designed modularly with universal, adaptable components and automatic pressure metering technology. It is used in a wide range of casting processes like chill casting, cold chamber die casting, continuous casting and break-mould casting. Further applications are methods of rapid solidification technology. In the powder metallurgy and spray casting a microstructure with fine and homogeneous phase distribution is generated. Based on the represented processing technology new magnesium alloy systems with lower density, improved ductility and corrosion resistance will be shown. A special peculiarity of magnesium alloys is their application in the medicine sector where they are used as implant materials for surgery.

Lattice constants, bulk modulus and heat of solution of a Mg₇Li₁ alloy are calculated by using an ab initio pseudopotential method based on the density function theory and the supercell approach. Lattice constants and bulk modulus at zero pressure are obtained by using the stress calculation and generalized Hooke’s law. For exchange and correlation method, the Wigner formula are used and the results are compared with ones using the Perdew and Zunger formula. Calculated lattice constants and bulk modulus are in good agreements with experimental values. We also calculate three kinds of ordered structures on this alloy in order to examine the influence of the structure. The results of three structures are different, respectively, and the influence is discussed from the atomic and electronic structure. The valence charge distribution is decreased around Li atoms. Especially, this region of the low charge distribution spreads in c direction.

Nanocrystalline magnesium alloys having high tensile strength, high elevated-temperature tensile strength, high-strain-rate superplasticity and high thermal stability have been developed in Mg₉₇Zn₁Y₂ (at%) alloy by rapidly solidified powder metallurgy (RS P/M) processing. The tensile yield strength and elongation that were dependent on the consolidation temperature were in the ranges of 480 to 610 MPa and 5 to 16%, respectively. Young’s modulus of the RS P/M alloy was 45 GPa. The specific tensile yield strength was four times as high as that of a commercial AZ91-T6 alloy, and was higher than those of conventional titanium (Ti–6Al–4V) and aluminum (7075-T6) alloys. The RS P/M alloys exhibited excellent elevated-temperature yield strength that was 510 MPa at 423 K . The RS P/M alloy also exhibited high-strain-rate superplasticity at a wide strain-rate range from 1×10⁻² to 1×10⁰ s⁻¹ and at a low temperature of 623 K . It is expected that the Mg₉₇Zn₁Y₂ RS P/M alloy can be applied in some fields that requires simultaneously the high specific strength at ambient and elevated temperatures and high workability.

Magnesium alloys are generally brittle owing to their HCP structure. In this study, improvement of tensile mechanical properties under dynamic loading has been demonstrated for a pure magnesium and a ZK60 magnesium alloy. The solution-treated ZK60 alloy exhibits yielding at a dynamic strain rate of 1.8×10³ s⁻¹, which was not observed in a pure magnesium. The yield stress of the ZK60 alloy increases at the dynamic strain rate with a similar slope of Hall-Petch relation at a quasi-static strain rate. Enhancement of ductility can be also achieved by refining grain structures for the ZK60 alloy. The high ductility of the fine-grained alloy is due to the absence of macroscopic cracking at mechanical twin boundaries. It is found that the absorption energy per weight in the fine-grained ZK60 is twice higher than that of high strength aluminum alloys.

The tensile properties at room temperature ∼ 573 K of Mg–9Al–1Zn processed by normal extrusion were compared with those processed by equal channel angular (ECA) extrusion. The strength at room temperature was strongly affected by the grain size. However, the effect of the grain size rapidly decreased with testing temperature. At room temperature, the normal extruded alloy showed the stronger grain size dependence of 0.2% proof stress than the ECA extruded alloy because of the microscopic orientation effect. Also, the grain size dependence was reduced when grain boundaries were in a non-equilibrium state.

A ZK60 alloy was subjected to severe plastic deformation by torsion straining under high pressure at ambient temperature. The data of microhardness measurements, X-ray analysis and TEM observation showed that intense plastic straining resulted in formation of nanometer-scale structure characterized by presence of high internal stress fields. The effect of phase composition of the magnesium alloy on the formation of ultrafine grain structure was examined. It was shown that prior aging promoted the formation of fully grained structure during severe plastic deformation. At the same time, the size of new grains formed was found to be smaller in the quenched state of the magnesium alloy than in the aged state. Grain formation during intense plastic straining was interpreted in terms of low temperature dynamic recrystallization. The mechanisms of ultrafine grain structure formation in hcp material during severe plastic deformation and the role of non-basal slip in this process are discussed.

Dynamic recrystallization (DRX) behavior was systematically examined in two commercial Mg–Al–Zn alloys in order to clarify the relationship between deformation conditions and the resulting microstructure. The materials were deformed by upset forging at temperatures ranging from 473 to 673 K at an initial strain rate of 3.3×10⁻² s⁻¹. Grain refinement was observed during deformation. It was found that the dynamically recrystallized grain size decreases with an increasing Zener–Hollomon parameter and/or a decreasing initial grain size. A phenomenological constitutive equation was developed in order to provide a guideline for the control of the grain size of hot deformed AZ61 alloy.

The individual contributions of Gadolinium and Yttrium to age hardening and high temperature strength of magnesium alloys containing both elements are investigated using alloys containing different Gd:Y mole ratios of 1:0, 3:1, 1:1, 1:3 and 0:1 with a constant Y+Gd content of 2.75 mol%. All investigated alloys exhibit remarkable age hardening by precipitation of β^′′ phase with DO₁₉ crystal structure and β^′ phase with bco crystal structure even at aging temperatures higher than 200°C. Both precipitates are observed in peak-aged specimens. The precipitates contributing to age hardening are fine and their amount increases as Gd content increases, and this results in increased peak hardness, tensile strength and 0.2% proof stress but decreased elongation. On the other hand, higher Y content increases the elongation of the alloys but results in decreased strength.

High temperature creep properties of Mg–Gd–Y–Zr alloys have been evaluated quantitatively. Creep test was carried out under a stress range of 50–100 MPa and a temperature range of 250–300°C. Within the limits of the creep test conditions used in this study, the activation energy for creep of investigated alloys is in the range of 160–240 kJ/mol, and the stress exponent is in the range of 3.7–5.2. Accordingly, the creep mechanism of the investigated alloys is considered to be power law creep. Creep resistance of the investigated alloys depends on chemical composition, it improves with increase in gadolinium content, which implies an increase in the quantity of precipitates. The creep resistance of the high gadolinium-containing alloys exceeds that of the existing heat resistant magnesium alloy, WE54A alloy.

Polymer plating of triazine dithiols on magnesium alloys and applications of the plated alloys were investigated, focusing on important factors for improvement of corrosion resistance and adhesion properties of magnesium alloys. Three kinds of triazine dithiols; 6-dihexylamino-1,3,5-triazine-2,4-dithiol monosodium salts (DHN), 6-diallylamino-1,3,5-triazine-2,4-dithiols (DAN), and 6-mercapto-1,3,5-triazine-2 4-dithiol (TTN), were prepared for imparting functions such as corrosion resistance and direct adhesion to magnesium alloys. In polymer plating of DHN, film thickness initially increases with current density, reaches a maximum at 1 A/m², and then decreases. Film thickness initially increases with DHN concentration, reaches a maximum at 8 mol/m³, and then decreases. Polymer plating of DAN yields good polymer film on magnesium in terms of adhesion when 10 mol/m³ NaOH/Na₂CO₃ aqueous solution is used as an electrolytic solution. An 8 mol/m³ DHN aqueous solution was found to impart the best corrosion resistance for reaction resistance and corrosion current density. In direct adhesion of EPDM to polymer-plated magnesium alloys, a film thickness of near 100 nm imparts the highest peel strength. In injection molding adhesion of PPS polymer to polymer-plated magnesium alloys, high tensile shear strength is obtained when mold temperature exceeds 140°C.

The structure and the composition of surface films formed by chemical conversion coating in Dow7 on pure magnesium and magnesium die cast AZ91D have been studied by XRD, XPS, SEM and TEM combined with ultramicrotomy to clarify the growth mechanism. The films are amorphous containing no definite crystallites detectable by XRD . The major constituents of the films are MgF₂ and MgO_x(OH)_y. The content of NaMgF₃ increases very much with increasing finishing time. In addition, small amounts of Cr₂O₃ and NH₄⁺, whose contents also increase with finishing time, are found. In the film formed on AZ91D, a substantial amount of Al, presumably present as spinel (MgAl₂O₄) or as it’s hydroxide, and small amounts of FeO_x(OH)_y and Mn⁴⁺ are found in addition to above film components on pure magnesium. The film surface of AZ91D shows a granular structure with each granule corresponding to a single grain of the substrate. Cylindrical porous cell structure of chemical conversion coating films on magnesium, which is formed by anodic reaction, has been confirmed for the first time by direct TEM observation. Film growth proceeds mainly by the formation of MgF₂ and MgO_x(OH)_y at film/metal interface by anodic reaction and subsequent film dissolution followed by precipitation of NaMgF₃, Cr₂O₃ and NH₄⁺ in the film. The porous film is composed of cell colonies in the size of sub-microns having central mother pores, which are branching into fine pores.

The influence of the nitrogen ion implantation on the corrosion resistance of magnesium has been studied. The nitrogen solid solution and Mg₃N₂ were detected by the grazing incidence X-ray diffraction analysis. TEM observations revealed the morphology of Mg₃N₂ and cavities in magnesium matrix. Anodic polarization measurements showed that implantation with 1×10²¹ N₂⁺⁄m² at 150 keV greatly reduce the current density at all applied voltage. The higher accelerating voltage showed better corrosion resistance. The amount of Mg₃N₂ was increased with increasing dose resulting in the degradation of corrosion resistance. The corrosion resistance of implanted specimen was also degraded by annealing.

Coating on surfaces of magnesium alloy with pure magnesium by applying a retort method to purification and deposition of magnesium has been attempted for improving corrosion resistance. Magnesium with three nine purity was evaporated in a vacuum furnace and deposited on a substrate, AZ31 magnesium alloy, put at a lower temperature zone in the furnace. Elemental conditions for the deposition coating technique have been investigated to obtain homogeneously coated specimens in microstructural and morphological points of view. The coated specimen has shown a superior corrosion resistance compared with uncoated AZ31 alloy.

Corrosion behavior in magnesium coated AZ31 alloy prepared by vapor deposition technique was examined by means of in situ laser microscopic observations in 1% NaCl solution in comparison with those in 3N–Mg, 6N–Mg and uncoated AZ31 alloy. General corrosion and evolution of hydrogen bubbles at stationary points were observed in all the specimens. Filiform corrosion occurred in the AZ31 alloy and in the 3N–Mg, while it did not occur in the 6N–Mg and deposition coated specimen. General corrosion seemed to suppress the evolution of hydrogen bubbles in the 6N–Mg and deposition coated specimen. Superior corrosion resistance in the deposition coated specimen was confirmed, which is comparable to that in the 6N–Mg.

Electrochemical behaviors of Fe, Cr and Ni have been investigated in molten salt systems as a basic study on the electrorefining of Mg metal; The electrochemical behaviors of FeCl₂, NiCl₂ and CrCl₃ were examined by voltammetry, and the electrochemical dissolutions of these metals were also studied. A MgCl₂–CaCl₂–NaCl or MgCl₂–NaCl–KCl melt was used as electrolytic bath. The electrode reactions of these elements were discussed on these results, and it was concluded that these elements could be eliminated from Mg metal easily by electrorefining process in molten salts. Magnesium metal was obtained electrochemically not only above the melting point of Mg metal but also below the melting point. However, Mg electrodeposit with good morphology could not be obtained under any electrolytic condition below the melting point of Mg metal.

To establish the solid-state recycling technology of remarkably inflammable magnesium alloys, Bulk Mechanical Alloying (BMA) process based on the cyclic plastic deformation has been applied to demonstration experiment. Even starting from large machined chips of AZ91D alloys as starting materials, microstructures of original AZ91D can be refined in the solid state via BMA . Much lower energy consumption is necessary for this solid state recycling compared to the conventional casting process. BMA is one of the suitable recycling technologies to be morphology-free from starting materials.

The automotive use of magnesium is currently restricted to low-temperature structural components. Its use in elevated-temperature structural components such as transmission and engine parts requires the development of cost-effective alloys that can meet the elevated-temperature (150–175°C) performance requirements of these components for strength and creep resistance. Rare-earth additions such as Ce Nd, Pr to Mg are known to improve the creep performance. An example of a rare-earth containing magnesium diecasting alloy with good creep resistance at 150°C is the Mg–4Al–2 rare-earth alloy AE42. This paper discusses a new class of magnesium alloys containing alkaline earth elements such as Ca and Sr, which offers excellent creep performance.

Chill-cast Mg–10%Y–2%Nd alloy was internally hydrided and examined for tensile properties up to 773 K by short-time tensile and creep tests at strain-rates of 1.4×10⁻⁹ to 2.78×10⁻⁴ s⁻¹, in comparison with those of the T6-treated alloy. The results obtained were analyzed for the relationship of stress, σ (0.2% yield stress or applied creep stress), vs. strain rate, \\dotε (initial strain rate or minimum creep rate). Generally, the stress exponent, n (=dlog\\dotε⁄dlogσ)-value of the hydrided alloy is larger than that of the T6-treated alloy at the same test temperature. Consequently, the strength of the hydrided alloy becomes higher than that of the T6-treated alloy at strain rates lower than 10⁻⁵ s⁻¹ at and above 723 K . This fact can be attributed to the stable hydride precipitates in the internally hydrided alloy.

A fracture mechanism and forming limit in cylindrical deep-drawing below 473 K was investigated for commercially rolled sheet of magnesium alloy AZ31. A small punch radius, of which the ratio to thickness was 2.5, was adopted intentionally. At room temperature, a brittle fracture under a very small punch load occurred at an early stage on the punch shoulder near the bottom. It was confirmed by stretch bending tests that the fracture was caused by bending under high tension, which deeply concerned with high yield stress and little local elongation of this material. The forming limit was dominated by this mechanism up to 453 K in semi-static drawing.

Rotation-Cylinder method (RCM) as a primary manufacturing process and thixoforming as a secondary forming process are discussed. RCM allows the production of U-shaped laminar melt surface with Rankine vortex, thus significantly reducing particle agglomeration and entrapped slags. Thixoforming is of particular interest as a secondary forming process because of the low fluidity of the composites and the significant wear produced on machining tools and extrusion and forging dies. Some indications about mechanical properties of the composites are also presented.

Bulk mechanical alloying was successfully applied to solid-state synthesis of Mg₂Ni, which is a promising hydrogen storage alloy. Through XRD, TEM and PCT measurement, the synthesized product was characterized to be a single-phase Mg₂Ni, excluding co-synthesized MgNi₂ or residuals of nickel or magnesium as in conventional processing. The synthesized Mg₂Ni has a fine microstructure with the grain size of 10–20 nm. Thermodynamic data for hydride formation are coincident with reference data, so that nano-structuring in the solid-state synthesized, stoichiometric Mg₂Ni has little role to make essential change in hydrogen storage capacity and mechanism. Physical modification by enrichment of nickel via the bulk mechanical alloying enables us to obtain the nickel-enriched, nano-structured Mg₂Ni and to demonstrate the phase transformation of Mg₂NiH₄ takes place from high temperature phase (HT-phase) to low temperature phase (LT-phase) with decreasing the holding temperature. Negative shift of formation entropy of Mg₂NiH₄ from HT-phase to LT-phase is corresponding to the hydrogen ordering, where a part of octahedral vacancy sites in Mg₂Ni structure is only occupied by hydrogen atoms. Formation of LT-phase is also verified by XRD analysis working together with PCT measurement. Positive shift of formation enthalpy of Mg₂NiH₄ from HT-phase to LT-phase drives a significant increase of plateau pressure in the PCT diagram. Non-equilibration of nano-structured Mg₂Ni leads to control of hydride formation process. Use of these non-equilibrium phase materials helps us to understand various unknown properties of metal hydrides.

The influence of the raw powder characteristics and cold compaction on the solid-state synthesis of Mg₂Si from elemental Mg–Si mixture powder has been examined by thermal analysis using DSC thermogram. The Mg₂Si synthesis progresses at the lower temperature range when employing the fine Si powder, because of the increase in the contacting area between the Mg and Si particles. MgO film, covering the Mg powder surface, prevents the Mg₂Si synthesis due to its thermal stability, however SiO₂ surface film is not so effective because of the deoxidization by Mg during heating. Concerning the effect of the compaction pressure, when the hard Si particles are embedded in the Mg powder after compacting the mixture powder, the mechanical breakage of the thermally stable MgO surface film occurs. Therefore, the enlargement of the contacting clean area between the particles causes the progress of the Mg₂Si synthesis at the lower temperature. Based on these results, the bulk mechanically alloyed Mg–Si composite powder, in which the fine Si particles are distributed uniformly in the Mg matrix, shows a decrease of the reaction temperature by 350 K compared to the mixture powder.

The high-pressure synthesis of new hydrides of the Ca–Mg–Ni system and crystal structural and thermal analyses have been conducted. The high-pressure synthesis was carried out at 1073 K for 2 h at 5 GPa using a cubic-anvil-type apparatus. An additional hydrogen source was used to stabilize the hydride phases. A series of (Ca_1−xMg_x)₂NiH_δ was found to form solid solutions in the range of x=0–0.4. The crystal structure of the (Ca_1−xMg_x)₂NiH_δ solid solutions was of the CsCl-type. The lattice constant of the CsCl-type phase decreased from 0.35542(2) nm (x=0) to 0.35478(2) nm (x=0.4) with increasing Mg content. For the samples with a Ni-rich composition, in addition to the CsCl-type phase, a Mg₂Ni₃H_3.4 phase was also present at high pressures. The CsCl-type phases were thermally stable at less than 646 ±2 K, regardless of Mg content.

The protium absorbing/desorbing characteristics of the rapidly-solidified and mechanically-ground Mg₂Ni–Mn alloys with different Manganese content were investigated. High Manganese additions do not result in great changes in plateau pressure of Mg₂Ni phase. But in mechanically-ground specimens of Manganese-added alloys, a new plateau appears during absorption of protium at a pressure higher than the plateau pressure of Mg₂Ni phase. In rapidly-solidified and mechanically-ground specimens of Manganese-added alloys, protium content at low temperature increases. By exposing specimens for many hours in argon atmosphere at room temperature after hydriding at 250°C, the desorbed protium content increases and the protium desorption temperature decreases. This implies that the amount of microtwinning decreases with increasing elapsed time, and then protium desorbing temperature changes to lower values. Formed MgH₂ phase in high Manganese-added alloys desorbs protium at low temperature probably due to contractive strain generated at the interface between the MgH₂ phase and intermingled Mg₂NiH₄ phase as the later desorbs protium.

Mg–Ti alloys and pure Mg were successfully gas nitrided at 723 K for 129.6 ks under 0.1 MPa N₂. Ti was selected as alloying element due to its high nitrogen affinity. Fine Ti particles in 10 \\micron range were homogeneously distributed in the alloys. Bulk Mechanical Alloying process was used to produce the sample having activated state of Mg, providing high diffusion path for nitrogen. Mg₃N₂ produced is not stable and reacts with moisture, resulting in fine powders of Mg(OH)₂ on the surface and NH₃(g) as by product.

A new artificial bone concept by magnesium alloys is proposed to think much importance on its homogenization with a surrounding natural hard and soft tissue. Magnesium is an essential element for human body, so that magnesium bone implants can be expected to be toxicity free even though magnesium dissolved into human soft tissue. In addition, magnesium base artificial bone has vivo-adaptively to growing bone cells once vivo-coating is formed on the surface of magnesium in the inside of soft tissue. In the present paper, its chemical behavior in Hank’s solution (HBSS (+)) is described to simulate biochemical reactions of magnesium in the human body. An effect of heat treatment of magnesium on its chemical behavior is also investigated. Specimens of 10×20×2 mm³ were used for examining chemical behaviors of commercial grade pure magnesium (3N–Mg) in a HBSS (+) for various holding time (25–700 h). Specific mass gain of each specimen was measured, the surface microstructure was observed by a scanning electron microscope, identification of reaction products were examined by x-ray diffraction measurements. Chemical compositions of reaction products were also analyzed by an energy dispersion x-ray spectrometry. Mass change of heat-treated 3N–Mg, which was heat-treated at 803 K for 90 ks increased with immersing time in HBSS (+) though that of other heat-treated 3N–Mg unstably decreased in HBSS (+). Magnesium reacted with HBSS (+) and then a magnesium apatite was precipitated on the heat-treated 3N–Mg specimen surface. The magnesium apatite should be described as (Ca_0.86Mg_0.14)₁₀(PO₄)₆(OH)₂.

The objective of this study is to improve the sliding wear resistance of the AZ91E magnesium alloy, by means of dispersion of hard particles on its surface. Two kinds of powder and injection methods were applied. SiC powder, fed by Ar gas stream, was injected directly into a melt pool formed by laser beam. And the hypereutectic aluminum-silicon alloy was also injected in order to crystallize the fine Mg₂Si intermetallic compound. These modified surfaces were evaluated by sliding wear test. SiC powder is dispersed uniformly at the surface of the AZ91E alloy, in proportion to the powder feeding rate. In the case of hypereutectic aluminum-silicon alloy, fine dendritic Mg₂Si compound is crystallized uniformly. The matrixes of the modified layers changes from Mg solid solution to Al solid solution according to the powder feeding rate. And hardness of the layers are also changed. The wear depth of the SiC injected layer is almost zero, while that of the pin is increased by the sliding wear test. The modified layer using Al–30Si powder with the high powder feeding rate shows shallow wear depth and wear depth in the pin is also low.

This paper deals with compression test simulations of open cellular magnesium alloy under dynamic loading to analyze the characteristics of energy absorption. Metallic foams are new lightweight materials that have some excellent mechanical and chemical properties. The major characteristics of open cellular Mg alloy are ultra low density (about 0.05 kg/m³) and energy absorption ability. Therefore this material can be used as suitable material for transportation systems. Most studies of metallic foams have been done using the Al close-cell foams so far and there are few studies for Mg open-cell foams. To make the open-cell metallic foams, the polyurethane form is used as the base shapes of metallic foams. As a result, it is very difficult to manufacture foams into the required shapes. If shapes of each cell can be controlled, the mechanical properties of the foams can be designed as expected. In this paper, we simulated the compression test of the shape controlled open-cell materials under dynamic loading made of Mg alloy by Finite Element Method (LS-DYNA). The analyzed sample is 14 mm cubic which includes 27 cell units. The strain rates of simulations are 10², 10³ and 10⁴ s⁻¹. The sample is compressed between static and moving rigid walls. Simulation results show that the difference of strain rate affects the compression behavior and the total amount of absorbed energy.

The characteristics of high temperature deformation in Mg–Y alloys were investigated and compared with those of pure Mg. The effect of Y on slip and twinning deformations was also discussed in respect to the lattice parameter with temperature. Pure Mg displays a gradual decrease in proof stress with increasing temperature. However, the proof stress of as-quenched Mg–Y alloys remains unchanged until about 573 K . The strain rate sensitivity of pure Mg increases above 400 K, whereas in Mg–Y alloys, it becomes larger above 550 K . It is deduced that the increase of the m value is due to the dependence of non-basal slip on temperature. Single slip line and twinning were observed in pure Mg deformed at low temperature. Above 473 K, much cross slip and twinning occurs. In the case of Mg–8Y alloy, cross slip was observed at 673 K only and there was no twinning at any temperature. This is interpreted due to the observation that the c⁄a is constantly maintained by the addition of Y with increasing temperature, resulting in the restraint of the non-basal slip and twinning.

This study investigated the effect of cooling rate on microstructure and fracture characteristics of β-rich α+β type Ti–4.5Al–3V–2Mo–2Fe alloy rolled plate. Particular attention was paid to the roles of the local and continuous secondary phases within prior β grain in the static fracture toughness. A variety of microstructures containing different types, morphologies, sizes and volume fractions of secondary phases were obtained in matrix β (within prior β grain) with varying cooling rates; namely, water-quenching (WQ), air-cooling (AC), furnace-cooling (FC) and slow furnace-cooling (SFC) from various solutionizing temperatures in the α+β field. The types of secondary phases are martensite α (orthorhombic α^′′), acicular α and plate-like α observed in WQ, AC and FC specimens, respectively. While, there is no or lack secondary phase observed in matrix β for SFC specimen. Deformation-induced martensite (α^′′), DIM, was observed in WQ specimens after testing. The results showed that the fracture toughness, J_IC, and calculated flow stress, σ_f, of the microstructures containing a secondary phase depend mainly on the type and width of the secondary phase. The J_IC of microstructures containing a secondary phase, in general, is superior to that of the microstructures lacking secondary phases. Both the martensite α and DIM appear to increase J_IC. In a particular condition, J_IC decreases slightly with increasing width of acicular α for microstructure containing predominantly local acicular α, but increases monotonously with further increases in the width of acicular α. J_IC increases considerably with increasing width of plate-like α. The increase of J_IC is mainly due to increasing the effect of extrinsic toughening mechanism.

The phase stability of molybdenum disilicide (MoSi₂, C11_b structure) relative to other phases, C40 and C54 phases, in the pseudo-binary systems of MoSi₂ and other types of disilicides of transition metals including Cr, V, Nb, Ta and Ti was investigated by establishing the MoSi₂–TSi₂ (T=Cr, V, Nb, Ta, Ti) pseudo-binary phase diagrams. It was found that V, Nb, Ta and Ti which substitute for Mo in MoSi₂ strongly stabilize the C40 phase while Cr only shows a weak C40 structure-stabilizing effect. The phase stability was also discussed on the basis of geometrical change in the three phases when composition varies. Change in lattice parameter of each phase indicated that phase stability of C11_b, C40 and C54 structures greatly depend on the relative stacking spacing of the equivalent hexagonal atomic plane for the three structures. Based on the results of phase diagram investigation, the present work attempted to design a C11_b/C40 lamellar microstructure. Existence of the equivalent hexagonal stacking atomic plane among C11_b, C40 and C54 phases makes it possible to design a coherent C11_b/C40 two-phase microstructure. Assuming the equivalent atomic planes in C11_b and C40 phases coincide with each other, high lattice coherency between C11_b and C40 phases is available in MoSi₂–CrSi₂ system with a lattice misfit of less than 0.5% and in the other four systems with a misfit of 1.3 to 2.7%. A lamellar microstructure was observed in MoSi₂–TaSi₂ and MoSi₂–NbSi₂ systems but no lamellar microstructures were obtained in MoSi₂–CrSi₂, MoSi₂–VSi₂ and MoSi₂–TiSi₂ systems.

Cast ingots of quasicrystal forming composition Al₆₂Cu₂₆Fe₁₂ and Al₇₂Ni₁₂Co₁₆ were prepared in an inductive melting apparatus using graphite crucible in the aim of investigating the effect of carbon on the stability of the single icosahedral (i) and decagonal (d) quasicrystalline phases, respectively. Owing to this process, the contamination of carbon was expected to occur during melting by diffusion from the crucible into the quasicrystalline ingots. The detected amounts of carbon in the i- and d-phase were 0.15 mass% and 0.88 mass%, respectively. Carbon introduced by this method was first found to disintegrate spontaneously these monolithic i-AlCuFe and d-AlNiCo ingots, i.e., a detrimental effect of self-fragmentation of quasicrystals induced by carbon. Our experimental results indicated that the self-fragmentation occurring in the carbon contaminated i- and d-phases did not appear immediately after solidification, but after a certain incubation time. During the self-fragmentation process, the quasicrystalline samples with an initial size of 20 mm diameter bar ingot were disintegrated to form small fragments of about 200 \\micron after 50 days. Fractography was carried out to analyze fracture modes in the i- and d-phases. The fracture surfaces of the i-phase exhibited a transgranular cleavage pattern without trace of cleavage steps and river patterns, while the d-phase showed both transgranular cleavage and intergranular fracture. Such unique fracture modes in the i- and d-phases are thought to arise from an internal mechanism of carbon-assisted fracture.

In order to elucidate the reduction kinetics and mechanism of natural ilmenite ore with carbon monoxide, reduction experiments have been carried out using thermogravimetric technique in the temperature range between 1173 and 1323 K with three kinds of natural ilmenite ores from Australia, Malaysia and China. The reduction rate was analyzed in terms of the mixed-control kinetics by applying a shrinking unreacted-core model, on the basis of the observation of cross section of partially reduced ilmenite and X-ray diffraction patterns. The determined reaction rate constant k_c and effective diffusivity D_e are expressed in the temperature range between 1173 and 1323 K by the following equations: k_c/m·s^-1&=exp{-113×10^3/(RT)+5.03} (Australian)
k_c/m·s^-1&=exp{-47.0×10^3/(RT)-3.04} (Malaysian)
k_c/m·s^-1&=exp{-71.6×10^3/(RT)+0.554} (Chinese)
D_e/m^2·s^-1&=exp(-4250/T-7.54) (Australian)
D_e/m^2·s^-1&=exp(-3860/T-7.94) (Chinese) where R: gas constant (J·mol⁻¹·K⁻¹), T: temperature (K). The activation energy of the reaction is 113, 47.0, 71.6 kJ·mol⁻¹ for Australian, Malaysian and Chinese natural ilmenite ore, respectively. The calculated reduction curves using the rate parameters reproduced the experimental data well. The reduction rate increases with increasing reduction temperature. With respect to the rate-determining step for Australian and Chinese ilmenite ores, the relative contribution of resistance of a mass transfer step through gas film is smaller than that for synthetic ilmenite, while the relative contribution of resistance of a diffusion step of carbon monoxide through pores of the product layer is larger than that for synthetic ilmenite. As for Malaysian ilmenite ore, the overall rate is mainly controlled by a chemical reaction step.

The breakaway behavior of surface oxide film on Al–Si–Mg alloy powder particles was investigated by X-ray photoelectron spectroscopy using synchrotron radiation (SR-XPS), which is a topmost-surface analysis method with less than 1.5 nm analysis depth and high sensitivity. During the heating of the powders up to 823 K, the dissolved magnesium in the powder particles was concentrated in the surface oxide below 670 K, and saturated in the composition with a Mg/Al ratio of approximately 0.5. Over 670 K, the metallic aluminum appeared on the topmost surface of the particles. It was speculated that the surface reaction was due to the reduction of surface oxide film by the concentrated magnesium or the breakage of the surface oxide film by the crystallization and growth of MgAl₂O₄.

In order to investigate the low cycle fatigue behavior of lamellar structured Ti–46.6Al–1.4Mn–2Mo (at%) alloy, total strain range controlled creep-fatigue and continuous fatigue tests (R=−1, strain rate=4×10⁻³ s⁻¹) were carried out at 800°C. A drastic reduction of fatigue life is observed in creep-fatigue test compared with continuous fatigue test. Microstructural and compositional changes during the creep-fatigue test were investigated using SEM, TEM and AES (auger electron spectroscopy). It was found that this reduction of fatigue life in creep-fatigue test was understood to be due to the additional creep damage occurring during tensile hold time. Recent reports indicate that the lamellar TiAl alloy has a different creep deformation mechanism from general metallic materials. For the lamellar TiAl alloy, it has been reported that creep deformation is controlled not by the self-diffusion assisted dislocation climb but by the dislocation generation due to the α₂→γ phase transformation at the lamellar interface. This implies that creep damage induces microstructural phase change during which moving dislocations are generated for the continuous deformation. Therefore, it is very important to investigate the effect of creep deformation on creep-fatigue damage in terms of microstructural change. In the present study, microstructural analysis using SEM, TEM and AES (auger electron spectroscopy) for compositional analysis were conducted. It is found that, under creep-fatigue test, the creep damage causes α₂→γ phase transformation at the grain boundary, from which intergranular cracks are initiated to control the creep-fatigue behavior.

Fe–3%C–4%Si alloys were melted and solidified under various cooling rates and pressures up to 1000 MPa using a piston-cylinder type high pressure apparatus. Carbide eutectic was observed in the inner region of the specimens slowly cooled under a pressure of about 800 MPa. The observed transition from graphite eutectic (Gray) to carbide eutectic (White) was explained by the pressure dependence of the stable and metastable eutectic temperatures. Solidification analysis involving the pressure dependence of eutectic temperatures agreed well with the observed chill fraction. The effective pressure for Gray to White transition was estimated to be more than 200 MPa for commercial grade cast irons. From the stress analysis for cylindrical castings having different sizes, it was suggested that the maximum pressure generated during solidification depends on the strength of outer solidified shell and can be greater in spheroidal graphite cast iron than in flake graphite cast iron.

We examined the microstructural characters of L1₂-type (Al, X, Y)₃Ti (X, Y=Mn, Cr, Ag, Fe) phase alloys, with compositions selected from the regions connecting ternary L1₂-type (Al, X)₃Ti and (Al, Y)₃Ti single-phase regions in the quaternary phase diagrams using laser microscopy, scanning electron microscopy and electron probe microanalysis. The residual strain, which was induced during milling of the alloys, was measured by examining the X-ray diffraction peak profile of the alloy powders. Most of the (Al, X, Y)₃Ti alloys heat-treated at 1450 K for 24 h exhibited the L1₂ single-phase structure, while a few alloys containing Ag showed two-phase structures consisting of L1₂ and Ag-rich phases. The lattice parameters of (Al, Mn, Ag)₃Ti and (Al, Cr, Ag)₃Ti increased remarkably with increasing Ag content, while those of (Al, Mn, Cr)₃Ti, (Al, Mn, Fe)₃Ti and (Al, Cr, Fe)₃Ti decreased gradually with increasing Cr or Fe content. The change of the lattice parameter with the contents of ternary and/or quaternary elements could be explained by the atomic size effect. The size and fraction of porosity, which were formed during heat treatment at 1450 K for 24 h, depended on the composition of the alloys. The changes of the size and fraction can be explained by the Kirkendall effect, irrespective of alloy systems. The increase of the residual strain by variation of the composition in the Al–Ti–Mn–Ag and Al–Ti–Cr–Ag systems may be explained by the fact that the decrease of size and fraction of the porosity increased the residual strain. However, in the Al–Ti–Mn–Cr, Al–Ti–Mn–Fe and Al–Ti–Cr–Fe systems, the change could be explained by the fact that the residual strain increased as the sum of the contents of Al and Ti decreased.

Sintering to full density enables powder metallurgy products to compete with castings and forgings. Subsieve powders with high inherent sinterability provide one means to attain densification, but at a substantial cost penalty when compared with readily available coarse compaction grade powders. Unfortunately, for large powders the sintering stress that causes densification is small and often insufficient to overcome the inherent compact strength that resists rapid densification. In such cases, only slow diffusion-controlled densification occurs. The current analysis identifies an option for sintering densification of large particles based on a comparison of the sintering stress and component strength during heating. Rapid densification occurs when the in situ strength is reduced to levels comparable to the sintering stress. On this basis, alloy systems are identified for full density sintering using thermal softening concepts.

The solid state reaction, (1⁄2)Nd₂O₃+(1⁄2)Cr₂O₃→NdCrO₃, was studied between 1473 K and 1773 K in air to determine the parabolic rate constant. The electrical conductivity of the NdCrO₃ phases separately equilibrated with Nd₂O₃ and Cr₂O₃ were measured between 1673 K and 1773 K in an argon-oxygen mixture. The standard Gibbs energy change for the solid state reaction was estimated from the activity dependence of the electrical conductivity of NdCrO₃. According to Wagner’s theory, the diffusion coefficient of Nd³⁺ in NdCrO₃ can be evaluated from the parabolic rate constant for the solid state reaction, the standard Gibbs energy change of the solid state reaction, and the defect equilibrium reaction in NdCrO₃. From these data, the diffusion coefficient of Nd³⁺ in NdCrO₃ was given by D_Nd^3+/m^2s^-1 = 1.3 ×10^-6 ·exp\\left(-\\frac343 kJ mol^-1RT \\ight) (P_O_2/Pa)^3/16 a_Cr_2O_3^1/8where a_Cr2O3 is the activity of Cr₂O₃ and P_O2 is the oxygen pressure. The obtained diffusion coefficient of Nd³⁺ in NdCrO₃ was compared with that of Y³⁺ in YCrO₃ and that of La³⁺ in LaCrO₃.

A study of phase decomposition in the Cu–34% Ni–4%Cr and Cu–45% Ni–10%Cr alloys was carried out in order to analyze the kinetics and morphology of phases. Samples were solution treated, quenched and aged at 600, 700 and 800°C for different times. The solution treated and aged samples were characterized by X-ray diffraction analysis, analytical transmission electron microscopy and field ion microscopy. The X-ray diffraction results indicated the presence of band sides on the 200 diffraction peak and an increase in its intensity with aging time. The TEM microanalysis showed the chemical composition of the decomposed phases increased with aging time. These facts suggest that the phase decomposition took place by the mechanism of spinodal decomposition. The decomposed phases are coherent and aligned on the ⟨100⟩ directions during the first steps of aging, which promote the increase in hardness. The coherency between the decomposed phases disappeared with prolonged agings, which facilitates the overaging. The morphology of decomposed phases changed from equiaxial to cuboids and then to an elongated plate oriented on the ⟨100⟩ directions with the increase in aging time. The equilibrium phases corresponded to a mixture of Cu-rich and Cr-rich phases. Their compositions showed a good agreement with those predicted by the miscibility gap in the equilibrium Cu–Ni–Cr phase diagram. The chromium-richer alloy showed the fastest kinetics of phase decomposition and coarsening, as well as the highest maximum of hardness.

The fluxless wetting properties were evaluated for TSM (Top Surface Metallurgy)-coated glass substrate by wetting balance method. The wettability of TSM was estimated with new indices from the wetting curve for one side-coated specimen. For TSM, it was more effective to use Cu as a wetting layer with Au as a protection layer than to use Au as a wetting layer alone. The degree of wetting of Sn–5Sb and Sn–57Bi solder was better than that of Sn–37Pb and Sn–3.5Ag solder. However, the wetting rate of Sn–37Pb solder was higher than those of Pb-free solders. The wetting angle of TSM-coated glass substrate to solders could be calculated from the force balance equation by measuring static state force and tilt angle. The wetting angles of Cr/Cu/Au TSM to Sn–37Pb, Sn–3.5Ag, Sn–5Sb and Sn–57Bi were 69.4^°, 67.2^°, 60.1^° and 53.0^°, respectively.

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 amorphous-forming composition range (AFCR) was calculated for 338 ternary amorphous alloy systems on the basis of the database given by Miedema’s model in order to examine the applicability of the model, to analyze the stability of the amorphous phase, and to determine the dominant factors influencing the ability to form an amorphous phase. The mixing enthalpies of amorphous and solid solution phases were expressed as a function of alloy compositions on the basis of chemical enthalpy. Based on the Eshelby and Friedel model, an elastic enthalpy term was added to the model for the solid solution. Furthermore, an average melting temperature of the constituent elements was added to the model as the topological enthalpy in an amorphous phase. An amorphous phase was assumed to have been formed at the composition where the enthalpy of an amorphous phase was smaller than that of a solid solution. The AFCR was calculated for 335 systems except for the Al–Cu–Fe, Al–Mo–Si and Au–Ge–Si systems. The calculated results are in agreement with the experimental data for Cu–Ni- and Al–Ti-based systems. For typical amorphous alloy systems exemplified by the Zr-, La-, Fe- and Mg-based systems, it was recognized that the calculated AFCR had been overestimated as a result of the model being simplified. We have also shown that the elastic enthalpy term arising in a solid solution phase stabilizes the amorphous phase, and the stabilization mechanism is particularly notable in Mg-based amorphous alloy systems. Short-range order plays an important role in the formation of Al-, Fe- and Pd-metalloid based systems. The following factors have a great influence on amorphous-forming ability: (1) three empirical rules for the achievement of high AFA, (2) melting temperature and viscosity at the melting temperature, (3) elastic enthalpy arising in a solid solution, and (4) short-range order observed in an amorphous phase. The importance of the latter two factors was only identified as a result of the present study.

High-purity single crystalline B₄C, several cm in size, has been prepared by a Floating Zone method. The electrical conductivity, Hall mobility, Seebeck coefficient and thermal conductivity are measured from room temperature to 1023 K . The electrical conductivity is about four times higher than that of sintered B₄C . The mobility has increased with increasing temperature, indicating hopping conduction of charge carriers. The Seebeck coefficient is 240–260 \\microVK⁻¹, almost independent of temperature. The thermal conductivity is about twice higher than that of sintered B₄C . The thermoelectric figure-of-merit is also higher than that of sintered B₄C, being 4.2×10⁻⁵ K⁻¹ at 1100 K.

The SiC-dispersed β–FeSi₂ compounds were synthesized by pressureless sintering with Cu and Si addition without a subsequent heat treatment for β phase formation. The effects of Cu and Si addition on the phase transformation, sintering behavior, and thermoelectric performance of these samples were investigated. Cu addition significantly accelerated the β phase formation during the pressureless sintering, and the samples with Cu were mostly composed of the β phase with a small amount of the residual α and ε phases. Cu addition was effective for densification; Cu–Si liquid phase formation resulted in a high relative density of over 90%. The thermoelectric power and thermal conductivity of the pressurelessly sintered samples with Cu addition were quite lower and higher, respectively, than those of the hot-pressed sample. The differences were caused by a larger amount of metallic phases, such as the ε and Cu–Si phases. The figure of merit of these samples was about 1/3 of that of the hot-pressed sample, indicating that Cu addition completely ruined the enhancement of thermoelectric performance obtained by SiC dispersion. The addition of Si to the samples with Cu decreased the amount of the metallic ε phase and increased the amount of the semiconducting β phase through the reaction ε+Si→β. The thermoelectric power and thermal conductivity of the samples with Cu were markedly improved by Si addition, which resulted in a significant increase in the figure of merit. The values of the figure of merit for the samples with Cu and Si addition were almost the same as that of the hot-pressed sample with SiC dispersion.