Table of contents

The Ministry of Education, Culture, Sports, Science and Technology of Japan started the Priority Assistance for the Formation of Worldwide Renowned Centers of Research – Global COE Program. This program is based on the competitive principle where a third party evaluation decides which program to support and to give priority support to the formation of world-class centers of research. Our program Center of Excellence for Advanced Structural and Functional Materials Design was selected as one of 13 programs in the field of Chemistry and Materials Science. This center is composed of two materials-related Departments in the Graduate School of Engineering: Materials and Manufacturing Science and Adaptive Machine Systems, and 4 Research Institutes: Center for Atomic and Molecular Technologies, Welding and Joining Research Institute, Institute of Scientific and Industrial Research and Research Center for Ultra-High Voltage Electron Microscopy.

Recently, materials research, particularly that of metallic materials, has specialized only in individual elemental characteristics and narrow specialty fields, and there is a feeling that the original role of materials research has been forgotten. The 6 educational and research organizations which make up the COE program cooperatively try to develop new advanced structural and functional materials and achieve technological breakthrough for their fabrication processes from electronic, atomic, microstructural and morphological standpoints, focusing on their design and application: development of high performance structural materials such as space plane and turbine blades operating under a severe environment, new fabrication and assembling methods for electronic devices, development of evaluation technique for materials reliability, and development of new biomaterials for regeneration of biological hard tissues.

The aim of this international conference was to report the scientific progress in our Global COE program and also to discuss related research topics. The organizing committee gratefully thanks participants for presenting their recent results and for discussions with our COE members and international attendees.

November 2008

Professor Tomoyuki Kakeshita Chairman of the Conference Vice Dean, Graduate School of Engineering, Osaka University, Division of Materials and Manufacturing Science, Graduate School of Engineering Leader of Global COE Program, Osaka University, ''Center of Excellence for Advanced Structural and Functional Materials Design''

Organization

Chairman: T Kakeshita (Osaka University)

Advisory Board: H Mehrer (University Münster, Germany), E K H Salje (University of Cambridge, United Kingdom), H-E Schaefer (University of Stuttgart, Germany), P Veyssiere (CNRS-ONERA, France)

Organizing Committee: T Kakeshita, H Araki, H Fujii, S Fujimoto, Y Fujiwara, A Hirose, S Kirihara, M Mochizuki, H Mori, T Nagase, H Nakajima, T Nakano, R Nakatani, K Nogi, Y Setsuhara, Y Shiratsuchi, T Tanaka, T Terai, H Tsuchiya, N Tsuji, H Utsunomiya, H Yasuda, H Yasuda (Osaka University)

Executive Committee: T Kakeshita, S Fujimoto, Y Fujiwara, A Hirose, T Tanaka, H Yasuda (Osaka University)

Conference Secretariat: Y Fujiwara (Osaka University)

Proceedings Editors: T Kakeshita and Y Fujiwara (Osaka University)

All papers published in this volume of Journal of Physics: Conference Series have been peer reviewed through processes administered by the proceedings Editors. Reviews were conducted by expert referees to the professional and scientific standards expected of a proceedings journal published by IOP Publishing.

Al alloy sheets were deformed by ECAP(equal channel angular pressing), asymmetry and frictional rolling without lubrication and subsequent heat treating, the textures and r-values of the samples measured by x-ray diffractometer and tensile tester, respectively. The r and |Δr|-values variations after ECAP, asymmetry rolling and frictional rolling without lubricant process and subsequent heat treating of Al alloy sheets are compared and analyzed by using the changes of inverse pole figures and f(g) values of ODFs.

Ruthenium (Ru) has been added to the latest 4th Generation Ni-base superalloys to improve phase stability and modify creep life. Various coatings are routinely applied to these advanced alloys to protect the turbine blade at elevated temperature, however, this creates several problems such as the precipitation of brittle Topologically Close-Packed (TCP) phases and the formation of Secondary Reaction Zones (SRZ). The SRZ forms under the plat-aluminized coating of turbine blades and consists of γ, γ and TCP phases growing into substrate by the migration of high-angle grain boundaries. Surface residual stress and chemical super-saturation of alloying elements are associated to SRZ formation. In the thin sections of high-pressure turbine blades this is critical in determining blade performance and longevity. It is essential to know how Ru additions affect coating and SRZ morphologies during exposure. In this study, we focus on the effects of three variables on the SRZ formation: Ru concentration, alloy composition in Ru-containing alloys and surface finish. A series of Platinum-Aluminised superalloys containing 2–5wt% Ru and having various surface finishes was studied after isothermal exposure at 1100°C for up to 500h. The alloys were classified into two groups by their distinctive SRZ morphology. At the lowest Ru levels sporadic formation of SRZ was observed, whilst a continuous SRZ was formed in the higher Ru alloys. EBSD analysis revealed that the latter group have a higher nucleation rate of individual SRZ grains and also showed more rapid SRZ growth. The precipitation of TCPs in the substrate also inhibited the growth of the SRZ towards the end of the exposure further reducing the penetration of the SRZ into the substrate. It is concluded that Ru-additions to Ni-base superalloys are effective in impeding TCP phase formation in the substrate, but increase both the extent and the rate of SRZ formation beneath coating.

Microstructure and mechanical properties of an electrodeposited nanocrystalline Ni (nc-Ni) were investigated. Electrodeposition was done on two different kinds of Cu substrates. One was a Cu sheet having coarse grains, and the other was a sheet having a nanocrystalline Cu (nc-Cu) layer on the surface. The substrates were chemically removed after the electrodeposition of Ni. Then, sheets of fully dense nc-Ni with a thickness of approximately 100 μm were obtained. The surface of the sheets which had touched the substrate showed a mat surface appearance, while the other surface was mirror-like. In the nc-Ni electrodeposited on the coarse grained Cu substrate, the mirror-like surface showed homogeneous equiaxed grain structure, while the mat surface showed inhomogeneous structure. In the nc-Ni electrodeposited on the nc-Cu substrate, the microstructure of the mirror-like surface was similar to that in the nc-Ni deposited on the coarse grained Cu. However, the mat surface showed the homogenous structure, different from that in the nc-Ni deposited on the coarse grained Cu. This result indicates that the grain size and homogeneity of the microstructure of nc-Ni sheets were affected by the microstructure of the substrate. The electrodeposited nc-Ni performed strength four times higher than that of the conventionally coarse-grained pure Ni with grain size of 28 μm. In the tensile test, the total elongation of the nc-Ni electrodeposited on the coarse grained Cu was smaller than that of the nc-Ni electrodeposited on the nc-Cu. It is suggested that the homogeneity of microstructure significantly affects the mechanical properties of nc-Ni.

The control of microstructure in Ni₃V single crystals such as variant and anti-phase boundary (APD) was attempted by quenching from the disordered state followed by annealing at several temperatures. In the heat-treatments, the microstructure strongly varied depending on the quenching speed from the disordered state. In slow-quenching, the lamellar structure composed of two variants was developed after annealing, as reported in many polycrystalline samples. However, only one of three variants was preferentially grown in the specimen rapidly quenched from the disordered state followed by annealing. The yield stress of slow-quenched specimen showed more than twice the value of the fast-quenched specimen.

It has been proposed by Marian et al. [1] that a [001] interstitial-type dislocation loop can be formed in body-centered cubic iron via the collision between a 1/2[111] loop and 1/ 2[111] loop, which undergo one-dimensional glide diffusion, and the subsequent shear reaction. However, the formation of [001] loops through this reaction has not been reproduced by other works even though the two 1/2<111> loops collided with each other. In the present paper, the origin of the difficulty in this reaction is discussed within the framework of isotropic elasticity theory. The sign of the driving force for the reaction is heavily dependent on the reaction path. The two 1/2<111> loops colliding to form a [110] junction can transform to a single [001] loop when a shear loop generated within the 1/2[111] loop propagates in sync with the other shear loop within the 1/ 2[111] loop. However, unsynchronized motion of the two shear loops significantly suppresses the propagation of the shear loops, which might be caused by the thermal fluctuation at finite temperatures. This will be one of the origins of the difficulty in the formation of [001] loops through the collision between the two 1/2<111> loops.

Unidirectionally solidified Al₂O₃-YAG(Y₃Al₅O₁₂: yttrium-aluminum-garnet) eutectic ceramic composites have been recognized as encouraging heat-resistance materials because of the superior mechanical properties at high temperatures. In addition to the excellent mechanical properties at high temperatures, some interesting solidification phenomena have been reported in the Al₂O₃-Y₂O₃ system. The Al₂O₃-YAG equilibrium eutectic at 2099 K and the Al₂O₃-YAP metastable eutectic at 1975 K exist in the Al₂O₃-Y₂O₃ system. The heating the metastable eutectic up to temperatures above the metastable eutectic temperature produced the undercooled melt. Solidification in the equilibrium path accompanied the melting of the metastable eutectic. The solidification process using undercooled melt resulted in the fine and uniform eutectic structure. In this study, the effect of the initial Al₂O₃-YAP particles size on the undercooled melt formation was examined. The Al₂O₃-YAP particles with diameters more than several μm resulted in the transformation through the undercooled melt. EBSD analysis showed that the domains of Al₂O₃ grains with same crystallographic orientation were observed and that their domain size depended on the Al₂O₃-YAP particles size. On the other hand, for the Al₂O₃-YAP particles with a diameter of 500 nm, the each Al₂O₃ grain with diameter of about 1 μm had the different crystallographic orientations, which suggested that the transformation from metastable eutectic to equilibrium eutectic occurred in the solid state. The increase in the Al₂O₃-YAP free surface area suppressed the undercooled melt formation.

In the present work, we report the formation of a porous oxide/hydroxide surface layer on the Ti-29Nb-13Ta-4.6Zr (TNTZ) alloy achieved by the combination of an alkali immersion and a potential pulse polarisation process. The alkali treatment has been employed for pure titanium to produce amorphous and porous layer prior to hydroxyapatite (HAp) growth. But, in the case of TNTZ, immersion in 5M NaOH at the open circuit potential (OCP) at 60°C for 24 hours, did not yield any uniform layer, instead a thick deposited layer with highly cracked one. The cracks were attributed to the growth of a tantalum enriched particulate. In order to avoid the crack formation, the electrochemical behaviour of the alloy and the pure alloying elements (Ti, Nb, Ta and Zr) was investigated to produce a uniform surface with the application of a square wave modulated potential pulse polarization, leading to the formation of a relatively uniform porous layer on the alloy.

MgO coating was formed on nickel superalloy substrate. The coating process consisted of two steps: cathodic electrolytic deposition of MgOH, and heat treatment. The heat treatment was necessary to calcinate and to sinter the deposit. The deposit morphology and composition, before and after heat treatment, were studied by X-ray diffraction and the scanning electron microscopy. The influence of deposition parameters on deposition yields and deposit morphology is studied and discussed.

Intergranular stress corrosion cracking (IGSCC) and transgranular stress corrosion cracking (TGSCC) of Type316L stainless steel are examined by a slow strain rate test (SSRT) in a dilute sodium sulphate solution at 288 °C. Crystallographical orientation measurements on Type316L are performed by electron backscatter diffraction (EBSD) in order to grasp crack initiation during SSRT. The relationship between crystal orientation and crack initiation of IGSCC and TGSCC can be characterized using EBSD data.

Mechanical properties of the ultrafine grained (UFG) steels having mean grain sizes much smaller than 1 μm were systematically shown from the original experimental data. The UFG steels performed very high strength that reached to 2–4 times of the starting materials having conventionally coarse-grained structures. On the other hand, the uniform tensile elongation was commonly limited within a few percents when the materials have single-phased UFG structures. The limited uniform elongation was explained in terms of early plastic instability. On the basis of the understanding, the ways to manage both strength and ductility in the UFG steels were indicated. The effectiveness of making the UFG structures multi-phased was confirmed.

Applications of ARB to magnesium alloys were limited due to low deformability. The authors recently found that the rollability of the alloys is significantly improved in highspeed rolling. It is supposed that the severe plastic deformation of magnesium alloy sheets is feasible if rolling in ARB processes is conducted at high speed. In this study, AZ31B and ZK60A sheets are processed by ARB up to five cycles at 423K with a speed of 1000m/min. Vickers hardness increases with increasing number of ARB cycles, while the tensile strength shows the maximum after the second cycle. The grain size is reduced significantly at the first cycle and decreases gradually from the second cycle. The mean grain sizes after five cycles are 1.6μm for AZ31B and 1.8μm for ZK60A. It is concluded that ARB using high-speed rolling is effective for grain refinement of magnesium alloys.

Micro and nano-joining has been identified as a key enabling technology in the construction of micromechanical and microelectronic devices. The current article reviews recent progress in micro and nano-joining. In particular, laser micro-welding (LMW) of crossed 316 LVM stainless steel (SS) wire was compared to conventional resistance micro-welding (RMW) and was successfully employed in welding a Pt-Ir /SS dissimilar combination. Welding of Au nanoparticles was realized using femtosecond laser irradiation and its application in the surface enhanced Raman spectroscopy was investigated. Brazing between carbon nanotube (CNT) bundles and Ni electrodes was attained in vacuum, resulting in the development of a novel CNT filament of incandescent lamps.

The Friction Stir Processing (FSP) was applied to the surface hardening of cast irons. Flake graphite cast iron (FC300) and nodular graphite cast iron (FCD700) were used to investigate the validity of this method. The matrices of the FC300 and FC700 cast irons are pearlite. The rotary tool is a 25mm diameter cylindrical tool, and the travelling speed was varied between 50 and 150mm/min in order to control the heat input at the constant rotation speed of 900rpm. As a result, it has been clarified that a Vickers hardness of about 700HV is obtained for both cast irons. It is considered that a very fine martensite structure is formed because the FSP generates the heat very locally, and a very high cooling rate is constantly obtained. When a tool without an umbo (probe) is used, the domain in which graphite is crushed and striated is minimized. This leads to obtaining a much harder sample. The hardness change depends on the size of the martensite, which can be controlled by the process conditions, such as the tool traveling speed and the load. Based on these results, it was clarified that the FSP has many advantages for cast irons, such as a higher hardness and lower distortion. As a result, no post surface heat treatment and no post machining are required to obtain the required hardness, while these processes are generally required when using the traditional methods.

Highly precise fabrication of welded materials is in great demand, and so microstructure and distortion controls are essential. Furthermore, consideration of process mechanics is important for intelligent fabrication. In this study, the microstructure and hardness distribution in multi-pass weld metal are evaluated by computational simulations under the conditions of multiple heat cycles and phase transformation. Because conventional CCT diagrams of weld metal are not available even for single-pass weld metal, new diagrams for multi-pass weld metals are created. The weld microstructure and hardness distribution are precisely predicted when using the created CCT diagram for multi-pass weld metal and calculating the weld thermal cycle. Weld distortion is also investigated by using numerical simulation with a thermal elastic-plastic analysis. In conventional evaluations of weld distortion, the average heat input has been used as the dominant parameter; however, it is difficult to consider the effect of molten pool configurations on weld distortion based only on the heat input. Thus, the effect of welding process conditions on weld distortion is studied by considering molten pool configurations, determined by temperature distribution and history.

A Zr₅₅Cu₃₀Al₁₀Ni₅ bulk metallic glass plate was successfully welded below its crystallization temperature by friction stir welding. The flash formation and heat concentration at the shoulder edge was minimized using a wider tool and the angle of the recessed shoulder surface was 3°. To analyze the crystallization of the base material and stir zone, the microstructure and mechanical properties were analyzed using DSC, XRD, TEM, and micro-hardness. As a result, it was found that the amorphous structure and original mechanical properties were maintained in the whole joints.

The characteristics of interfacial microstructures with additional elements in dissimilar 6000 system aluminum/steel joints were basically evaluated using tensile test, EPMA, TEM and nanoindentation. For Si (and Cu)-added alloy (S1 and SC), EPMA analysis showed that Si (and Cu) was enrichment in the reaction layers, which were formed during diffusion bonding. SAED pattern clarified that the reaction compounds at the interface changed from AlFe intermetalic compounds to AlFeSi intermetalic compounds by Si addition. Nanoindentation technique was successfully applied to the interfacial microstructures to understand directly the nanoscopic mechanical properties in the interfacial microstructures. The hardness and Young's modulus of Al₃Fe intermetalic compounds was lower than those of Al₂Fe₅ intermetalic compounds. Moreover, the hardness and Young's modulus of AlFeSi(Cu) compounds were lower than those of Al₃Fe, indicating that the crystal system changed from orthorhombic structure to cubic structure. Joint strength of SC/steel joints was higher than that of the aluminum alloy with no additional element (Base)/ steel joint, indicating that interfacial microstructure was modified by the addition of Si and Cu to the 6000 system aluminum alloy. These results suggest that the nanoscopic mechanical properties at the interface microstructures affect greatly the macroscopic deformation behavior of the aluminum /steel dissimilar joints.

The convergent-beam electron diffraction (CBED) method we proposed recently for enantiomorph identification has successfully been applied to some amino acid crystals such as glutamic acid and threonin. Enantiomorph identification (either left-handed or right-handed form) can readily be made within the framework of the proposed method by noting the asymmetric intensity distribution of Bijvoet pairs of reflections in the CBED pattern taken along an appropriate zone-axis orientation. Although the proposed method usually requires only a single CBED pattern, some effort to eliminate the ambiguity of 180°-rotation of the CBED pattern about the incident beam is needed for enantiomorph identification for these organic crystals because of the lack of HOLZ (higher-order Laue zone) reflection disks.

We have investigated photoluminescence (PL) and linear and induced absorption of Si nanocrystals (NCs) in a SiO₂ matrix. By measuring PL intensity dependence on the excitation photon fluence, we conclude that (i) the excitation wavelength independent saturation level is reached when on average a single photon is absorbed per NC, and (ii) excitation cross-section is proportional to the linear absorption only in the low-energy range; for higher energy it increases faster than absorption, thus indicating more efficient excitation which appears due to space-separated quantum cutting (SSQC). In order to shed more light on the mechanism behind the SSQC, we have compared spectral and temporal characteristics of induced absorption for excitation wavelengths below and above the SSQC threshold. These were found to be very similar, thus indicating that the SSQC process is most likely very fast, possibly taking place already during photon absorption by NCs.

We synthesized hexagonal diamond directly from highly oriented pyrolytic graphite (HOPG) using a femtosecond laser pulse without catalyst. A femtosecond laser pulse with wavelength of 800 nm, pulse width of 130 fs, the intensity of 2×10¹⁵ W/cm² was irradiated onto the HOPG surface in air. Crystalline structures of the fs laser-affected region in the HOPG were analyzed using grazing-incidence XRD method. We found that the hexagonal diamond which is the metastable high-pressure phase of carbon appeared in the HOPG which was irradiated by the femtosecond laser normal to the basal plane. We suggest that the femtosecond laser-driven shock wave induces the graphite – hexagonal diamond transformation and that the hexagonal diamond is synthesized due to the rapidly cooling in the shock heated region.

The effects of static magnetic field and hydrogen gas atmosphere on the solidification of silicon with hydrogen atom were examined using electromagnetic levitation method in hydrogen atmosphere. Since the convection in the levitated silicon melt which is caused by the electromagnetic field is restrained by a high static magnetic field, it is possible to examine a solidification of silicon from an equilibrium silicon melt with hydrogen without an inflluence of convection. The cooling rate of the melt increases with increasing hydrogen partial pressure. Since the undercooling of the melt does not change with the static magnetic field, the grain size on the surface does not change with the static magnetic field. However, the morphology of the grain surface drastically changed from flat and smooth to undulation.

This paper presents the three-dimensional alignment of β-FeSi₂ particles in a resin under an anisotropic magnetic field. The alignments obtained under a static magnetic field and a rotating magnetic field proved that the magnetic susceptibilities of β-FeSi₂ are χ_c>χ_b>χ_a. The magnetic anisotropy was sufficiently large under a magnetic field of 10T for the crystallographic alignment. Under an anisotropic magnetic field (oscillating magnetic field), a-, b- and c-axes of β-FeSi₂ particles suspended in a resin were aligned each other. A pseudo single crystal β-FeSi₂ with the orthorhombic structure was fabricated by the oscillating magnetic field.

Fe-Si melt is a candidate for use as an alloy solvent for rapid liquid phase growth of SiC because of the high solubility of carbon in molten iron. In this work, the equilibrium phase relationship between SiC and the liquid phase of the Fe-Si-C system was investigated at 1523 K by the phase equilibration technique, and was further studied with the thermodynamic calculation. It was found that Fe- 36 mol% Si melt equilibrates with SiC at the temperature and possesses the higher carbon solubility than silicon-based melt. The first trial of the SiC crystal growth experiment was then carried out with Fe- 36 mol% Si melt by means of temperature difference method, and formation of SiC layer was obtained on the graphite substrate. Accordingly, it was found possibility for rapid growth of SiC by the solvent growth method with Fe- 36 mol% Si melt.

Band structure of β-FeSi₂ epitaxial and polycrystalline films was investigated by photoreflectance (PR) measurements. In the PR spectra, two critical points were obtained at E_g = 0.918 eV and E_i = 0.895 eV in the films. The E_g consistent with the direct transition energy at Y point in the Brillouin zone of β-FeSi₂. The E_i was assigned as a transition-energy between conduction band and the acceptor-level originating from Si vacancies in β-FeSi₂. The energy from top of valence band to the acceptor-level was determined to be 23 meV.

Growth of GaAs by MOVPE (Metal-Organic Vapor-Phase Epitaxy) at a much lower temperature than conventional conditions of 650°C required to produce highly resistive buffer layers and molecular-layer-level abrupt heterojunctions and doped-layer interfaces. It has been widely recognized that growth at 550∼600°C causes poor morphological surfaces on the grown layers. This paper describes a fundamental improvement of the low-temperature growth of GaAs, resulting in smooth surfaces with very low impurity concentrations. Excellent GaAs layers can be grown at 500°C by increasing partial pressure of AsH₃ during growth. The method has been applied to high quality epitaxial layers for electronic device production.

We have fabricated two types of 1.5 μm light-emitting devices based on Er,O-codoped GaAs (GaAs:Er,O) using organometallic vapor phase epitaxy (OMVPE). In a device exhibiting a room-temperature lasing at the GaAs band-edge, the threshold current density increased with Er concentration. The Er intensity rose up steeply and then decreased gradually in the spontaneous emission region. In the stimulated emission region, it remained almost constant, reflecting ultrafast energy transfer to Er ions. In a device exhibiting a room-temperature lasing from strained GaInAs quantum wells (QWs) embedded in GaAs:Er,O, on the other hands, the lasing wavelength was designed to tune to the energy separation between the second excited states ⁴I_11/2 and the ground state ⁴I^15/2 of Er³⁺ ions. The Er intensity revealed initially steep increase with injected current density in the region for spontaneous emission from the GaInAs QWs. In the stimulated QW emission region, however, the intensity continued to increase with the current density.

We investigated the relationship between optical properties and crystal quality of Eu-implanted GaN-based semiconductors. Emission peaks that are due to intra-4f shell transitions in Eu³⁺ ions were observed in Eu-implanted GaN, InGaN, and AlGaN. Emission wavelengths from the Eu ions were constant regardless of the matrix materials. The emission peak intensity in the Eu-implanted AlGaN was higher than that in the Eu-implanted InGaN or GaN because of less matrix damage by ion implantation.

Er-doped ZnO (ZnO:Er) thin films were grown by metalorganic chemical vapor deposition. Subsequent annealing improved the crystallinity of the ZnO:Er by suppressing nonradiative centers in the films. The annealed ZnO:Er showed clear 1.54 μm PL originating from the ⁴I_13/2 → ⁴I_15/2 transitions of Er³⁺ ions. This was in addition to luminescence from the ZnO host. Er-related PL intensity increased when prepared with annealing temperatures of 800–900 °C, while annealing at 1000 °C reduced the intensity. PL of 1.54 μm with high intensity was obtained in the ZnO:Er film that exhibited band-edge emission of free- or bound-excitons.

The origin of the size-dependent photoluminescence (PL) of CuInS2-ZnS alloyed quantum dots and the effect of a ZnS thin layer coating on PL properties were studied. The PL emission band shifted to the shorter wavelength region with a decrease in the crystal size. A time-resolved measurement indicated that the PL lifetime for the lower energy component of the emission band was longer than that of the higher energy component. Although this is a characteristic feature of the donor-acceptor pair (DAP) recombination, the size dependent shift of the band was too large to be attributed to DAP recombination. The emission was attributed to an electronic transition from the quantum confined conduction band to an acceptor level. By coating with a ZnS thin layer the PL quantum yield increased up to 31%.

For future advancement in magnetic and spin electronic devices, it is essential to understand the magnetic behaviour and spin dynamics of a single nanomagnet. However, no effective method has been available for detection of such extremely small magnetic moment. To overcome this difficulty, we have newly developed a highly sensitive magnetic detection technique utilizing the anomalous Hall effect (AHE). This technique has systematically revealed the magnetic behaviors of various magnetic materials in the nanoscale regime. In this article, we demonstrate how single nanomagnets of L10 FePt and Co/Pt multilayers behave depending on their size, especially focusing on the magnetization reversal process and the bistability condition necessary for digital memory use.

Magnetic logic devices have been investigated by micromagnetics simulation and experiment. The simulation shows that the magnetic logic devices composed of 4 elliptical permalloy dots perform both the NAND and NOR logic operations. The experiments indicate that the distance between adjacent dots must be shorter than 50 nm and 70 nm along the long axis and the short axis of the dots, respectively. The micro-fabricated test devices show the above logic operations.

We investigated the basic properties of Mn-Zn ferrite thin films fabricated by pulsed laser deposition (PLD) with the aim of controlling their saturation magnetizations (M_s) and electrical resistivities. The M_s and electrical resistivities varied dependent on the substrate temperature during deposition and post-anneal conditions. X-ray diffraction (XRD) measurement revealed the difference is due to the formation of a Mn-Zn ferrite and an Fe₂O₃ phase, which is attributed to the amount of oxygen in the post anneal atmosphere.

The structure and magnetic behavior of Mn-AlN (Al_1-xMn_xN, x = 0.03, 0.04) films deposited on thermally oxidized Si (001) substrates and sapphire (0001) substrates were studied. Mn-AlN films deposited on each substrate had a würtzite-type AlN phase with a preferentially oriented c-axis. Mn-AlN films that were deposited on Si (001) substrate exhibited paramagnetic behavior. In addition to paramagnetic behavior, weak ferromagnetic behavior with curie temperatures higher than room temperature were observed for Mn-AlN films deposited on sapphire (0001) substrates.

We have investigated the role of the Ru underlayer for the segregated structure formation of CoCrPt-SiO₂ granular thin film to improve the recording performance of CoCrPt-SiO2 perpendicular magnetic recording media. The coercivity of the granular film decreases with decreasing Ru layer thickness, which comes from the change in the segregated structure. Formation of an oxide grain boundary is needed to obtain higher coercivity. According to a morphology analysis, the Ru layer roughness is closely related to the coercivity. The results indicate that the oxide grain boundary formation of the granular layer is enhanced by the roughness of the Ru underlayer. The shapes of the Ru grains probably serve as a template in CoCrPt grain growth and oxide grain boundary formation. As one of the methods for enhancing the roughness of the Ru underlayer, an island shape growth on the low surface energy materials is applicable. The roughness of the Ru layer was increased using a Pd/MgO/Pd seed layer and high coercivity was successfully obtained even when the Ru layer thickness was 3 nm.

The effect of antiferromagnetic Cr₂O₃ thin film on the magnetic properties of ultrathin Co film has been investigated. To achieve the investigation, we have also investigated the fabrication of Cr2O3 thin film of high quality using MBE. The crystalline quality of Cr oxide film strongly depends on the in-plane epitaxial variants of Cr(110) before the oxidation. We have successfully fabricated Cr₂O₃(0001) film by oxidizing three fold-symmetric Cr(110) film. On Cr₂O₃(0001) thin film, the magnetization of Co is stabilized parallel to the Cr spin direction below the Néel temperature of Cr₂O₃.

We investigated magnetic properties of BiCrO₃ having the C2/c symmetry and Bi_0.9Y_0.1CrO₃ having the Pnma symmetry with a = 5.5451(1) Å, b = 7.7185(2) Å, c = 5.3882(1) Å. Bi_0.9Y_0.1CrO₃ shows magnetic anomalies at T_N = 154 K due to an antiferromagnetic transition with weak ferromagnetism and frequency-dependent anomalies near 50 K with very large temperature shifts on both the real and imaginary parts of the ac susceptibilities (43 K at 0.5 Hz and 56 K at 300 Hz on χ"). Based on the results for Bi_0.9Y_0.1CrO₃, we could understand some peculiarities in magnetic properties of BiCrO₃. In particular, an orthorhombic modification of BiCrO₃ was found to exist in the monoclinic form. We also discuss magnetic properties of BiMnO₃ that shows more complicated behavior.

We propose a magnetic force imaging technique based on MFS. This method is directly visualize a two dimensional motion of the domain structure of the sample, without a disturbance of the magnetic structure of the sample by MFM tip stray field, and improve the interpretation of the result of MFS. Magnetic force imaging by MFS is useful to investigate magnetic domain structures and magnetization switching mechanism of magnetically soft materials.

The present work reports the deposition behavior of Pd or Ag particles on TiO₂ nanotube layers with an average diameter of 100 nm and length of 500 nm. The deposition of Pd or Ag particles is carried out by photo-induced reduction of metal ions. TiO₂ nanotube layers are exposed to UV irradiation in an electrolyte containing Pd ions. Pd deposition behavior strongly depends on the crystal structure of TiO₂ nantoube layers. As-anodized nanotube layers are in the amorphous state and can be converted to anatase without any changes in structural integrity. On as-formed amorphous nanotube layer, no Pd particles are deposited whereas in the case of anatase nanotube layer many Pd particles are formed by UV irradiation in PdCl2 electrolyte. However, the size of Pd particles is larger than the diameter of nanotubes and increases with irradiation time. In order to circumvent this problem, TiO₂ nanotube layer is immersed in an electrolyte, then rinsed with deionized water, followed by exposure to UV irradiation. This results in the deposition of Ag nanoparticles with the size of 10 nm on and inside nanotubes. Photocurrent measurements reveal that the presence of Ag nanoparticles on TiO₂ nanotube layer enhances the absorption efficiency of dye, leading to higher photocurrent.

Vanadium doped TiO₂ photocatalysts (V/TiO₂) were synthesized by sol-gel method and hydrothermal treatment using tetraisopropyl orthotitanate (TPOT) and vanadium acetylacetonate in the presence and absence of NH₄F. In comparison to the vanadium non-doped TiO₂ (pure TiO₂), V/TiO₂ exhibited the long-tailed absorption in the visible light region above 380 nm. Degradation of isobutanol diluted in water proceeded efficiently on V/TiO₂ even under visible light (λ > 430 nm) irradiation, although pure TiO₂ did not show any photocatalytic activities. It was also found that the hydrothermally synthesized V/TiO₂ (HT-V/TiO₂) in the presence of NH₄F possessed high crystallinity of anatase phase and hydrophobic surface, showing the higher photocatalytic activities as compared to those of sol-gel V/TiO₂ as well as HT-V/TiO₂ prepared in the absence of NH₄F.

Nanosized Pd and PdAux alloy particles were deposited on various titanium dioxides (TiO₂; anatase + rutile, anatase, and rutile) using a simple and unique microwave-assisted method. The structure of the obtained samples was characterized by X-ray diffraction (XRD), N₂ adsorption-desorption and X-ray absorption fine structure (XAFS) methods. Results of XAFS indicated that PdAux alloy particles had formed on TiO₂ supports. Catalytic activity was estimated by a direct synthesis reaction of hydrogen peroxide (H₂O₂) from hydrogen and oxygen gases at room temperature. A high reaction rate could be obtained using these PdAux alloy catalysts compared with conventional Pd catalysts.

The encapsulation of fcc FePt nanoparticles (NPs) having a mean diameter of ca. 6 nm with the Ti-SiO₂ thick shell gives a new nanocomposite (FePt@Ti-SiO₂), which serves as an efficient nanocatalyst for the liquid-phase selective epoxidation. Recovery from the reaction mixture was readily attained by applying an external permanent magnet, and the spent catalyst could be recycled without any appreciable loss in activity.

Colloidal undoped and Y₂O₃-doped CeO₂ nanocrystals (NCs) with diameters of 2.1∼2.6 nm were synthesized by the thermolysis of cerium and yttrium acetylacetonate dissolved in oleylamine. A surface modification of the NCs, i.e., removing the oleylamine adsorbed on the NCs surfaces was attempted by heating to 200 °C under an oxygen atmosphere and by treatment at 200 °C in pure water and H₂O₂ aqueous solution in an autoclave. Heating under an oxygen atmosphere resulted in a complete removal of oleylamine. However, the particle size increased to above 5 nm after heating. For the autoclave treatment, especially in the H₂O₂ aqueous solution, the oleylamine was almost removed without distinct particle growth.

Plasma generation and control technologies for meters-scale ultra-large-area plasma sources have been developed with multiple low-inductance antenna (LIA) modules, as a promising candidate of ultra-large-area and high-density plasma sources for next-generation processing of hybrid flexible devices. Properties of argon-oxygen mixture plasmas sustained with multiple LIA units have been investigated and surface modifications of polymer substrates using the plasmas have been performed. Ion energy distribution at the sheath edge of the argon-oxygen mixture plasmas showed considerable suppression of ion energies as small as or less than 10 eV. We have examined effect of plasma exposure on surface modification and/or degradation of polymer. Surface analysis of polytetrafluoroethylene (PTFE) exhibited nano-surface modification of the polymer surface without suffering degradation of molecular structures beneath the nano- surface layer.

First-principles electronic structure calculations have been performed in order to investigate the effect of hydrogen on the stability of Ni vacancy after hydrogen desorption in LaNi₅. Our calculated binding energy of hydrogen to the Ni vacancy indicates that up to two H atoms are trapped at the Ni vacancy. The formation energy of the Ni vacancy trapping H atoms was found to decrease with the number of H atoms and to turn negative when two H atoms are trapped. These results suggest that the Ni vacancy is energetically stabilized by trapping two H atoms in hydrogen desorption process.

Recently, a new process has been proposed for fabricating a LSI interconnection; filling trenches and via holes with Cu using high-pressure annealing treatment. It is already known that a Cu film produced by physical vapor deposition (PVD) has a lower reflowability compared to a Cu film produced by electrochemical deposition (ECD). Additionally, it has also been recognized that the addition of Sb to the PVD-Cu film improves the reflowability. However, the factors responsible for the reflowability of Cu films have not yet been studied. In this work, we evaluated a PVD pure-Cu film and a PVD Cu-0.5at%Sb film by using a slow positron beam. Addition of Sb led to the introduction of lattice defects in the as-deposited film. These defects that were observed in the PVD-CuSb dilute alloy film were identified as frozen-in vacancies that were produced during deposition.

In this work we investigate the complex atomic structures of two CMAs in the Eu-Ag-In system, these are quasicrystal approximants and their structures are similar to those of the 1/1 and 2/1 approximants found in the Yb-Cd system. The similarities between the Yb-Cd and Eu-Ag-In systems indicate that the RE-Ag-In systems in several ways can be treated as pseudo-binary. The investigated phases however also show some interesting differences; when Cd is replaced by the elements Ag and In partial chemical disorder is introduced into the system. As a response, the positional disorder which is usually observed in certain atomic positions for these types of approximants tends to disappear. The complete structures refined from single crystal data are described in detail and the differences with other related phases are highlighted. The 2/1 approximant in the Eu-Ag-In system is the first to exhibit such a degree of structural perfection in terms of positional order, a feature which makes this phase very suitable for further studies of chemical order and also can give insight into the local structures of related disordered phases. Electron diffraction studies on the 1/1 approximant phase have previously shown the presence of weak superlattice reflections, indicating a superstructure similar to that of the related phase Eu₄Cd₂₅. However, no superlattice reflections could be observed by single crystal diffraction and it is thus concluded that the phase has a disordered average structure with possible short to medium range superlattice ordering.

The Gas Diffusion Layer(GDL) of fuel cell, are required to provide both delivery of reactant gases to the catalyst layer and removal of water in either vapor or liquid form in typical PEMFCs. In this study, the fabrication of GDL containing Micro Porous Layer(MPL) made of the slurry of PVDF mixed with carbon black is investigated in detail. Physical properties of GDL containing MPL, such as electrical resistance, gas permeability and microstructure were examined, and the performance of the cell using developed GDL with MPL was evaluated. The results show that MPL with PVDF binder demonstrated uniformly distributed microstructure without large cracks and pores, which resulted in better electrical conductivity. The fuel cell performance test demonstrates that the developed GDL with MPL has a great potential due to enhanced mass transport property due to its porous structure and small pore size.

Soldering or anisotropic conductive adhesives (ACAs) are used for assembling of devices with high-density area-array terminals. However, solder bump processes are high-cost and interconnects of ACAs are less-reliable. The purpose of this study is to develop a novel method. Self-organization assembly method that uses active resin containing solder fillers may allow reliable interconnects at low-cost. Fundamental process of Self-organization assembly are movement, coalescence, and wetting of molten fillers in resin. The focus of this paper is on the movement of fillers. In-situ observations of coalescence behavior for 40μmϕ molten fillers in the resin revealed irregular movement of the fillers at a velocity of several μm/s. Numerical analysis, using improved volume fraction method, indicated that a 10% degradation of interfacial energy on one side of a 40μmϕ filler could move the filler at a velocity of several mm/s. This degradation of the interfacial energy was resulted the remaining oxide film.

We developed a technique for the formation of nonplanar surfaces of inorganic optical materials by a combined process of nonlinear lithography and plasma etching. This technique can be used to fabricate structures even on non-flat substrates, which is difficult using current semiconductor technology. Three-dimensional patterns were written directly inside a positive-tone photoresist using femtosecond laser-induced nonlinear optical absorption. The patterns were then transferred to underlying nonplanar substrates by the ion beam etching technique. For the lithographic process, we obtained a minimum feature size of 900 nm, which is below the diffraction limit. We demonstrated the fabrication of silica-based hybrid diffractive-refractive lenses. Fresnel zone plates with smooth surfaces were obtained on convex microlenses. When a 633-nm-wavelength He-Ne laser was coupled normally to the hybrid lens, the primary focal length was measured as 630 μm. This hybridization shifted the focal length by 200 μm, which agreed with the theoretical value. Our process is useful for the precise fabrication of nonplanar structures based on inorganic materials.

We firstly observed the surfaces of periodic structures consisting of Ge nanoparticles in channel waveguide cores of GeO₂-B₂O₃-SiO₂ glasses. Such periodic structures were formed only by direct laser writing and subsequent annealing. Refractive index patterns were written in the glass films by irradiation with KrF excimer laser pulses. Then, the patterns were inverted by thermo-induced precipitation of Ge nanoparticles at unirradiated areas. Brown channel cores were obtained due to predominant precipitation of the nanoparticles in the cores. The periodic structures of 530 nm pitch and microchannels were directly observed by using wet-etched samples. The periodic structures were confirmed only inside the cores.

In this paper the magnetocaloric behaviour of Ni-Mn-based Heusler alloys is discussed in relation to their shape-memory and superelastic properties. We show that the magnetocaloric effect in these materials originates from two different contributions. The first, associated with the mechanism which makes feasible the magnetically induced rearrangement of martensite variants, controls the magnetocaloric effect at low applied fields, while the latter is dominant at higher fields and is essentially related to the possibility of magnetically inducing the martensitic transition. The occurrence of the inverse magnetocaloric effect associated with these two contributions is also considered.

Magnetic field is one of intensive variables which influence phase transformations in solids. In the present study we will show how the characteristics of martensitic transformations are influenced by magnetic field, such as transformation temperature, morphology and kinetics.

We have investigated the rearrangement of crystallographic domains (martensite variants) in Ni₂MnGa ferromagnetic shape memory alloy and CoO antiferromagnetic oxide by applying magnetic field up to 8.0 MA/m. From the result of optical microscope observation of Ni₂MnGa single crystal, when a magnetic field is applied along [001]_p (p represents a parent phase), the rearrangement of crystallographic domains occurs and the single domain state is obtained below T_Ms = 202 K. The same rearrangement occurs but partially when a magnetic field is applied along [110]_p. On the other hand, when a magnetic field is applied along [111]_p, the rearrangement does not occur. In case of the CoO single crystal, when a magnetic field is applied along [001]_p below T_Ms = 293 K, the rearrangement occurs at 170 K ≤ T ≤ 293 K, but does not occur at T < 170 K. When a magnetic field is applied along [110]_p and [111]_p, the rearrangement does not occur below T_Ms. In order to explain the rearrangement in the alloy and the oxide, we have evaluated the magnetic shear stress, τ_mag, which is derived from the difference in magnetic energy among crystallographic domains and have compared it with the shear stress required for the twinning plane movement, τ_req. As a result, we have found that the rearrangement occurs when the value of τ_mag is larger than or equal to the value of τ_req for the present alloy and oxide.

Pseudoelastic behaviour of Fe-23.8at%Ga single crystals compressed with different loading axes at room temperature was examined focusing on the activated deformation mode. In the crystals, {101}<111> slip and {211}<111> twin were mainly activated depending on the loading axis. Pseudoelastic behaviour of the crystals depended strongly on the deformation mode. If {101}<111> slip was operative, sole and paired 1/4<111> superpartial dislocations moved dragging antiphase boundaries (APB). During unloading, the APB pulled back the superpartials due to its tension resulting in pseudoelasticity. In contrast, twinning and untwinning of {211}<111> pseudo-twins also led to large strain recovery accompanying a serrated flow during loading and unloading. It is suggested the energy of the pseudo-twin interface was the driving force for the twinning pseudoelasticity.

Neutron diffraction measurements using single crystal and powder synchrotron X-ray diffraction measurements revealed that an intermediate (I-) phase and a martensite (M-) phase in Ni₂MnGa have incommensurate modulated structures. The modulation vectors of the I- and M-phases are nearly equal to q = 0.341 <220> at 210 K and q = 0.427 <220> at 100 K, respectively. Moreover, displacements of Ni, Mn and Ga atoms in both the I- and M-phase are expressed by sinusoidal waves with the same phase and almost the same amplitudes.

The influence of a magnetic field on microstructure formation through a disorder-order transformation has been investigated in Co-Pt and Fe-Pd alloys. Single crystals of disordered Co-50Pt(at%) and Fe-55Pd(at%) were subjected to an ordering heat-treatment under a magnetic field. When the ordering heat-treatment is performed without applying a magnetic field, three equivalent variants are formed. On the other hand, when the ordering heat-treatment is performed under a magnetic field of 0.5 T (in CoPt) – 4 T (in Fe-55Pd) and higher as applied along the [001] direction of the disordered phase, a single variant with an easy axis along the field direction is obtained. The induced anisotropy energy of the ordered phase under a magnetic field of 1 T was 4.1 kJ∣m^-3 at 773 K for CoPt and 45.3 kJ∣m^-3 at 673 K for Fe-55Pd.

Transformation behaviour in Ti-(50-x)Pd-xFe shape memory alloys was investigated by electrical resistivity and specific heat measurements. The alloys with x = 14, 16 and 18 exhibit an obvious first order martensitic transformation but the alloys with x ≥ 19 do not exhibit a first order transformation down to 4.2 K. However, the alloys with x ≥ 19 show a negative temperature coefficient in electrical resistivity below a specific temperature T_min. In addition, specific heat measurements at cryogenic temperatures revealed that the Debye temperature of the first order transformation-suppressed state is significantly lower than that of the martensite state. These anomalies are suggested to be caused by an instability of the electronic structure of the B2-phase.

We investigated the grain size effect on martensitic transformation behavior in Fe-30at.%Ni powder and ribbon specimens. The powder specimen with a particle size of 5 um does not show an athermal martensitic transformation but does show an isothermal martensitic transformation after an incubation time of about 10⁴ s at 205 K. On the other hand, the powder specimen with a particle size of 20 um shows an athermal martensitic transformation at 150 K. The value of M_s is much lower than that of the single crystal and of bulk specimens. However, the M_s temperature of a ribbon specimen with an average grain size of 15 um is found to be almost identical to that of the single crystal and of bulk specimens. Considering these results, the athermal martensitic transformation is suppressed by the decrease in particle size if grains do not have grain boundaries.

We investigated martensitic transformation behaviour in sensitized SUS304 austenitic stainless steel to determine the stability of the austenitic phase at low temperatures. We found that a specimen that was sensitized at 973 K for 100 h exhibits an isothermal martensitic transformation when the specimen is held in the temperature range between 60 and 260 K. We constructed a time-temperature-transformation (TTT) diagram corresponding to the formation of 0.5 vol. % α'-martensite. A magnetization measurement was used to evaluate the volume fraction of a'-martensite. The TTT diagram shows a double-C curve with two noses located at about 100 and 200 K. In-situ optical microscope observations reveal that the double C-curve is due to two different transformation sequences. That is, the upper part of the C-curve is due to a direct γ → α' martensitic transformation and the lower part of the C-curve is due to a successive γ → ψ → α' martensitic transformation. The direct γ → α' transformation occurs in the vicinity of grain boundaries while the successive γ → ψ' → α' transformation occurs near the centre of grains. A scanning electron microscope observation reveals that carbide particles of M₂₃C₆ are formed in the grain boundaries. The concentration difference between the centre of the grains and regions near grain boundaries is the reason for the difference in the isothermal transformation sequence for the sensitized SUS304 stainless steel.

We have investigated a quadrupolar interaction by measuring the temperature dependence for elastic constants of C_B = (C₁₁ + 2C₁₂)/3 (Γ₁-symmetry), C' = (C₁₁ – C₁₂)/2 (Γ₃-symmetry) and C₄₄ (γ₅-symmetry) in DyCu. We have evaluated the quadrupolar interaction coefficient K(0) by analyzing critical fields of metamagnetic transitions. The following results are obtained; (i) C₄₄ in the paramagnetic phase exhibits softening at temperatures near T_Nwhile C_B and C' do not. This indicates that the dominant components of the quadrupolar moment are O_yz, O_zx and O_xy with Γ₅-symmetry. (ii) From an analysis of critical fields of metamagnetic transitions based on a mean-field approximation we have determined the quadrupolar interaction coefficient to be K(0)|Q|²/k_B = –16.8 K. The negative value of K(0) indicates that the quadrupolar interaction has an antiferro-type of Γ₅-symmetry.

As to the electrochemical formation of Dy-Ni alloy films in a molten LiCl-KCl-DyCl₃ system at 700 K, the growth of DyNi₂ film and behavior of anodic dissolution of Dy from the formed DyNi₂ film were investigated. The DyNi₂ films were formed by potentiostatic electrolysis at 0.55, 0.62 and 0.70 V with Ni electrodes. The growth rates of DyNi₂ films are higher at less noble potential, i.e., 0.47 8m min^-1 at 0.55 V, 0.32 8m min^-1 at 0.62 V and 0.14 8m min^-1 at 0.70 V. From RBS analysis, it was suggested that the Dy-Ni alloy film was formed for 10 or 30 s during electrodepositing Dy at 0.30 V with a Ni electrode. Moreover, the growth rate of Dy-Ni alloy film was faster than that of Dy-Fe alloy film. Anodic electrolysis of the formed DyNi₂ film with thickness of 15 μm was conducted at 0.90 V, 1.30 V and 1.90 V, respectively. The formed DyNi₂ were transformed to other phases, i.e., DyNi₃, DyNi₅ and Ni, by selective anodic dissolution of Dy. The transformed Ni film was about 10 μm in thickness and had a porous structure with a pore diameter of 1∼2 μm.

Cellular metallic materials are a new class of materials which have been the focus of numerous scientific studies over the past few years. The increasing interest in cellular metals is due to the fact that the introduction of pores into the materials significantly lowers the density. These highly porous materials also possess combinations of properties which are not possible to achieve with other materials. Besides the drastic weight and material savings that arise from the cell structure, there are also other application-specific benefits such as noise and energy absorption, heat insulation, mechanical damping, filtration effects and also catalytic properties. Cellular metallic materials are hence multi-functional lightweight materials.

Membrane separation is regarded nowadays as a preferred method for production of purified hydrogen. Palladium (Pd) is an attractive membrane material due to its ability to dissociate molecular hydrogen into atoms. It is usually deposited on the porous substrate that can provide good mechanical support and reduce the thickness of the membrane for maximizing hydrogen permeability. Pd membrane used for hydrogen separation must be thin enough to increase hydrogen flux and reduce cost while remaining thick enough to retain adhesion, attrition resistance and mechanical integrity during high temperature cycles. In this paper, the progress of electroless deposition of Pd around the pore area at surface of porous stainless steel was recorded and a bridge structure that was formed during the membrane deposition around the pore area of the substrate was illustrated. After that, the porous substrate was modified using micro-or nano-size metal or metal oxide particles in order to reduce pore size in the substrate surface. The experimental results obtained from hydrogen permeation through the Pd membranes having the thickness from 400 nm to 18 μm built on both modified and original porous stainless steel substrates demonstrate that these thin membranes are solid and they can be used at the temperature of 550°C and hydrogen pressure difference of 3.447x10⁵ Pa. The proposed processing will allow optimizing the design and fabrication of thin Pd membranes on different porous substrates for hydrogen separation.

Lotus-type porous metals were fabricated by unidirectional solidification in pressurized gas atmosphere. The elongated pres are evolved by insoluble gas resulted from the solubility gap between liquid and solid when the melt is solidified. Recently we developed a novel fabrication technique, in which gas compounds are used as a source of dissolving gas instead of the high pressure. In the present work this gas compound method was applied to fabrication of lotus aluminium. Hydrogen decomposed from calcium hydroxide, sodium bicarbonate and titanium hydride evolves cylindrical pores in aluminium. The porosity is about 20%. The pore size decreases and the pore number density increases with increasing amount of calcium hydroxide, which is explained by increase in pore nucleation sites.

Lotus-type porous copper was fabricated by unidirectional solidification through thermal decomposition of titanium hydride. Effects of additive method and additive amount of titanium hydride on pore formation were investigated. The porosity of lotus copper depends on additive method and additive amount of titanium hydride. The pore formation effectively occurs in the method that titanium hydride decomposes in molten copper. For all the additive methods of titanium hydride, the porosity increases and pore diameter does not change with increasing additive amount of titanium hydride. While, for adding large amount of titanium hydride, the porosity became constant. This is because hydrogen solubility in liquid phase does not change owing to bubbling of hydrogen gas.

Lotus-type porous magnesium was fabricated through thermal decomposition of MgH₂ powders as a source of hydrogen. Liquid magnesium was cast into a mold in which MgH₂ powders were placed and solidified unidirectionally in the mold, which achieved the growth of unidirectional elongated pores in magnesium matrix. The fabrication method is safer than a conventional method using pressurized hydrogen atmosphere, because the risk of explosion can be avoidable. The effect of the amount of MgH₂ powders and the distance from the bottom of the ingot on the porous structure was investigated, which clarified that the two factors have the large influence on the pore growth.

Lotus-type porous aluminium was fabricated using a thermal decomposition of magnesium hydroxide in a vacuum. The effect of addition of magnesium hydroxide on the porosity and the average pore diameter were investigated. The hydrogen dissolved in molten aluminum through thermal decomposition of magnesium hydroxide when the melt solidified evolves pores. Furthermore, magnesia formed through thermal decomposition of magnesium hydroxide in the melt serves as nucleation sites for the pores. The porosity and the number of the pores increase with increasing amount of magnesium hydroxide, and the average pore diameter is not much varied with amount of magnesium hydroxide.

Lotus-type porous A1-14 wt.%Si alloy was fabricated by continuous casting at a transference velocity of 10 mm∣min^-1 in vacuum and by adding Ca(OH)₂ pellets to the crucible. The porosity and the pore diameter increased by varying the amount of Ca(OH)₂ from 0.2 g to 0.6 g. In the case of 0.6 g Ca(OH)2, the average porosity was about 30 % and the average pore diameter was about 3.8 mm. XRD patterns of the pellets after continuous casting showed the Ca(OH)₂ pellet did not decompose completely during continuous casting. The TG-DTA analysis showed that the Ca(OH)₂ pellet decomposes more slowly than Ca(OH)₂ powder. These results suggest that the Ca(OH)₂ pellets gradually decomposed in the crucible during the continuous casting, which is suitable for the supply of hydrogen over extended periods.

A1-4.5wt%Cu was unidirectionally solidified by continuous casting technique under hydrogen pressure of 0.1MPa at the transference velocities ranging from 1 to 50 mm∣min^-1. The fabricated slabs have microstructures of columnar α-dendrite and eutectic, which are typical for hypo-eutectic Al-Cu alloys. Elongated pores are observed in the eutectic region surrounded by several columnar α-dendrites. The shapes of the pores are affected by that of the surrounding α-dendrites. The average pores diameter is several ten μm smaller than the average dendrite arm spacing, which decreases with increasing solidification rate. Therefore, the pore diameter varies from about 200 μm to 30 μm with increasing transference velocity from 1 to 50 mm∣min^-1. The porosity of the fabricated samples is in the range between 2 ∼ 6 %. There is not significant dependence of the porosity on the transference velocity in the range of the present study. The porosity is similar to the reported value of Al-Si, where the area fraction of eutectic region is smaller than that of α-dendrite.

A porous magnesium spinel (MgAl₂O₄) with cylindrical pores was fabricated by unidirectional solidification in pressurized 1%H₂-99%Ar mixed gas. A small amount of Al₂O₃ phase was formed in porous MgAl₂O₄ bulk. Two different kinds of pores with large cylindrical and small facet shape were formed in the solidified samples. The former pores were dominant in the porous structure. The cylindrical pores were formed at solid-liquid interface due to hydrogen solubility gap. On the other hand, the small facet shape pores were formed by vaporization of MgO component on the cooling step in the solidification process. The pore length of the cylindrical pores decreased with increasing total pressure due to increasing the chance to nucleate pores during the solidification.

The surface of lotus-type porous copper plates that had cylindrical open pores in the thickness direction (porosity 50.4%, average pore diameter 144.4 μm) were processed by wire-brushing. Open pores on the surface of the lotus copper are closed by a newly formed nonporous thin layer. Electron backscatter diffraction patterns of the processed plate cross section show that the deformed surface consists of ultra-fine grains and that a nonporous layer was formed on the deformation of the surface layer. The Vickers hardness of the wire-brushed lotus copper is higher than that of the wire-brushed non-porous copper. The Vickers hardness increases with the increase in the rate of revolution of the wire–brush due to grain refinement. The increment of the ultimate tensile strength of lotus copper by wire-brushing is larger than of non-porous copper. The increment of ultimate tensile strength of the lotus copper reaches maximum when the newly deformed layer closes all the pores on the surface. These results show that wire-brushing is an effective process for the improvement of mechanical properties for lotus metals.

Lotus-type porous carbon steel (AISI1018) with aligned cylindrical pores was fabricated by continuous casting method in a mixture gas atmosphere (PN₂=0.8MPa, PAr=1.7MPa). Compressive yield strength of the nonporous and the lotus carbon steels was measured in the direction parallel to the transference direction. The compressive stress of the lotus carbon steel is lower than that of the nonporous carbon steel because of the existence of pores. The specific compressive yield strength (σ_0.2/ρ) of as-cast lotus carbon steel is higher than that of as-cast nonporous carbon steel, which is due to the solid solution hardening of nitrogen. Although the microstructure of carbon steel changes from Widmannstätten to homogeneous ferrite and pearlite by normalizing, the yield strength does not change significantly by normalizing. The microstructure of lotus carbon steel changes from Widmannstätten to martensite by quenching so that the yield strength increases significantly.

The structural stability of hollow Cu₂O and NiO nanoparticles, which were obtained via oxidation of Cu and Ni nanoparticles in air, was studied by transmission electron microscopy (TEM). Hollow Cu₂O and NiO were observed to have shrunk at 473 and 623 K in annealing under 5.0×10^-5 Pa, respectively, where the reduction reactions from oxides to metals started. As a result of shrinking associated with reduction, hollow oxides turned into solid metal nanoparticles after annealing at higher temperatures for a long time. On the other hand, hollow oxides shrunk and collapsed through annealing in air at high temperatures. It was found that shrinking of hollow oxides during annealing in air occurs at temperature where the diffusion coefficients of slower diffusing species reach around 10^-22 m²s^-1. We can conclude that hollow oxide nanoparticles tend to shrink and collapse at high temperatures because hollow structures with extra surface energy are energetically unstable.

Wettability of carbon and SiC by liquid Cu – Si alloys was measured by the sessile drop method and was mainly investigated from a thermo-dynamical view point. Penetration phenomena of liquid Cu – Si alloy into porous carbon was also investigated. It was found that it is possible to expect whether wetting system or non-wetting system from thermo-dynamic calculation.

The metal-to-metal bonding has been successfully achieved via the bonding process using Ag metallo-organic nanoparticles at a bonding temperature of around 300-, which can be alternative to the current microsoldering in electronics assembly using high-temperature solders. However, further reduction of bonding temperature and/or bonding pressure is needed. In the present research, a novel bonding process through in-situ formation of Ag nanoparticles instead of the filler material of the Ag metallo-organic nanoparticles has been developed. The Ag nanoparticles can form by the reduction of Ag₂O particles. In this study, the Ag₂O particles were mixed with triethylene glycol as a reducing agent to form a paste for bonding. The Au coated cylindrical specimens were bonded using the paste. The Ag nanoparticles formed at around 130 to 160 through the reduction process of Ag2O particles with triethylene glycol. The Ag nanoparticles were immediately sintered each other due to a great surface energy per volume. A transmission electron microscope observation revealed that the sintered Ag metallurgically bonded to the Au substrate at around 160 and a dense Ag layer formed after further heating. The tensile strength of the joint bonded at 250 under a bonding pressure of 5MPa was around 60MPa

MeV electron irradiation introduces Frenkel pairs in a crystalline phase and free volume in a glassy phase, resulting in solid-state amorphization of metallic crystals and crystallization of metallic glasses, respectively, in some alloy systems. C16-Zr₂Ni intermetallic compound transformed to a f.c.c.-type nano-crystalline phase through an amorphous state under electron irradiation. Namely, a metastable nano-crystalline structure was formed by a Crystal-to-Amorphous-to-Crystal (C-A-C) transition in the C16-Zr₂Ni thermal equilibrium phase. An overlap of amorphization and crystallization was observed at 298K, while such an overlap could not be observed at 103K. This behavior can be explained by the difference in the temperature dependence of electron dose required for amorphization and crystallization.

Ionic liquids are mixtures of organic and inorganic salts which are liquids at room temperature. Several potential applications of ionic liquids in the field of materials processing are electrowinning and electrodeposition of metals and alloys, electrolysis of active metals at low temperature, liquid-liquid extraction of metals. Results using 1-butyl-3-methylimidazolium chloride with AlCl₃ at low temperatures yielded high purity aluminium deposits (>99.9% pure) and current efficiencies >98%. Titanium and aluminium were co-deposited with/without the addition of TiCl₄ with up to 27 wt% Ti in the deposit with current efficiencies in the range of 78–85 %. Certain ionic liquids are potential replacements for thermal oils and molten salts as heat transfer fluids in solar energy applications due to high thermal stability, very low corrosivity and substantial sensible heat retentivity. The calculated storage densities for several chloride and fluoride ionic liquids are in the range of 160–210 MJ/m³. A 3-D mathematical model was developed to simulate the large scale electrowinning of aluminium. Since ionic liquids processing results in their low energy consumption, low pollutant emissions many more materials processing applications are expected in future.

The authors have applied hydrothermal reactions to develop recycling processing of slag or glass. As an example, under hydrothermal conditions such as 200 300°C and 30 40MPa with H₂O, powders made of glass can be sintered to become solidified glass materials containing about 10mass% H₂O. When the glass containing H₂O is heated again under normal pressure, the glass expands releasing H₂O to make porous microstructure. H₂O starts to emit just above the glass transition temperature. Therefore, when we have a glass with low glass transition temperature, we can make low temperature foaming glass. The SiO₂-Na₂O-B₂O₃ glass is a candidate to be such a foaming glass. In this paper, we describe our recent trial on the fabrication of the low temperature foaming glass by using hydrothermal reaction.

Thermodynamic analyses were performed to evaluate the composition ranges for metastable immiscibility including spinodal decomposition in multi-component silicate systems, where glass was regarded as a super-cooled liquid phase. Experimental studies on two-phase separation in glass were carried out, and the phase separation was observed in those silicate glasses corresponding to calculated metastable miscibility gaps.

The authors also attempted the fabrication of porous glass using phase separation in oxide glasses containing B₂O₃. A multi-component borosilicate glass composition was prepared based on the silicate glass composition where spinodal decomposition had been confirmed. One of the glass phases formed by the spinodal decomposition in the boro-silicate glass was selectively removed by acid leaching and porous glass was successfully obtained.

Solid-liquid interfacial energies in alloys are important properties for the process design of materials production such as casting and crystal growth. Inadequate information exists on solid-liquid interfacial energies. Our aim for this study was thus to establish a method to measure solid-liquid interfacial energies in alloys from equilibrium interfacial shapes of solid-liquid-gas phases. Experiments were carried out on Cu-B and Ag-Bi systems where copper and silver were treated as solid phases, respectively. Since the determined values for solid-liquid interfacial tensions for both systems agreed with reported values or estimated values, it was clarified that solid-liquid interfacial energies can be measured by observing the interfacial shape.

The effect of transformation-induced microscopic residual stress on fatigue crack propagation behavior of ferrite-martensite lamellar steel was discussed. Fatigue tests of prestrained and non-prestrained specimens were performed. Inflections and branches at ferrite-martensite boundaries were observed in the non-prestrained specimens. On the other hand, less inflections and branches were found in the prestrained specimens. The experimental results showed that the transformation induced microscopic residual stress has influence on the fatigue crack propagation behavior. To estimate the microscopic residual, a numerical simulation method for the calculation of microscopic residual stress stress induced by martensitic transformation was performed. The simulation showed that compressive residual stress was generated in martensite layer, and the result agree with the experimental result that inflections and branches were observed at ferrite-martensite boundaries.

Herein, our recent work on synthesis of nanocomposites including carbon nanotube-based, polymer-based and porous material-based composites via chemical reactions in supercritical fluids (SCFs) is introduced. The described examples highlight that SCFs allow facile synthesis of nanocomposites, leading to some new nanomaterials with special structures that are very difficult to achieve by conventional methods. The potential applications of these nanocomposites were also discussed.

We have newly developed micro-stereolithography system to realize freeform fabrication of micrometer order 3D metal structures. In this process, the photo-sensitive resin paste mixed with nanometer sized ceramic and metal particles was spread on a glass substrate with 10 μm in layer thickness by using a mechanical knife edge, and two-dimensional images of UV ray were exposed using DMD (Digital Micro-mirror Device) with 2 μm in part accuracy. Through the layer by layer stacking process, micrometer order three-dimensional objects were formed. Dense metal structures could be obtained by dewaxing and successive sintering of the formed objects. In our recent investigation, micro photonic crystals with lattice structures of alumina or pure copper were fabricated in order to control electromagnetic wave propagation in a terahertz (THz) frequency range. The micro photonic crystals with a diamond structure perfectly reflected the THz wave by Bragg diffraction

The aim of this study is to evaluate the use of our nano-HA/PCL composite 3D scaffolds as graft materials for mastoid cavity obliteration in an animal model. Nano-HA particles were synthesized by chemical precipitation technique and mixed them with PCL solution to make composite paste. 3D scaffolds were fabricated by a paste extruding deposition process. The nano-HA/PCL 3D scaffolds showed good in vivo bone regeneration behaviour in a rabbit model after 4 and 8 week implantation. To characterize the 3D scaffolds as a grafting material for mastoid obliteration, mastoid cavities were introduced in rats and implanted the scaffolds. After two week implantation, histological examination showed good tissue ingrowth and new bone formation behaviour. It can be argued that our nano-HA/PCL composite 3D scaffold is a promising alternative material for mastoid obliteration.

Biological apatite (BAp) c-axis orientation strongly depends on stress distribution in vivo and tends to align along the principal stress direction in bones. Dentulous mandible is subjected to a complicated stress condition in vivo during chewing but few studies have been carried out on the BAp c-axis orientation; so the adaptation of BAp crystal orientation to stress distribution was examined in rat dentulous mandible during bone growth and mastication. Female SD rats 4 to 14 weeks old were prepared, and the bone mineral density (BMD) and BAp crystal orientation were analyzed in a cross-section of mandible across the first molar focusing on two positions: separated from and just under the tooth root on the same cross-section perpendicular to the mesiodistal axis. The degree of BAp orientation was analyzed by a microbeam X-ray diffractometer using Cu-Kα radiation equipped with a detector of curved one-dimensional PSPC and two-dimensional PSPC in the reflection and transmission optics, respectively. BMD quickly increased during bone growth up to 14 weeks, although it was independent of the position from the tooth root. In contrast, BAp crystal orientation strongly depended on the age and the position from the tooth root, even in the same cross-section and direction, especially along the mesiodistal and the biting axes. With increased biting stress during bone growth, the degree of BAp orientation increased along the mesiodistal axis in a position separated from the tooth root more than that near the tooth root. In contrast, BAp preferential alignment clearly appeared along the biting axis near the tooth root. We conclude that BAp orientation rather than BMD sensitively adapts to local stress distribution, especially from the chewing stress in vivo in the mandible.

It is of great importance to understand how bone defects regain the microstructure and mechanical function of bone and how the microstructure affects the mechanical function during the bone healing process. In the present study on long bone defects, we investigated the relationship between the recovery process of fracture toughness and biological apatite (BAp)/collagen (Col) alignment as an index of the bone microstructure to clarify the bone toughening mechanisms. A 5-mm defect introduced in the rabbit ulna was allowed to heal naturally and a three-point bending test was conducted on the regenerated site to assess bone toughness. The bone toughness was quite low at the early stage of bone regeneration but increased during the postoperative period. The change in toughness agreed well with the characteristics of the fracture surface morphology, which reflected the history of the crack propagation. SEM and microbeam X-ray diffraction analyses indicated that the toughness was dominated by the degree and orientation of the preferred BAp/Col alignment, i.e. bundles aligned perpendicular to the crack propagation clearly contributed to the bone toughening owing to extra energy consumption for resistance to crack propagation. In conclusion, regenerated bone improves fracture toughness by reconstructing the preferred BAp/Col alignment along the bone longitudinal axis during the healing process of long bones.

A Ti-15Mo-5Zr-3Al alloy with a bcc structure are promising materials for biomedical application and were examined. The focus of this study was on the effect of heat treatment on microstructure and plastic deformation behaviour using a single crystal. The single crystal was successfully obtained by a floating zone method at a crystal growth rate of 2.5 mm/h. A slip at the <111> dislocation was present irrespective of heat treatment at 573K or 673K for 1.2 ks or 300 ks. The yield stress at the [149] loading axis varied significantly depending on microstructure, especially from the precipitation of the a phase. Al addition suppresses generation of the σ phase and increases the yield stress at the same time.

The preferential orientation of biological apatite (BAp) is a possible bone quality parameter for the comparison of the bone mechanical property. The preferential BAp orientation undergoes sensitive changes according to the change in the in vivo stress distribution, bone turnover rate etc., resulting in a variation of bone function. Osteoporosis is a metabolic bone disease characterized by reduced bone mass and deterioration of bone microstructure. The effect of osteoporosis on the preferential BAp orientation is however unknown. In this study, a microbeam-X-ray diffraction (μXRD) study was carried out on a trabecula extracted from osteoporotic and normal human vertebral bones and the degree of orientation for the BAp c-axis along its craniocaudal axis was analysed based on our previous report. A micro-computed tomography (μCT) measurement was also performed to analyze trabecular density and structure. In osteoporotic human vertebra, the trabecular number is markedly lower than that in normal vertebra. To sustain increased stress because of bone loss, the primary trabeculae, which are aligned parallel to the craniocaudal axis, tend to selectively remain while the secondary trabeculae, which are perpendicular to the craniocaudal axis, mostly disappear. Moreover, the primary trabecula from osteoporotic vertebra showed a significantly higher degree of BAp preferential orientation than the normal bone. This suggests that the remaining primary trabecula in osteoporotic vertebra is further reinforced by an increase in applied stress in vivo by enhancing the preferred BAp c-axis orientation along the trabecular direction.

Research on how implant surface shape contributes to long-term stability after implantation is important in the field of orthopaedics. In particular, technology that controls various bone quality parameters and voluntary bone inducement in implant surroundings should be developed for the next generation of implants and this will improve the patient's quality of life (QOL). For this research, we focused on the inducement of the appropriate alignment for biological apatite (BAp) crystallites and related collagen (Col.) fibres as a bone quality parameter. In this study, we predicted that when stress is applied to bone, the BAp/Col. preferential alignment can be formed if osteocytes are in an environment that is aligned with the principle stress vector. We tested this idea by introducing grooves in the principal stress direction on the surface of an implant. This work thus analyzes the effect of stress transmission by a load at the proximal femur on the bone inside and near the grooves by using mechanical simulation in which groove angles can be changed on the implant surface. Coordinate data from the mechanical simulation of the combined bone/implant environment was verified against the coordinate data obtained by CT scans of actual canine bone. Results suggest that the tendency of stress transmission differs depending on the position and angle of the grooves and based on a vector diagram of the maximum and minimum principal stresses. The simulation was able to predict bone dynamics in vivo and enabled a best design of an implant to control the BAp/Col. alignment as an index of bone quality.

A model for the nucleation and growth processes of Sn whisker is offered. High density of localized screw dislocations by deformation form the dense spiral steps of atomic scale on Sn surface. The spiral steps would induce the nucleation of Sn whisker. Edge dislocations localized at the same region where dense screw dislocations exist supply Sn atoms to the Sn whisker through pipe diffusion. Both screw and edge dislocations would bend along almost one direction, namely, to relax the external shear stress. The image force also helps to bend the dislocations perpendicular to the whisker side-surface. The bending of dislocations at root of whisker leads the bend of whisker. The pipe diffusion of Sn atoms through edge dislocations from bulk Sn toward whisker is suppressed at the bent part of edge dislocation, resulting in release of Sn atoms inside whisker and leading to the growth of whisker near its root.

Phases, oxygen release and absorption behaviour was investigated for Zr-Ce-Sn-Pr-O mixed oxides (Zr_2/8Ce_x/8Sn_y/8Pr_(6-x-y)/8)O_2-z (x=1.5, 4.5, y=0.5) together with the effect of Cu addition. Major phases of Sn-added samples were similar to those of Sn-free (Zr_2/8Ce_x/8Pr_(6-x)/8)O_2-z phases, that is, CaF₂-like λ and C phases for Pr-rich and Ce-rich compounds, respectively. All samples contained a SnO₂ or a Sn phase. When Cu was added, repetitive oxidation and reduction appeared to promote the fine dissolution of Sn and Cu through bulk phases. OSC values of both (Zr_2/8Ce_1.5/8Sn_0.5/8Pr_4/8)O_2-z and (Zr_2/8Ce_4.5/8Sn_0.5/8Pr_1/8)O_2-z were larger than those of the corresponding Sn-free samples. For the Ce-rich sample, Cu addition improved both the OSC value and its on-set temperature.

The reactive diffusion between Ta and a bronze was experimentally examined using sandwich diffusion couples consisting of Ta and a Cu-9.3Sn-0.3Ti alloy. The diffusion couples were isothermally annealed at a temperature of T = 1053 K for various times up to t = 1200 h. During annealing, a layer of Ta₉Sn is formed at the interface in the diffusion couple. The mean thickness of the Ta₉Sn layer increases in proportion to a power function of the annealing time. The exponent of the power function is equal to unity at t < 167 h but smaller than 0.5 at t > 167 h. This means that the transition of rate-controlling process occurs at t = 167 h. For the reactive diffusion between Ta and the bronze at T = 1053 K, the growth of Ta₉Sn is controlled by the interface reaction at t < 167 h but by the volume and boundary diffusion at t > 167 h.