Metallurgical chemistry's principle is that “the chemical process is governed by chemical potential.” Successful chemical process technology follows a route that does not go against the governance of chemical potential. In order to realize the technological objective, raw materials and a reactor are necessary, and after the principle is established, the method is supported by the reactor and advances in the materials that comprise such apparatus. Therefore, the technology is a fusion of material, apparatus, experience, and science, all of which are parts of the foundation of a technological method. The author's involvement is described as an academician from the postwar recovery to the technological rearmament period. In the postwar recovery period, a systematic point-of-view was introduced to a technological principle using phase diagrams that clarified chemical potential, which was a new concept in the field of thermodynamics. In the technological rearmament period, technological evaluation was conducted from a philosophical standpoint.

The superplastic behavior of fine-grained metals is well described by the deformation model in which grain boundary sliding (GBS) is accommodated by slip. This slip accommodation process involves the sequential steps of glide and climb, with climb assumed to be the rate-controlling process. The climb distance during GBS is often considered to be on the order of grain size in the conventional theoretical models. However, these models have not been able to predict quantitatively the strain rates actually observed in fine-grained superplastic materials. Therefore, the deformation model was reviewed by comparing the theoretical and phenomenological equations in order to accurately understand the mechanism of the accommodation process. The analyses revealed that the climb process is governed by the effective diffusivity. The climb distance through the grain boundary is of the order of the grain size, and that through the lattice close to the dislocation core size was quantitatively in agreement with the phenomenological relation.

Commercial 7075-T6 aluminum alloy was subjected to friction stir welding (FSW), resulting in development of a fine-grained structure with average size of about 3 μm in the nugget zone. Static annealing at temperatures ranging from 623 to 773 K for 30 min showed that the fine grain microstructure was stable at temperatures not higher than 723 K. Increase in annealing temperature up to 773 K led to an abnormal grains growth, followed by the development of mm-scale grains. The specimens obtained from the nugget zone demonstrated a superplastic behavior at temperatures ranging from 623 to 723 K and at strain rates ranging from 1 × 10⁻⁴ to 1 × 10⁻² s⁻¹. Large elongation of about 440% was observed at a temperature of 673 K and at a strain rate of 1 × 10⁻³ s⁻¹.

The tensile properties and blow forming characteristics of 5083 Al alloy recycled by solid-state recycling were investigated from the viewpoint of oxide contamination. Three types of machined chip with different volumes were recycled by hot extrusion and hot rolling in air. Oxide layers, which were contaminants from the machined chip surface, were distributed in the extrusion direction for the recycled specimens. Oxygen concentration in the recycled specimens increased with the total surface area of the machined chips per unit volume. From the result of the tensile tests performed at 773 K, the elongation to failure of the specimen made of smaller machined chips was lower, than that of the specimens made of larger machined chips, in spite of their same strain rate sensitivity of 0.5. Similarly, in the blow-forming tests at 773 K, the specimen made of smaller machined chips exhibited a lower formability. The low elongation to failure and formability of the recycled specimens made of smaller machined chips are likely to be attributed to a greater contamination of oxide. Thus, oxide contamination has a detrimental effect on the superplastic properties of recycled Al alloy.

The superplastic blow forming limits for a 5083 aluminum alloy were investigated at a strain state from the plane strain to the balanced biaxial at a temperature of 773 K and a strain rate of about 1 × 10⁻³ s⁻¹. It was discovered that the equivalent strain at the forming limit was in a range of 1.3 ± 0.1 for every strain state and was consistent with that obtained by the uniaxial tensile tests. Namely, the forming limit for the blow forming could be estimated from the equivalent strain-to-fracture in the uniaxial tensile test. The cavity volume fraction and the cavity growth rate at the forming limit, however, increased by changing the strain state from uniaxial to balanced biaxial through a plane strain. The relationship between the forming limit and the cavity volume fraction is dependent upon the strain state.

Equal-channel angular pressing (ECAP) was successfully applied to samples of an Al-1% Mg-0.2% Sc alloy in the form of plates. Pressings were conducted at room temperature using route B_CZ in which the plates are rotated in the same direction by 90° around the vertical axis between each pass. Following ECAP, the grain size was measured as ∼0.5 μm. Tensile specimens were cut from the plates and pulled to failure at a temperature of 673 K. Superplastic elongations were achieved and the measured ductilities were essentially independent of the orientation within the plate. These results provide the first demonstration of the potential for directly producing a plate in a superplastic condition through the application of ECAP. Thus, this new procedure removes the necessity of rolling conventional as-pressed bars or rods into sheets or plates for subsequent superplastic forming operations.

The change in crystallographic orientation distribution during high temperature deformation for an Al-Mg-Mn alloy sheet consisting of the coarse-grained surface and the fine-grained center layers has been investigated in order to reveal the deformation mechanism. The grain size dependence of the deformation behavior is discussed in the identical deformation condition by using the specially-prepared sheet. The grain structures in the coarse-grained surface layer of the sample deformed at 713 K are elongated in the tensile direction corresponding to the macroscopic elongation to failure. The structures related to the maximum elongation in both of the surface and center layers have preferred orientations of the tensile deformation. Further, the intragranular misorientation, grain boundary misorientation and high strain rate deformation are analyzed in detail.

The applicability of high-strain-rate superplasticity for forming magnesium parts, especially structural component with a boss on a plate designed with the aim of producing mobile electric appliances, was examined. The required microstructure and grain refinement process for target forming was examined prior to the forming trials. Following the examination of the processing design, the high-strain-rate superplastic AZ91 magnesium alloy with a required grain size less than 3.2 μm was produced by hot extrusion under the condition that the Zener-Hollomon parameter was 3 × 10¹² s⁻¹. It was experimentally confirmed that a boss with a height greater than 5 mm, which is a requirement for electric appliances, could be formed within 10 s even at a low temperature of 523 K using this high-strain-rate superplastic magnesium alloy. Basic knowledge related to the formation of three-dimensionally shaped magnesium parts by utilizing high-strain-rate superplasticity was obtained.

Extruded Mg-6 mass%Al-1 mass%Zn (AZ61) alloy was grain-refined utilizing Equal Channel Angular Extrusion (ECAE) processing. Initially, the extruded bar of the alloy was ECAE-processed 2-times at 473 K. Subsequently, it was processed 4-times at 448 K. As a result, the grains are refined to less than 1 μm and a large amount of fine Mg₁₇Al₁₂ compound precipitates. Subsequently, the superplastic properties of the ECAE-processed specimens were investigated. Large fracture elongations of over 300% are obtained at 423 K and 448 K, which is below T_m/2 (T_m: melting point of the alloy), at strain rates above 1 × 10⁻⁴ s⁻¹ and 1 × 10⁻³ s⁻¹, respectively. That is, low temperature superplasticity occurs. Furthermore, at a high strain rate of 1 × 10⁻² s⁻¹, superplasticity occurs with the elongation of 242% and 398% at relatively low temperatures of 473 K and 523 K, respectively. In the extraordinarily elongated specimens, significant grain boundary sliding is observed with strain rate sensitivity of 0.3∼0.4. The activation energy for superplastic deformation is about 91 kJ/mol, which is close to that for grain boundary diffusion of pure magnesium. It is concluded that the superplastic deformation mechanism of the investigated alloy would be grain boundary sliding accommodated by dislocation slip controlled by grain boundary diffusion.

It has been reported that superplastic Zn-22 mass%Al alloy has some excellent properties needed for seismic damping devices. In addition, seismic damping devices using this alloy has already been put into practical use for a high-rise building. In the present investigation, we tried to apply this alloy to seismic damping devices for a general residence and examined the formability and the characteristics after the formation. No cracks were observed in the sample formed even at room temperature. The surface roughness, R_a, was improved from 0.602 μm before the formation to 0.418 μm after the formation. As the results of FVM analysis, it was predicted that the formed sample could exhibit the higher stress and strain at both sides of the inferior part of the gauge section. However, the average grain size, the ratio of the major axis length to minor axis one in a grain, and the hardness in the region exhibiting high stress and strain were almost the same characteristics as those in the initial sample.

The cavitation behavior in room-temperature deformation was investigated for Zn-22 mass%Al alloy by metallographical observations, and compared with that in high-temperature deformation. It was predicted from metallographical observation that the critical cavity radius of transition in growth mechanism from diffusion-controlled to plasticity-controlled would exist. In the range of plasticity-controlled growth mechanism, the cavity growth parameter, η, exhibited 1.65 and 2.45 in the deformation at high temperature and room temperature, respectively. In primary period of straining, cavities in room-temperature deformation could be nucleated more easily compared with those in high-temperature deformation. The experimental cavity growth rates in both conditions were almost same at cavity size of about 1 μm.

Ceramics composites are suitable as the materials for a micro electro mechanical system (MEMS) parts in terms of its mechanical and physical properties. We apply the PM process in combination with mechanical alloying (MA) and Spark Plasma Sintering (SPS) to produce micro-parts using a ceramic composite, TiC/Ti₅Si₃. Powders of elements Ti and SiC whose composition is Ti-20 mass%SiC are blended for MA. After the alloying, the MA powder whose average particle size is 20∼30 μm, has amorphous-like structures. Results of compression-tests and TEM observations indicate the occurrence of unusual high-temperature deformation behaviors such as low flow stress at the lower deformation temperature. The deformation is attributed to a pseudo-superplasticity in which the phase transition of metastable microstructure occurs during the deformation. The pseudo-superplasticity observed in the SPS compact improves the formability in fabrication of the micro-parts. The MA powder is filled into a micro-size mold produced by LIGA process, and casts together by SPS in order to fabricate a new micro-parts using TiC/Ti₅Si₃. Optimization of the pseudo-superplasticity enables the fabrication of the micro-parts using TiC/Ti₅Si₃.

The superplastic deformation behaviour of a Ni-1 mass%SiC nanocomposite produced by pulse electrodeposition was investigated at temperatures of 410°C and 450°C and strain rates ranging from 8.3 × 10⁻⁴ to 5.0 × 10⁻² s⁻¹. A maximum elongation of 836% was obtained at 450°C and a strain rate of 1.67 × 10⁻² s⁻¹, which is the first observed result of the high strain rate and low temperature superplasticity for Ni-SiC nanocomposites. Scanning electron microscopy and transmission electron microscopy were employed to examine the microstructure of the asdeposited and deformed samples. The superplastic behaviour of the Ni-1%SiC nanocomposite was analysed through observations of its fracture surfaces and microstructures. The results showed that SiC nanoparticles play an important role in the stability of the microstructure of the Ni-SiC nanocomposite. A low volume fraction of cavity is necessary for a large elongation. The mechanisms of high strain rate and low temperature superplasticity of the composite are discussed in the paper.

Superplastic flow behavior in 1 mol% of GeO₂ and 1 mol% of NdO_1.5 co-doped ZrO₂-3 mol%Y₂O₃ (3Y-TZP) was examined at 1400°C under an initial strain rate of 1 × 10⁻⁴ s⁻¹. 1 mol% of GeO₂ or NdO_1.5-doping slightly enhances high-temperature ductility in 3Y-TZP, but 1 mol% of GeO₂ and 1 mol% of NdO_1.5 co-doped TZP exhibits large elongation to failure of more than 600% at 1400°C. The large ductility in TZP due to Ge⁴⁺ and Nd³⁺ co-doping can be explained from reduction in the flow stress. High-resolution electron microscopy (HREM) and energy-dispersive X-ray spectrometer (EDS) analysis revealed that Y³⁺, Ge⁴⁺ and Nd³⁺ cations segregate in the vicinity of grain boundaries in the present materials. The segregation width of the dopant cation across the grain boundaries in GeO₂ and NdO_1.5 co-doped TZP is larger than that in GeO₂ or NdO_1.5 singly doped TZP. The reduction in the flow stress due to GeO₂ and NdO_1.5 co-doping is probably related to the increment in the segregation width.

Superplastic behavior in a fine-grained, GeO₂-doped 3 mol% yttria-stabilized tetragonal zirconia polycrystal (3Y-TZP) with the dopant level of 0.2—3 mol% was examined at 1400°C under an initial strain rate of 1.3 × 10⁻⁴ s⁻¹. Microstructure and chemical composition at the grain boundaries was examined by high-resolution electron microscopy (HREM) combined with an X-ray energy dispersive spectrometer (EDS). No secondary phase was observed along the grain boundaries, though EDS analysis indicated the segregation of Ge cations along the grain boundaries. The Ge content at the grain boundaries tends to increase with increasing the total amount of GeO₂ addition, but saturate over the doping level of 2 mol%. Dependence of flow stress reduction on the total amount of GeO₂ addition has a good correlation with Ge content at the grain boundaries. This fact indicates that the GeO₂-doping effect on the flow stress in 3Y-TZP is caused mainly from the grain boundary segregation of Ge cations.

A novel small angle neutron scattering (SANS) method was applied for characterization of cavities in a 3Y-TZP subjected to superplastic deformations under different conditions, and the results were compared with conventional methods, i.e. SEM analysis and density measurement (Archimedes) method. Morphology of cavities formed in the 3Y-TZP specimens varied depending on the deformation condition and the dimension of the specimen. Effects of cavity morphology on the cavity characterization were investigated and were discussed for each method. It was found that: (1) The accuracy of the data obtained from SEM analysis was very sensitive to the morphology of cavities, while (2) Archimedes and SANS methods gave similar results, and showed good reliability for evaluation of bulk properties of cavities.

Fast neutrons (energy > 1.6 × 10⁻¹³ J) were irradiated to tetragonal zirconia polycrystals containing 3 mol% yttria (3Y-TZP) at the fluence levels of 2.5 × 10²⁴ (Light irradiation) and 4.3 × 10²⁴ (Heavy irradiation) m⁻². The irradiation caused no significant swelling in the 3Y-TZP specimens. After the neutron irradiation, superplastic characteristics were examined by tensile tests at a temperature range from 1623 to 1773 K with initial strain rates ranging from 5.0 × 10⁻⁴ to 1.67 × 10⁻² s⁻¹. It was found that the elongation to fracture of the irradiated specimens was quite small in comparison with that of unirradiated ones. The apparent activation energy for the superplastic flow of the irradiated 3Y-TZP was fairly high, i.e. 785 ± 35 and 693 ± 26 kJ·mol⁻¹ for Light and Heavy irradiations, respectively. It appears that the induced defects, nuclear transmutation and radiation-induced segregation in the 3Y-TZP due to the neutron irradiation are responsible for these property changes.

The electrical resistivity values of Pd₄₀Cu₃₀Ni₁₀P₂₀ alloy in the supercooled liquid and liquid states have been measured by the direct current four probe technique and the experimental uncertainty of electrical resistivity was found to be ±2.2%. The electrical resistivity values of Pd₄₀Cu₃₀Ni₁₀P₂₀ alloy show negative temperature dependence in the supercooled liquid and liquid states. It was also suggested from analysis with the Wiedemann-Franz law that principal mechanism for thermal conduction of Pd₄₀Cu₃₀Ni₁₀P₂₀ alloy is likely to be controlled by electron in the supercooled liquid and liquid states.

Magnetism in clusters having upto 15 atoms of non-magnetic element, Rh is studied using the ab initio ultrasoft pseudopotential method and generalized gradient approximation for the exchangecorrelation energy. The lowest energy structures are found to have no atom at the center upto n = 13 and have low symmetries. The well known icosahedral structure for 13 atoms does not have the lowest energy. A transition to more compact structures with an atom at the center occurs beyond 13 atoms. The calculated magnetic moments are in better agreement with experiments than obtained before. An H atom on these clusters favors a bridge side and generally reduces the magnetic moment.

To expand our knowledge of electronic structures and magnetic properties for ferromagnetic shape memory Co-Ni-Al alloys, which undergo a thermoelastic martensitic transformation from a B2 to an L1₀ structure, we have performed first-principles band calculations for the composition of Co:Ni:Al = 1:1:1, i.e., (Co_1/3Ni_2/3)(Co_1/3Al_2/3) and (Ni_1/3Co_2/3)(Ni_1/3Al_2/3). Their electronic structures have been calculated for the supercell structure, including a cubic-tetragonal distortion and a spin polarization. The obtained total energy indicates that (Co_1/3Ni_2/3)(Co_1/3Al_2/3) may be more appropriate than (Ni_1/3Co_2/3)(Ni_1/3Al_2/3). In the paramagnetic state of (Co_1/3Ni_2/3)(Co_1/3Al_2/3), it has been found that the transformation from the B2 to the L1₀ structure comes from the change of the environment of the Co atoms (Co[2e]) on the original Al sites and, other Co and Ni atoms. These features are reflected in their d-orbital density of states (DOS). The Co[2e] atoms also play an important role in the magnetic transition between the paramagnetic and ferromagnetic states. The Co[2e] atoms carry magnetic moments corresponding to those of fcc Co (hcp Co) and the energy gain due to the spin polarization is brought. The origin of the spin polarization can be attributed to the similarity of their environment. This is confirmed by the similarity of their d-orbital DOSs. The “band energy” estimated from total DOS shows that the changes of DOS near the Fermi level bring the band-Jahn-Teller-type stabilization of the distorted structure.

Binary C14 Laves phase alloy CaMg₂ is expected to absorb hydrogen more than 6 mass%, when the H/M ratio reaches 2. We have found that the binary CaMg₂ did not absorb hydrogen at ambient temperature and hydrogen pressure but Ni added CaMg_1.8Ni_0.2 which is also C14 Laves phase absorbed 6.0 mass% of hydrogen at room temperature. CaMg_1.8Ni_0.2 formed hydrides of Mg and Ca, and the C14 Laves phase disappeared. Hydrogen desorption was not observed from hydrogenated CaMg_1.8Ni_0.2. (Ca_0.8La_0.2)Mg_2.2Ni_0.1 absorbed 5.1 mass% hydrogen and H/M ratio reached 1.8 at room temperature. After hydrogenation, the lattice of the Laves phase expands 4% in a-axis and 3% in c-axis. The volume expansion is 13%. The Laves phase structure is kept and hydrogen was absorbed in the lattice. The hydride of (Ca_0.8La_0.2)Mg_2.2Ni_0.1 decomposed from 470 K with increase at temperature and hydrogen was released from the products of decomposition from 610 K.

The precipitation behavior and orientation relationships of icosahedral quasicrystalline phase (I-phase) and Laves phase precipitates in ferrite matrix have been investigated by transmission electron microscopy (TEM) in an Fe-10Cr-1.4W-4.5Co-0.3Si (at%) alloy. It is found that the precipitates of the alloy aged at 873 K are the I-phase but those of the alloy aged at 973 K are the Laves phase. Through a double aging experiment at both temperatures, it is shown that the transformation between the I-phase and the Laves phase occurs. Although a single orientation relationship is established between the I-phase and the ferrite matrix, three different types of the orientation relationships between the Laves phase and the ferrite matrix are obtained by the analysis of SAD patterns. The results can be explained by the coincidence between the five-fold symmetrical plane of the I-phase and the (1120) plane of the Laves phase on the phase transformation between them.

A new method to generate 3 dimensional finite element method (FEM) models of composites with random materials arrangement has been proposed. Some basic models are used to represent the structure of the composites, in each basic model, the dispersions are assumed to have the same geometry, and the structure of a basic model can be described by some structural parameters such as the geometry, number and volume fraction of the dispersion and so on. A program has been developed to automatically generate the geometric model and FEM mesh of the basic models according to given structure parameters. More complicated composite structure can be composed by combination of the basic models. The effective thermal conductivity of composite has been calculated based on the models generated with this method. Optimization of modeling parameters such as model scale and mesh refinement has been discussed with consideration of both calculation accuracy and computation efficiency. The calculated thermal conductivity has been compared to the values obtained by analytical method.

In this paper, the electromagnetic wave absorption properties of Co-Ti substituted Ba M-type hexagonal ferrite (BaFe_9.5(Co_1−yTi_y)_2.5O₁₉ (y = 0.4 ∼ 0.8)) produced by a modified coprecipitation method were investigated. This modified chemical coprecipitation method was the combination of the coprecipitation and the synthesis from a salt melt. The sample, Ba_9.5(Co_0.5Ti_0.5)_2.5O₁₉ sintered at 1423 K for 5 h, whose powder was produced by the modified coprecipitation method, exhibited high permeability (μ_r″ _max = 11) and reflection loss (R.L.) around −10 dB at a relatively wide frequency range (0.8 ∼ 9 GHz) comparable to the Ultra Wide Band (UWB). In order to enhance the sintering, the addition of Bi₂O₃ was carried out. The Ba_9.5(Co_0.4Ti_0.6)_2.5O₁₉ sample sintered at 1423 K for 5 h with a Bi₂O₃ content of 2 mass%, exhibited high permeability (μ″_{r max} = 25 and μ″_{r at 1 GHz} = 10). This sample also showed good microwave absorption properties (reflection loss: R.L. < −20 dB) with a small matching thickness (d_m) of 3.1 mm at 0.65 GHz and f · d product (f · d) of 2.0 GHz·mm. It can be said that the Ba M-type ferrite produced by the modified coprecipitation method has a possibility to become a material for the thinner microwave absorber in the several GHz range.

Pressure-composition (P-C) isotherms of RNi₅-H (R = Pr, Nd, Sm and Gd) systems show two hydrogen pressure plateaux during hydrogen absorption and desorption. These correspond to the transitions among three phases: one hydrogen solid solution (α phase; RNi₅H_∼0.5) and two hydrides (β phase; RNi₅H_3∼4 and γ phase; RNi₅H_5∼7). To explore the mechanism of the phase transitions, we investigated the structural changes in RNi₅-H systems with ex-situ XRD. During hydrogen desorption, the ex-situ XRD profiles show that the crystal structures in RNi₅-H systems change from hexagonal (γ phase) through monoclinic (β phase) to hexagonal (α phase). The temporary decrease in structural symmetry may be due to hydrogen occupation except in the basal plane of the crystal structure. Lattice expansion between the α and β phases normalized in the monoclinic structure decreases slightly with the increasing atomic number of R in RNi₅-H systems. Similarly to the correlation between the first plateau pressure and the unit cell volume of the α phase, the second plateau pressure is logarithmically related to the unit cell volume of the β phase. As far as the RNi₅-H (R: from La to Gd except Ce) system is concerned, we can empirically predict the second plateau pressure as well as the first plateau pressure from the unit cell volume of the alloys.

Composites of Al₂O₃-X (X: BaCrO₄, BaSO₄ and CaSO₄) systems were prepared by spark plasma sintering, and their friction and wear properties were evaluated from room temperature to 1073 K. With the addition of Na₂SiO₃, SiO₂ and Ag to the Al₂O₃-BaCrO₄ composites exhibited improved densification. In particular, the wear rates of the Al₂O₃-BaCrO₄ composites were notably reduced from room temperature to 1073 K with the addition of SiO₂. Al₂O₃-BaSO₄ composites, both with and without SiO₂ as a sintering additive, showed friction coefficients as low as 0.5 from room temperature to 1073 K, while the friction coefficients of the Al₂O₃-50CaSO₄-5SiO₂ (mass%) composites were as high as 1.2 at 473 K.

A near lamellar microstructure and two fine-grained fully lamellar (FGFL) microstructures in Ti-45Al-2Nb-2Mn+0.8 vol%TiB₂ alloy were prepared by selected heat treatments, and the fully lamellar microstructures were aged for stabilizing the lamellar plates. Microstructural examination and tensile creep tests at 760°C showed that the near lamellar microstructure possessed inferior creep resistance due to its coarse lamellar spacing and its larger amount of equiaxed γ grains at colony boundaries. The fine lamellar spacing as well as the fine lamellar colony size gave a major contribution to make the minimum creep rates smaller in the fully lamellar TiAl alloys. Since aging treatments stabilized the lamellar microstructures and delayed the degradation process during creep deformation, the aged samples exhibited lower minimum creep rate and longer creep life than the corresponding samples without the aging treatment. These results suggest that a fine as well as stabilized fully lamellar structure is a critical factor to improve the creep resistance of TiAl alloys in terms of short and long-term creep.

Magnetic field can affect the transformation temperature and microstructure if a transformed phase has different susceptibility with parent phase. Fe-C alloy is an ideal system to show the magnetic field effect since in this system, austenite (fcc structure) is a paramagnetic phase and ferrite (bcc structure) is a ferromagnetic phase below 770°C. In this paper, phase transformation temperature in Fe-C alloys in a magnetic field was measured from cooling curve. It was found that the transformation temperature for pure Fe from austenite to ferrite has a linear relationship with magnetic field strength, increasing about 0.8°C per unit of magnetic field of 1 T. For eutectoid transformation in Fe-0.8C alloy, similar relationship exists, the transformation temperature increases about 1.5°C per unit of magnetic field of 1 T. Experimental results do not agree well with that calculated by molecular field theory. An elongated and aligned microstructure by transformation in a high magnetic field was found in an Fe-0.4C alloy, but was not found in pure Fe and Fe-0.8C alloy.

This paper describes the effect of applied magnetic field on magnetic properties of Sm-Fe-N thick films prepared by the aerosol deposition (AD) method. At first, the magnetic field (0.17 T) was applied in the direction perpendicular to the film plane during the deposition. The remanence of Sm-Fe-N AD film deposited with the applied field, which was measured along the direction of the field, was 0.38 T. The remanence was smaller than that obtained without the applied field (0.40 T), which was considered to be due to lower film density. However, XRD analysis revealed that the ratio of X-ray peak intensities between (006) and (033) in the film deposited with the applied field was higher than that without the field and the c-axis of the Sm₂Fe₁₇N_x compound has a tendency to align along the direction of magnetic field. Secondly, the magnetic field (0—0.19 T) was applied in the direction parallel to the film plane during the deposition. The densities of the films were independent of the applied field. The remanences measured in the direction parallel to the applied field increased but those measured in perpendicular to the applied field decreased, with increasing the magnetic field during the deposition. The maximum value of remanence was 0.54 T, which was 29% higher than that without the applied field (0.42 T). Therefore, it is concluded that the c-axis of Sm-Fe-N AD films aligned along the direction of magnetic field during the deposition and the anisotropic feature increased with increasing the field.

The ⁷Li NMR Knight shift was measured for liquid Li-Tl alloys. Prior to this NMR measurement, DSC measurements were carried out to know the exact liquidus temperatures. Except for some trivial points, the phase diagram determined is almost same as the reported one; the solid compound at the maximum liquidus temperature was observed at 50 at% Tl; the eutectic point was detected at 84.8 at% Tl. The ⁷Li Knight shift, K, decreases rapidly with the addition of Tl up to 20 at% Tl. In the concentration range from 20 to 50 at% Tl, the K keeps almost constant value, which is 60% of ⁷Li Knight shift for the pure liquid Li. Such a decrease of the K is considered as an indication for the strong charge transfer from Li to Tl. The Zintl ion formation is expected in this concentration range of liquid Li-Tl alloys. These tendencies are similar to the previous studies for liquid Li-Ga and Li-In alloys. However, beyond 50 at% Tl, the K increases slightly and reaches to the constant value (70% of ⁷Li Knight shift for the pure liquid Li). Such a back donation of charge is absent for liquid Li-Ga and Li-In alloys. It is considered that the tendency of the Zintl ion formation for liquid Li-Tl alloys is slightly weaker compared with the cases of liquid Li-Ga and Li-In alloys.

Powders of La₂O₃ and ZrO₂ in ethanol based suspension were used as precursor materials for the wet mechanochemical (MC) synthesis of single phase La₂Zr₂O₇, a promising thermal barrier coating (TBC) material. Lanthania powder was first preheated at 1200°C prior to MC treatment in a planetary ball mill at 200, 300 and 400 revolutions per minute (rpm) milling speeds for 12, 18, and 24 hours using zirconia pot and balls. The slurries were then dried at 110°C for 24 hours in an oven followed by heat treatment at 1500°C for 1 hour. X-ray diffraction results showed that single phase La₂Zr₂O₇ was produced using 5 mm ball at 200 rpm milling for 12 hours and using 10 mm balls at 400 rpm for 24 hours. Smaller grinding balls and slower milling speed allowed more homogeneous mixing during wet mechanochemical treatment in a 250 mL milling pot. The average particle size of the products is lower than 1.8 μm. The yield of powder is greater than 98%.

Microstructures and tensile properties of a martensitic steel F82H (Fe-8Cr-2W-0.1C-0.2V-0.04Ta) were examined as a function of time and temperature of tempering. A heat treatment was performed at temperatures from 750 to 800°C for 0.5 h after the normalizing at 1040°C for 0.5 h. The tempering time at 750°C was varied from 0.5 to 10 h. The tensile specimens were irradiated at 250°C to a neutron dose of 1.9 dpa in the JMTR (Japan Materials Testing Reactor), and tensile tests were carried out at 250°C after the irradiation. The microstructures of the non-irradiated specimens were observed by a transmission electron microscope. The density of dislocations decreased with increasing time and temperature of the tempering, and the size of M₂₃C₆ carbide increased with it. While the yield stress of the non-irradiated specimens decreased with increasing time and temperature of tempering, the yield stress of the irradiated specimens tended to increase with increasing temperature of the tempering. The yield stress of the irradiated F82H steel changed from about 525 to 750 MPa and depended on the conditions of tempering treatment before irradiation.

Dependence of fracture properties and hardening was examined as a function of helium production in tensile specimens of a martensitic steel F82H (Fe-8Cr-2W-0.1C-0.04Ta) irradiated at 300°C to 2.3 dpa by neutron irradiation in the JMTR (Japan Materials Testing Reactor). The specimens used in this study were F82H, F82H+60 ppm¹¹B, F82H+30 ppm (¹¹B+¹⁰B) and F82H+60 ppm¹⁰B. The helium range produced from ¹⁰B (n,α)⁷Li reaction was from 5 to 330 appm in the specimens. The tensile testing was performed at 25°C. The radiation hardening due to helium production was detected at 330 appmHe. The degradation of fracture stress due to helium production was approximately evaluated from the fracture strength and the reduction area. Effect of specimen size on tensile and Charpy impact properties in F82H doped with 60 ppm boron and 200 ppm nitrogen was also examined. The JIS 14A and SS-J3 (Small Size—Japanese-3 type) were used for the tensile specimens, and half size (55 mm in length, 10 mm in height and 5 mm in width) and 0.5-1/3CVN (18 mm in length, 3.3 mm in height and 1.65 mm in width) were used for the Charpy impact testing. The tensile properties were a similar to each other. However, the ductile-brittle transition temperature measured in smaller size specimen was somewhat lower than that in the standard size specimen.

Strained Si has been attracting attention as a new material that has high carrier mobility. Such strained Si can be produced by epitaxial growth on strain-relaxed SiGe grown on a Si substrate, because SiGe has a larger lattice constant than Si. It is important to fabricate a highly relaxed SiGe buffer layer. We consider that defect distribution can be controlled and that the highly relaxed SiGe buffer can be prepared by ion implantation into the substrate. We carried out ion implantation under several conditions. Then, Si_0.7Ge_0.3 films with a thickness of 100 nm were grown at 650°C on the implanted substrates by the solid-source molecular beam epitaxy method (MBE). The strain relaxation of SiGe films was estimated by Raman spectroscopy. The crystallinity was observed by transmission electron microscopy. In the case of an implantation energy of 50 keV and an ion dose of 5 × 10¹³ cm⁻², the dislocation density was low and the relaxation was lower than 50%. In the case of 1 × 10¹⁵ cm⁻² or a higher ion dose, the SiGe film became partially polycrystalline. In contrast, we succeeded in forming an approximately 80%-relaxed single-crystal SiGe thin film when the ion dose was 5 × 10¹⁴ cm⁻². This means that ion implantation into the substrate before MBE growth is a good method of controlling defect introduction and producing relaxed single-crystal SiGe thin films.

This paper describes the behavior of hydrogen in iron chromium nickel alloys. The behavior of hydrogen was investigated by radioisotope of the hydrogen. The traces of hydrogen in the iron chromium nickel alloys were investigated by measuring the β-ray radiated from tritium. The β-ray can be identified by an auto radiograph method (ARG) and radiation detector. When the ARG method was used, the tritium was charged to the specimen of iron chromium nickel alloys. An emulsion was coated onto specimen of the iron chromium nickel alloys, which charged the tritium, and then the emulsion was exposed to β-ray emitted from tritium captured by iron chromium nickel alloys. The trace of the β-ray appeared as a black image on the emulsion film. It was noted that there were two different black images on the emulsion film. One was an Ag cluster, generated by the β-ray, and the other was stain generated at the development of emulsion. These images were investigated by energy dispersive X-ray spectroscope (EDX) and high-resolution electron microscope (HREM). Using these techniques, captured tritium quantity in the iron chromium nickl alloys was estimated by the analyzed data of EDX and dose of radiation of β-ray.

We report synthesis of zinc sulfide nanocrystals (NCs) via formation of polymetallic thiolate cages. Nearly monodisperse ZnS NCs with size ranging from 2.2 to 7 nm were obtained by thermolysis of S-Zn-dodecanethiol precursors. The electron diffraction pattern of zinc sulfide NCs indicates that precipitates are wurtzite or mixture of wurtzite and zincblende. TEM observation and UV-vis spectra reveal that the growth rate of ZnS NCs considerably depends on the annealing temperature. UV-vis spectra of ZnS NCs with size smaller than 3 nm show sharp excitonic features and a large blue shift from the bulk material. The photoluminescence spectra exhibit large red shift from the absorption band edges, being attributed to electron-hole recombination by surface traps. The narrow size distribution of ZnS NCs leads to formation of ordered self-assemblies with various well-defined structures, where non-closed-packing structure is predominant.

In order to obtain a sharply textured silver sheet with a single orientation as a substrate, which is favorable for high temperature superconductor (HTS) film having high J_c (critical current density), the sheet metallurgical processing of pre-heated rolling combined with two step annealing treatment has been performed. We used two kinds of starting material, which are silver ingots of commercial purity obtained by casting in air and vacuum, to examine the effect of oxygen on texture development. The main feature of warm rolling texture obtained in this study was a strong Brass {011}‹211› component with minor S {123}‹412› component, and in some cases, cube {001}‹100› component or Copper {112}‹111› component appears also depending on the warm rolling procedures. Upon crystallization, it has been found that cube {001}‹100›, {124}‹4, 12, 7› and {13, 6, 15}‹365› orientations were formed as the dominant components in silver sheets and the relative amount of their orientation components depended on the concrete annealing conditions applied after the warm rolling and on the oxygen content. A very sharp single-crystal like cube texture has been successfully obtained in the specimen, which was cast “in vacuum”, warm rolled at 95 percent and subsequently annealed 10 minutes at 150°C and 30 minutes at 500°C in nitrogen. Finally, we discussed the technological basis on sharp cube texture formation in fcc pure metals with low stacking fault energy.

Tungsten (W) is a candidate material for plasma-facing components in fusion reactors. Large amounts of solid transmuted elements of W such as Re and Os will be produced under fusion neutron irradiation. In order to investigate the effects of transmutation products, a series of W-Re-Os alloys were fabricated by an arc melting method. The nominal compositions of these samples were selected based on theoretical predictions of changes in chemical composition for the transmutation of W to Re and Os. Significant hardening, an increase in electrical resistivity and changes in the lattice constant were observed as a function of the Re and Os content. Microstructural observations revealed the existence of the σ phase precipitate, which did not significantly affect physical properties and hardness compared with solid solution effect. The main objectives of this work were to demonstrate the processes of material fabrication and to study unirradiated material properties.

Change in cohesive energies in a non-stoichiometric Ni₉Mn₃Ga₄ alloy was calculated as a function of tetragonality, c/a, and compared with that in a stoichiometric alloy. A dip around c/a ≅ 0.97, which is seen in a stoichiometric alloy, disappears in the non-stoichiometric alloy, and a dip around c/a ≅ 1.23 become deeper than that in the stoichiometric alloy. These results are in good agreement with the influence of Ni concentrations on c/a in Ni-Mn-Ga alloys.

The microstructure of friction stir welded 1050 aluminum and 6061 aluminum alloy was observed by a metallographic technique, transmission electron microscopy, electron backscatter diffraction pattern and optical microscopy. In the stir weld zone, the microstructure of welded 1050 aluminum was significantly different from that of the welded 6061 aluminum alloy. In the case of welding 1050 aluminum, there was a comparatively uniform microstructure in the stir weld zone, not a wedge-shaped microstructure formed in a stir welded 6061 aluminum alloy. EBSD indicated that there was almost same fraction of low angle boundaries among the thermo mechanically affected zone in welded 1050 aluminum and 6061 aluminum alloy. In the stir weld zone, equiaxed grains were created and the grain size of 1050 aluminum was little larger than that of 6061 aluminum alloy, suggesting that precipitates pinning effect affects the dynamic recrystallization in the stir weld zone.

The ceramic composites consisting of titanium nitride and aluminium nitride were prepared by mechanical alloying (MA) of AlN, Ti and ammonium carbonate. Comparing with conventional way, which used AlN and TiN powders as the initial materials, in this study, nearly full densification can be achieved by hot pressing or by spark plasma sintering at much lower temperatures. XRD analysis indicated that besides AlN and TiN, Al₂O₃ is also present after sintering. SEM observations showed that the average grain size of the sintered compacts is lower than 0.5 μm. The sample which had been subjected to MA using steel grinding medium for 52 h exhibited Vickers hardness of 18.9 GPa, which is comparable to the value that reported by other researchers.

The crystal structure of δ-phase in the Sb-Te binary system was investigated by high-resolution transmission electron microscopy. The observed high-resolution images could be described by a simple rhombohedral structure with the lattice displacement wave (LDW). The wavelength of LDW was slightly longer than double the (111) plane in the rhombohedral structure, and increased slightly with increasing Te concentration.

The beneficial aspects of intragranular ferrite formation on mechanical properties of welds have been reported in the literature for decades. In recent years, this concept has been successfully extended to medium carbon forging steel to refine the microstructure and optimise ductility and toughness. The aim of this work is to demonstrate that intragranular formation of ferrite could be enhanced by increasing the austenite grain size and/or optimising the nature of the inclusions. In this sense, the isothermal decomposition of austenite in allotriomorphic and idiomorphic ferrite for two medium carbon steels microalloyed with vanadium and titanium have been studied. The experimental results reported in this work allows to conclude that austenite grain size and the nucleation potency of inclusions are two parameters that should be considered linked to promote the full decomposition of austenite into intragranularly nucleated ferrite.

The gradual nanocrystallization and amorphization mechanisms in various Zr-X alloys during accumulative roll bonding (ARB) are explored. The effects of strain accumulation, the relative initial hardness of the elemental foils, the enhanced diffusion, and the critical nano size for the sudden transformation from the nanocrystalline phase to the amorphous state are examined. For elemental foils with compatible initial hardness, the nanocrystallization and amorphization rates appear to be higher. The estimated diffusion rates during ARB are higher by several orders of magnitude than the lattice diffusion in bulk materials. When the nano grains are refined down to around 3 nm, sudden transformation into the amorphous phase would occur.

The effect of In substitution for Ni on the glass forming ability has been studied in Cu₄₇Ti₃₃Zr₁₁Ni_8−xIn_xSi₁ (x = 0, 2, 4, 6, 8) alloys by using thermal analysis and X-ray diffractometry. Partial substitution of Ni by In in Cu₄₇Ti₃₃Zr₁₁Ni₈Si₁ promotes the glass forming ability. Cu₄₇Ti₃₃Zr₁₁Ni₆In₂Si₁ bulk metallic glass with diameter of 6 mm can successfully be fabricated by Cu-mold injection casting method. ΔT_x and K parameter show a good relationship with the measured maximum diameter of the amorphous specimen.

A₂O₃/YAG/ZrO₂ eutectic Melt-Growth-Composite (MGC) rods with two different microstructures were prepared by unidirectional solidification using the modified-pulling-down method (MPD) under different processing parameters. Microstructure and crystallographic texture were investigated by X-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM) and electron backscattered pattern (EBSP) method. High-temperature strength was evaluated by compression tests at 1773 K and 1873 K. Geometric pattern structure and Chinese script pattern structure are evolved by controlling processing parameters. MPD rods have strong preferred growing orientations in Al₂O₃ of ‹001› for the geometric pattern structure and of ‹300› for Chinese script pattern structure. Constituent phases in the MPD rod hold the orientation relationship. The yield stress for the geometric pattern structure is over 1 GPa at 1773 K, which is extremely higher than that for Chinese script pattern structure. High-temperature strength at 1773 K and 1873 K depends on strain rate and temperature in both the MPD rods.

Cold model experiments were carried out to effectively produce spherical solar cells made of single crystal silicon. Water was used as the working fluid. A single-hole nozzle was chosen to generate water droplets in the dripping mode. The wettability of the nozzle was changed by coating repellent on it. The diameter of a water droplet thus generated depended strongly on the wettability. The flow field in a droplet was visualized with a CCD camera and the velocity vectors were determined with particle image velocimetry. The flow pattern in the droplet was correlated based on the Weber number similitude. A Weber number range suitable for producing single crystal silicon of high quality was found.

The solubility of MgO in molten MgCl₂ and CaCl₂ has been studied. In pure MgCl₂, the MgO solubility was determined to be 0.63—2.90 mol% at 1073—1373 K, while 0.24—0.63 mol% in pure CaCl₂ at 1223—1523 K. The effects of MgCl₂ and CaO addition to CaCl₂ were investigated, and the solubility product of MgO was found to increase with the amount of MgCl₂ addition, while CaO addition did not affect significantly. They were explained in terms of the activity coefficient of oxide ion (O²⁻) as well as the enthalpy of mixing for the MgO-CaCl₂ system. From the experimental results, the corrosion possibility of MgO refractory by Cl-containing gas was considered.

The behavior of oxygen adsorption on the surface of liquid Cu-Ag alloys was investigated by measuring their surface tension (σ) with the sessile drop method in the oxygen partial pressures (p_o2) between 2.5 × 10⁻¹¹ and 2.5 × 10⁻³ Pa. The oxygen adsorption (the surface excess concentration of oxygen) was calculated from the slope of dσ/d ln p_o2 by applying Gibbs adsorption isotherm, for liquid Cu, Cu-5 at%Ag, Cu-10 at%Ag, Cu-20 at%Ag and Ag. It was found that the oxygen adsorption increased with the oxygen partial pressure, up to saturation on the surface. The oxygen adsorption on the surface of liquid Cu-20 at%Ag alloys exhibited almost the same behavior as that of pure liquid Ag, because surface saturation was not achieved for the Cu-20 at%Ag alloys, even at high oxygen partial pressures. Thermodynamic calculations using Butler's model indicated that the mole fraction of Ag in the surface of liquid Cu-Ag alloys drastically increases to 0.81 when the mole fraction of Ag in the bulk is only 0.2. Thus, it is considered that the outermost surface of liquid Cu-20 at%Ag alloys contains an enhanced level of Ag, which determines the oxygen adsorption behavior on liquid Cu-20 at%Ag alloys.

Two kinds of solidification paths from Y₃Al₅O₁₂ melt has been reported; one is stable Y₃Al₅O₁₂ (YAG) garnet, the other is metastable YAlO₃ perovskite (YAP) and subsequent YAP+Al₂O₃ eutectic. The reason for this, however, has been puzzled. The effect of cooling rate on this phase selection was addressed under containerless condition using an aero-acoustic levitator. A high-speed video camera (HSV) enabled us to directly observe the recalescence behavior. As the cooling rate increased from 15 to 350 K/s, the solidification of a metastable YAP and YAP+Al₂O₃ eutectic, a monophasic YAG, and an amorphous phase were successively obtained. At around the critical cooling rate of approximately 50 K/s for the formation of YAP and YAG, simultaneous recalescence of YAP and YAG was observed by HSV, and the sample obtained contained both the metastable YAP and stable YAG. The nucleation rate of YAG corresponds with that of YAP at the critical cooling rate and the growth velocity of YAP, which first nucleated in the undercooled melt, was slow enough for YAG to nucleate in the remaining undercooled melt, resulting the simultaneous recalescence. In general, the metastable phase nucleates at the higher cooling rate than the stable phase. However, in this system, the higher nucleation barrier of YAG than that of YAP led to the nucleation of YAG at the higher cooling rate.

The thermal and metallographic analysis of Inconel718 alloy revealed that the solidification proceeds in the order of primary γ at 1615 K, (γ + NbC) eutectic at 1561 K, and (γ + Ni₂Nb) eutectic at 1452 K. Additional equilibrium evaluation performed using Thermo-Calc predicts the following solidification sequences: primary γ at 1633 K, followed by (γ + NbC) eutectic at 1555 K. The (γ + Ni₂Nb) eutectic phase does not appear in the equilibrium calculation, indicating that the (γ + Ni₂Nb) phase crystallizes as a non-equilibrium eutectic phase. The partition coefficients of alloying elements in the primary γ and eutectic phases were also determined experimentally (k_(Exp.)) and compared with the values calculated using Thermo-Calc (k_(T-C)). The behaviors of alloying elements during solidification were estimated using k_(Exp.) and k_(T-C) in various solidification models. It was found that the results of the experiments are in agreement with the calculations obtained using the Clyne-Kurz model.

This paper describes the microstructure and the formation mechanism of solid-state diffusion bonded interfaces of silicon carbide (SiC) and titanium aluminide (TiAl). Two SiC specimens were diffusion bonded using a Ti-48 at%Al foil in vacuum. The interfacial microstructure has been investigated by means of scanning electron microscopy, electron probe microanalysis and X-ray diffraction. Four layers of reaction products are formed at the interface by diffusion bonding: a layer of TiC adjacent to SiC followed by a diphase layer of TiC+Ti₂AlC, a layer of Ti₅Si₃C_X containing Ti₂AlC particles and a layer of TiAl₂. However, the TiAl₂ layer is formed during cooling. The actual phase sequence at the bonding temperatures of 1573 K and 1673 K are SiC/TiC/(TiC+Ti₂AlC)/(Ti₅Si₃C_X+Ti₂AlC)/Ti_1−XAl_1+X/TiAl and SiC/TiC/(TiC+Ti₂AlC)/(Ti₅Si₃C_X+Ti₂AlC)/Ti₅Al₁₁/Ti_1−XAl_1+X/TiAl, respectively. Ti₅Al₁₁ and Ti_1−XAl_1+X rapidly transform to TiAl₂ during cooling. The layers grow obeying the parabolic law. The growth rate of the reaction layers change sensitively depending on the bonding temperature.

Increase of Si content increases wear resistant properties of hypereutectic Al-Si alloys. However, large primary Si phase is inevitable in Al-Si alloy when the alloy is produced by conventional casting processes, which deteriorate its machinability as well as its mechanical properties. The objective of this study is to produce graded Al/Al-Si nanocomposite coating onto an A1050 Al alloy substrate by mixing Al and Si nanoparticles with Supersonic Free-Jet PVD (SFJ-PVD). The SFJ-PVD has been developed as a new coating method in which a coating film is formed by depositing nanoparticles with very high velocity onto a substrate. This SFJ-PVD provides a high deposition rate and produces a mixture of different kinds of nanoparticles formed in different evaporation chambers on the substrate. The graded Al/Al-Si coating film is produced by depositing Al and Si nanoparticles formed in different evaporation chambers with the controlled evaporation rates of Al and Si respectively. A smooth, compact and defect-free microstructure is formed both at the interface between the substrate and the coating film and inside the coating film. The graded Al-Si coating film has very fine, varying from 10 to 20 nm in diameter, Si phases in Al matrix. It is confirmed with nano-indentation hardness tester that the graded Al/Al-Si coating film on A1050 substrate has graded hardness from 0.65 to 5.9 GPa corresponding to the gradual compositional change of Si up to Al-57.8 at%Si.

Rapid prototyping technology offers an ideal method for manufacturing ceramic workpieces. Not only can it produce parts without tooling, it poses no limitation in shape and complexity of the parts to be fabricated. This article aims to study the manufacturing technologies of a new rapid prototyping process named Ceramic Laser Fusion. A self-developed rapid prototyping system is employed to fabricate complex ceramic parts using slurry composed of silica powder as basic material, inorganic fire-resistant binder as additive, and water as solvent. Three-dimensional parts can be made by repeatedly executing the single-layer generating cycle which includes slurry feeding, layer paving, layer drying, remnants cleaning, platform descending and laser scanning. The laser fusion experiment shows that hollow ceramic pump impeller, complicated ceramic fan and turbine blades as well as ceramic molds for metal casting can be made at a production rate of 32 cm³/h. Ceramic Laser Fusion is capable of paving very thin layers; green portion provides solid-state support to prevent deformation of workpieces, and can be removed by water or sodium hydroxide solvent. Further advances in stacking thinner layers will lead to the production of parts with fine details and higher precision.

We have found a large magnetostriction of 1 × 10⁻³ for FSMA (ferromagnetic shape memory alloy) Fe-29.6 at%Pd ribbon of 70 μm thickness prepared by rapidly solidified melt-spinning. The ribbon contains FCT martensite twins at room temperature. The magnetic field induces a strain caused by conversion of variants in the FCT martensitic phase. This strain and shape recovery ratio increase with temperature and the former reaches a maximum value at 380—400 K, though FCT-FCC transformation temperature of the single crystal is 293—307 K. In order to confirm of this phenomenon, we investigated the phase transformation behavior of the ribbons by DSC, magnetization measurement and an acoustic elastometer method. During heating and cooling, the modulus defect ΔM/M and damping Q⁻¹ of the as-spun ribbon sample exhibit tow-step (330—350 K and 400—420 K) phase transformation. From these results, it is found that the low-temperature FCT-FCC transformation is consistent with the value obtained from a single crystal of Fe-29.6 at%Pd. On the other hand, the high-temperature transformation, which is due to local residual strains, is peculiar to rapidly solidified ribbon.

A rapid solidification process was found to form unidirectional crystal structures in (Bi,Sb)₂Te₃ and (Bi,Sb)₂(Te,Se)₃ based thermoelectric alloys. This fine microstructure with its unidirectional crystal orientation is expected to yield high value thermoelectric properties because the fine grain size is extremely effective in decreasing thermal conductivity by boundary scattering of phonons and because this crystal orientation is extremely effective in decreasing the electrical resistivity (increasing the carrier mobility). The rapidly solidified alloys were prepared by a single-roller liquid quenching method. These alloys were shaped as a thin foil with thicknesses ranging from 5 to 20 μm. The crystal orientation was analyzed by the x-ray diffraction method and the microstructure was observed by optical microscopy and SEM of the thin foil samples. The crystal grains of the rapidly solidified foils were very fine and highly oriented. The thermoelectric properties were measured for p-type Bi_0.4Sb_1.6Te₃ and the n-type Bi_1.9Sb_0.1Te_2.6Se_0.4 compacted alloys. The figures of merit of 3.5 × 10⁻³ K⁻¹ for p-type alloy and 3.3 × 10⁻³ K⁻¹ for n-type alloy were obtains. It is proposed that the rapid solidification is very useful technique for the improvement of thermoelectric properties.

Porous glassy Pd_42.5Cu₃₀Ni_7.5P₂₀ alloy rods of 7 mm in diameter and 50 mm in length were produced by melting the Pd-Cu-Ni-P alloy for 600 s at 853 K in a hydrogen pressure of 1.5 MPa and then water quenching. The pores have a nearly spherical shape and their average size is approximately 200 μm. The volume fraction of the pores was changed in the range of 0.36 to 0.64. The glass transition temperature and crystallization temperature of the porous alloy are 577 and 671 K, respectively, in agreement with those of the pore-free alloy. The porous alloy rod exhibits unique mechanical properties with the features of much higher ductility and flexibility which are significantly different from those for the pore-free alloy rod. The unique mechanical characteristics indicate the possibility of future uses as a new type of structural and functional materials.