In this study, factors considered to be causes of promotion of densification of sintered pellets identified during phase change are reviewed. As a result, conclusions shown below are obtained for each factor. In order for MA powder to soften, a temperature of 1,000 K or higher is required. In order to confirm the temporary increase in density throughout the sintered pellet, the temperature rise due to heat during phase change was found not to have a significant effect. While examining the thermal expansion using the compressed powder, which stopped densification at a temperature below the MA powder itself, and the phase change temperature, no shrinkage phenomenon contributing to the promotion of densification is observed. The two types of powder made of Ti-silicide through heat treatment are densified only in the high temperature region of 1,000 K or more; it can be estimated that this is the effect of fine grain superplasticity. In the densification of the amorphous powder, the dependence of sintering pressure and the rate of temperature increase are shown. It is thought that the specific densification behavior identified during the phase change of the Ti-37.5 mol.%Si composition MA powder reviewed in this study is the result of the acceleration of the powder deformation by the phase change from non-equilibrium phase to equilibrium phase.

The composition of martensite transformation in NiAl alloy is determined using pure nickel and aluminum powder by vacuum hot press powder metallurgy, which is a composition of martensitic transformation, and the characteristics of martensitic transformation and microstructure of sintered NiAl alloys are investigated. The produced sintered alloys are presintered and hot pressed in vacuum; after homogenizing heat treatment at 1,273 K for 86.4 ks, they are water-cooled to produce NiAl sintered alloys having relative density of 99 % or more. As a result of observations of the microstructure of the sintered NiAl alloy specimens quenched in ice water after homogenization treatment at 1,273 K, it is found that specimens of all compositions consisted of two phases and voids. In addition, it is found that martensite transformation did not occur because surface fluctuation shapes did not appear inside the crystal grains with quenching at 1,273 K. As a result of examining the relationship between the density and composition after martensitic transformation of the sintered alloys, the density after transformation is found to have increased by about 1 % compared to before the transformation. As a result of examining the relationship between the hardness (Hv) at room temperature and the composition of the matrix phase and the martensite phase, the hardness of the martensite phase is found to be smaller than that of the matrix phase. As a result of examining the relationship between the temperature at which the shape recovery is completed by heating and the composition, the shape recovery temperature is found to decrease almost linearly as the Al concentration increases, and the gradient is about -160 K/ at% Al. After quenching the sintered NiAl alloys of the 37 at%Al into martensite, specimens fractured by three-point bending at room temperature are observed by SEM and, as a result, some grain boundary fractures are observed on the fracture surface, and mainly intergranular cleavage fractures.

Titanium aluminides have attracted special interest as light-weight/high-temperature materials for structural applications. The major problem limiting practical use of these compounds is their poor ductility and formability. The powder metallurgy processing route has been an attractive alternative for such materials. A mixture of Ti and Al elemental powders was fabricated to a mechanical alloying process. The processed powder was hot pressed in a vacuum, and a fully densified compact with ultra-fine grain structure consisting of Ti3Al intermetallic compound was obtained. During the compressive deformation of the compact at 1173 K, typical dynamic recrystallization (DR), which introduces a certain extent of grain refinement, was observed. The compact had high density and consisted of an ultra-fine equiaxial grain structure. Average grain diameter was 1.5 μm. Typical TEM micrographs depicting the internal structure of the specimen deformed to 0.09 true strain are provided, in which it can be seen that many small recrystallized grains having no apparent dislocation structure are generated at grain boundaries where well-developed dislocations with high density are observed in the neighboring grains. The compact showed a large m-value such as 0.44 at 1173 K. Moreover, the grain structure remained equiaxed during deformation at this temperature. Therefore, the compressive deformation of the compact was presumed to progress by superplastic flow, primarily controlled by DR.

In order to observe the microstructure and morphology of porous titanium -oxide thin film, deposition is performed under a higher Ar gas pressure than is used in the general titanium thin film production method. Black titanium thin film is deposited on stainless steel wire and Cu thin plate at a pressure of about 12 Pa, but lustrous thin film is deposited at lower pressure. The black titanium thin film has a larger apparent thickness than that of the glossy thin film. As a result of scanning electron microscope observation, it is seen that the black thin film has an extremely porous structure and consists of a separated column with periodic step differences on the sides. In this configuration, due to the shadowing effect, the nuclei formed on the substrate periodically grow to form a step. The surface area of the black thin film on the Cu thin plate changes with the bias potential. It has been found that the bias of the small negative is effective in increasing the surface area of the black titanium thin film. These results suggest that porous titanium-oxide thin film can be fabricated by applying the appropriate oxidation process to black titanium thin film composed of separated columns.

In order to develop a process for manufacturing a composite structure of an intermetallic compound foam and a hollow material, the firing and pore form of the Al-Ni precursor in a steel pipe are investigated. When the Al-Ni precursor is foamed in a hollow pipe, if the temperature distribution inside the precursor is uneven, the pore shape distribution becomes uneven. In free foaming, no anisotropy is observed in the foaming direction and the pore shape is isotropic. However, in the hollow pipe, the pipe expands in the pipe axis direction and fills the pipe. The interfacial adhesion between Al3Ni foam and steel pipe is excellent, and interfacial pore and reaction layer are not observed by SEM. In free foaming, the porosity is 90 %, but it decreases to about 80 % in the foam in the pipe. In the pipe foaming, most of the pore shape appears elongated in the pipe direction in the vicinity of the pipe, and this tendency is more remarkable when the inside pipe diameter is small. It can be seen that the pore size of the foam sample in the pipe is larger than that of free foam, because coarse pores remain after solidification of the foam because the shape of the foam is supported by the pipe. The vertical/horizontal length ratio expands along the pipe axis direction by foaming in the pipe, and therefore circularity is reduced.