MATERIALS TRANSACTIONS
Online ISSN : 1347-5320
Print ISSN : 1345-9678
ISSN-L : 1345-9678
Volume 42, Issue 5
Displaying 1-31 of 31 articles from this issue
  • Hiroshi Ohtani, Isamu Satoh, Masamitsu Miyashita, Kiyohito Ishida
    2001 Volume 42 Issue 5 Pages 722-731
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The present study was undertaken to examine the phase equilibria in the Sn–Ag–Bi ternary system, a promising candidate for use as a Pb-free solder. A regular solution approximation as well as the sublattice model was applied to describe the Gibbs energy for the individual phases. Themodynamic parameters were achieved by combining the experimental results obtained by differential scanning calorimetry and energy dispersive X-ray microanalysis in the present study with data from the literatures. The calculation of phase diagrams showed that there are three invariant reactions in the liquidus surface of the ternary system and that the ternary eutectic point falls at 0.5%Ag, 53.9%Bi and 139°C.
    Download PDF (1262K)
  • Takahiro Miki, Naotaka Ogawa, Tetsuya Nagasaka, Mitsutaka Hino
    2001 Volume 42 Issue 5 Pages 732-738
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    Activities of the constituents in one of the Pb-free solder systems, Ag–In–Sn, were studied experimentally using a mass spectrometer. Ion current ratios of Ag to In were measured in the temperature range 1273–1523 K . From the experimental results and the assessed thermodynamic properties of molten Ag–Sn, Ag–In, and In–Sn binary alloys, the excess Gibbs energy of liquid Ag–In–Sn ternary alloy was determined.
    Download PDF (835K)
  • Seok-Hwan Huh, Keun-Soo Kim, Katsuaki Suganuma
    2001 Volume 42 Issue 5 Pages 739-744
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The effect of adding Ag up to 1 mass% on the microstructural and mechanical properties of Sn–Cu eutectic solder alloy was examined. Without Ag, primary β–Sn grains are surrounded by the eutectic network band of Cu6Sn5 needle precipitates/β–Sn. With increasing Ag content, the primary β–Sn grain size and the eutectic network size become finer. In the eutectic band fine Ag3Sn particles appear in addition to Cu6Sn5 precipitates. The DSC experiment revealed the presence of four endothermic reactions on heating for Sn–Cu–Ag alloys; the two peaks near 217°C correspond to the Sn–Cu–Ag ternary eutectic melting reaction and those at 223–225°C/224–226°C are for Sn–Cu binary melting. The 0.2% proof stress and tensile strength decrease with the addition of 0.1 mass%Ag and then gradually increase up to 1 mass%Ag. Even with 1% Ag, they are less than the values for a Sn–0.7Cu binary alloy. In contrast, elongation increases with increasing Ag content up to 1%. Thus, the addition of Ag to Sn–0.7Cu alloy can effectively improve its ductility. The strain rate dependence of 0.2% proof stress of Sn–0.7Cu–0.5Ag is similar to that of Sn–Ag eutectic alloy but is different from that of Sn–Cu eutectic alloy. A small amount of added Ag results in a change of the deformation mechanism of Sn–Cu alloy.
    Download PDF (2471K)
  • Tadashi Takemoto, Makoto Miyazaki
    2001 Volume 42 Issue 5 Pages 745-750
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    A wetting test was conducted using a wetting balance tester to evaluate the wettability of Sn–Ag and Sn–Zn lead-free solders with the addition of bismuth or indium up to 25 mass%. The base metal and flux used were oxygen-free high-conductivity copper and rosin-based flux containing 0.2 mass%Cl and 35 mass% solid rosin diluted with isopropyl alcohol. The wetting time decreased with increases in the amount of additional elements at a fixed test temperature. The wetting force had only a slight dependence on the element content; Sn–Zn alloys exhibited less wetting force than Sn–Ag. The plots between the excess temperature above liquidus, (test temperature)–(liquidus), and wetting time correlated well. An increase in the excess temperature decreased the wetting time, indicating that the wetting time is closely controlled by the temperature-rising stage of the wetting test specimen. Therefore, a comparison of the contact angles is recommended for precisely evaluating the wettability on lead-free solders. The interfacial tension between the soldering flux and molten solder was measured by a simple method using a wetting balance tester to obtain the contact angle from the wetting balance test. The calculated contact angle in Sn–Ag system solders is smaller than that in Sn–Zn system solders, indicating that Sn–Ag system solders exhibit superior wettability.
    Download PDF (1080K)
  • Chang-Bae Lee, Seung-Boo Jung, Young-Eui Shin, Chang-Chae Shur
    2001 Volume 42 Issue 5 Pages 751-755
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    In the present work, the wettability and interfacial tension between Cu-substrate and Sn–Bi solder were examined. The variables of this experiment used the content of Bi, types of flux and soldering temperature. Through analyzing the experimental result concerning wettability, it was found that the role of Bi was diminishing wettability of Cu substrate during soldering. The increasing Bi contents made liquid solder-Cu substrate interfacial tension (γsl) go up. Consequently, the increasing liquid solder-Cu substrate interfacial tension (γsl) led to the degradation of wettability in solder. The rates of wetting can be controlled by the interfacial energy between liquid solder and substrate. This paper describes the effect of Bi element on the wettability of Sn–Bi solders.
    Download PDF (735K)
  • Keisuke Uenishi, Toshio Saeki, Yasuhiro Kohara, Kojiro F. Kobayas ...
    2001 Volume 42 Issue 5 Pages 756-760
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    Effect of Cu in the eutectic Sn–37 mass%Pb or Sn–3.5 mass%Ag solder ball has been investigated on the microstructure and the shear strength of BGA joints with Ni/Au electroplated Cu pad after reflow and the subsequent heat exposure at 423 K . For joints using both Cu free Sn–Pb and Sn–Ag solder balls, a metastable (Au, Ni)Sn3 reaction layer was formed on the interface between the solder and the pad after reflow soldering and the subsequent heat storage at 423 K, while stable η-(Au, Cu, Ni)6Sn5 reaction layer was formed for the joints using Cu containing solder balls. The growth rate during heat exposure was much slower for η reaction layer than that for (Au, Ni)Sn3 reaction layer. All the joints after a heat exposure fractured on the lower shear load than as reflowed joints, but the shear strength of the joint using Cu containing ball was less degraded than that using Cu free solder balls, because η reaction layer grows slower and plays a roll as a good barrier layer to suppress the Ni–Sn interaction. Almost the same effect was confirmed for the BGA joints using Cu cored solder balls, since Cu dissolves into the solder during reflow soldering.
    Download PDF (2367K)
  • Kunihiro Noguchi, Mayumi Ikeda, Isao Shimizu, Yoshito Kawamura, Yasuhi ...
    2001 Volume 42 Issue 5 Pages 761-768
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The microstructure and the strength property of Sn–3.5 mass%Ag and Sn–2.5 mass%Ag–0.8 mass%Cu solder ball bonding formed at various cooling rates were investigated. The grain size of Sn phase and the lamellar space depended on the cooling rate Rc. The reaction layer was thinner and the grain size of Sn phase coarsened when solder balls containing Cu were used. The reaction layer formed between Sn–3.5 mass%Ag and Cu pad was composed of (Ni–Au, Ag)3Sn4, (Ni–Au, Ag)3Sn2 and (Ni–Au, Ag)3Sn precipitates. (Cu–Ni, Au, Ag)6Sn5 was formed in the bonding interface when the ball with Cu was used. The lamellar space in both balls became smaller at faster cooling rates and was proportional to Rc(−1⁄2). The micro Vickers hardness of the solder ball with Cu decreased slightly at the slower cooling rates because of the grain size and lamella coarsening. The shear strength of joint using Sn–2.5 mass%Ag–0.8 mass%Cu was a little higher than that using Sn–3.5 mass%Ag at the slower cooling rate because of the fine lamellar structure.
    Download PDF (3837K)
  • Jia Ning Hu, Hiroyuki Tanaka, Toshio Narita
    2001 Volume 42 Issue 5 Pages 769-775
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The hardness of a Sn–Bi eutectic solder was measured as a function of aging time at 233, 298, 323, and 373 K for up to 15.4 Ms, and the time dependence of the hardness could be divided into three stages. At 373 K the hardness decreased up to 259.2 ks (the initial stage) and then increased between 259.2 ks and 1.8 Ms (the transient stage), and beyond 1.8 Ms (the final stage) the hardness tended to decrease gradually. These critical times, 259.2 ks and 1.8 Ms at 373 K, shifted to later at 323 and 298 K, while at 233 K there was little change in the hardness up to about 1 Ms after which hardness decreased rapidly. At an aging temperature of 373 K the area fractions and widths of the Sn and Bi layers in the lamellae structure as well as their lattice constants were determined by microstructural observation, electron probe microanalysis, and X-ray diffraction analysis. The decrease in hardness in the initial aging stage may be due to the disappearance of Bi particles, followed by dissolution of Bi into the Sn matrix. The rapid increase in hardness in the transient stage could arise from changes in the Sn content in the Bi phase. The decrease in hardness in the final stage could be mainly caused by coarsening the lamellar structure and increases in the area fraction of Sn phase.
    Download PDF (2968K)
  • Yasuyuki Miyazawa, Tadashi Ariga
    2001 Volume 42 Issue 5 Pages 776-782
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The microstructural change and hardness, which strongly influence the life of soldered joints, were investigated by microstructure observations for three typical promising alloys as Pb free solder, Sn–3.5 mass%Ag, Sn–8.8 mass%Zn and Sn–57 mass%Bi alloys, and Sn–38.1 mass%Pb for comparison. The solders of Sn–Ag and Sn–Zn eutectic alloys and Sn–Pb eutectic alloy were found to age-soften during storage at 25–100°C (298–373 K). It seems that the coarsening of the microstructure and re-crystallization caused the age-softening of the Sn–Ag and Sn–Zn alloys. The hardness of the Sn–Bi eutectic alloy hardly changed during storage at 25–100°C (298–373 K) in spite of remarkable microstructural coarsening. It seems that the hardening effect caused by solid-solution hardening of Bi into Sn-rich solution and/or the precipitation hardening of Bi in the Sn-rich solution compensated the softening caused by the remarkable microstructural coarsening of the Sn–Bi alloy.
    Download PDF (3672K)
  • Won Kyoung Choi, Seung Wook Yoon, Hyuck Mo Lee
    2001 Volume 42 Issue 5 Pages 783-789
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The effect of In addition on the binary eutectic Sn–3.5Ag (compositions are all in mass% unless specified otherwise) solder alloy was investigated. The phase equilibrium and the thermodynamic behaviors were examined up to an indium level of 11.5. The addition of about 9In lowered the liquidus temperature below 210°C, and the microstructures of the solder and the morphology of secondary phases in the solder matrix changed accordingly. The secondary phases coarsened in aging at 150°C up to 500 h. However, the coarsening had little effect on the hardness of the solder alloys. Though the surface tension between the solder and the vapor increased with an increase in In content, the wetting angle over the Cu substrate decreased. This was explained by the increased driving force for formation of the intermetallic compound (IMC) at the interface between the Sn–3.5Ag–In solder and the Cu substrate and thereby the decreased surface tension between the solder and the substrate.
    Download PDF (2928K)
  • Ikuo Shohji, Fuminari Mori, Kojiro F. Kobayashi
    2001 Volume 42 Issue 5 Pages 790-793
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    Thermal cycle tests were performed for flip-chip joints with several lead-free solders in order to research the thermal fatigue behavior of the joints. A joint with Sn–3.5Ag solder was ruptured by solder cracking in the neighborhood of the joint interface. A joint with Sn–3.5Ag–0.76Cu solder was ruptured by interfacial cracking at the Al/Ni interface, and the lifetime before the rupture was shorter than that of the Sn–3.5Ag solder. Although the rupture modes were different, both joints showed a sufficient fatigue life to tolerate the alignment stress in the flip-chip bonding process. In the case of a joint with Sn–1.9Ag–0.52Cu–7.6Bi solder, an interfacial crack was observed in the joint immediately after reflow soldering. Bi segregation was found in the cracked interface. The cracked microstructure was similar to the structure of lifted-off joints often observed when soldering through-hole circuits with Bi-bearing solder. These thermal cycle test results with Sn–3.5Ag and Sn–3.5Ag–0.76Cu solders demonstrate that there is a possibility of using flip-chip joints with these solders for conventional electronic applications.
    Download PDF (1348K)
  • Akio Hirose, Toshio Fujii, Takeshi Imamura, Kojiro F. Kobayashi
    2001 Volume 42 Issue 5 Pages 794-802
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    QFPs (quad flat packages) with Sn–10Pb plated Cu lead or Au/Pd/Ni plated Cu lead were reflow-soldered using Sn–3.5Ag, Sn–3Ag–5Bi, Sn–3.5Ag–0.7Cu, Sn–3.5Ag–2.5Bi–2.5In and Sn–37Pb solders. The strength and the microstructure of the solder joints were examined after an exposure test at 398 K . Although the strength of the solder joint with the Sn–10Pb plated Cu lead using the Sn–3Ag–5Bi solder significantly decreased with increasing holding time at 398 K, the strength of the other Sn–Ag based solder joints was comparable to that with the Sn–37Pb solder before and after the high temperature exposure test. The reaction layers formed at the interface between the solder and the Cu pad consisted of Cu6Sn5 and Cu3Sn in the joints with Sn–10Pb plated Cu lead and consisted of only (Cu, Ni, Pd)6Sn5 in the joints with the Au/Pd/Ni plated Cu lead after the exposure at 398 K up to 7.2 Ms for all five solders. The growth kinetics of the reaction layers obeyed the parabolic law except for the joint with the Sn–10Pb Plated Cu lead using the Sn–3Ag–5Bi solder, in which the growth of the reaction layer deviated from the parabolic law and accelerated beyond the holding time of 1.8 Ms. The unusual growth of the reaction layer, which resulted from a liquid phase forming ahead of the reaction layer and penetrating the grain boundaries of the reaction products, caused the degradation in the strength of this solder joint. The liquation was caused by the enrichment of Bi and Pb ahead of the reaction layer during the high temperature exposure to the extent where melting occur at the holding temperature above the Sn–Bi–Pb ternary eutectic point of about 370 K . The Sn–3Ag–5Bi solder is, therefore, considered to be unsuitable for assembling packages with the Sn–10Pb plated lead because of degradation in reliability during the high temperature exposure above 370 K.
    Download PDF (2760K)
  • Shinichi Terashima, Tomohiro Uno, Eiji Hashino, Kohei Tatsumi
    2001 Volume 42 Issue 5 Pages 803-808
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The failure mechanism and improvement in the thermal reliability were studied of a new flip chip interconnection system using Au bumps on Si chips and Sn bumps on substrates. The thermal reliability of this flip chip depended on the diffusion behaviour at the Au/Sn bonded interface. Kirkendall voids, which were formed at the Au/Sn bonded interface, induced cracks and impaired the reliability. By the addition of Ag into the Sn bumps, the reliability was improved, due to the formation of an Ag concentrated layer at the Au/Sn bonded interface which acted as a barrier to Au–Sn further interdiffusion and suppressed the Kirkendall voiding. The AuSn phase grew faster than the other Au–Sn intermetallic compounds did at the Au/Sn bonded interface after annealing. Activation energy for AuSn growth was 29.3 kJ·mol−1.
    Download PDF (1755K)
  • Young-Eui Shin, Kyung-Woo Lee, Kyong-Ho Chang, Seung-Boo Jung, Jae&nbs ...
    2001 Volume 42 Issue 5 Pages 809-813
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    This study was focused on the numerical prediction of the thermal fatigue life of a μBGA (Micro Ball Grid Array) solder joint. Numerical method was performed to predict three-dimensional finite element analysis with various solder alloys such as Sn–37 mass%Pb, Sn–3.5 mass%Ag, Sn–3.5 mass%Ag–0.7 mass%Cu and Sn–3.5 mass%Ag–3 mass%In–0.5 mass%Bi during a given thermal cycling. Strain values, which were obtained by the result of mechanical fatigue tests for solder alloys, were used to predict the solder joint fatigue life using the Coffin-Manson equation. The numerical results showed that Sn–3.5 mass%Ag with the 50 degree ball shape geometry had the longest thermal fatigue life in low cycle fatigue. Also a practical correlation for the prediction of the thermal fatigue life was suggested by using the dimensionless variable γ.
    Download PDF (859K)
  • Jun Seok Ha, Choon Sik Kang, Jae Yong Park, Jae Pil Jung
    2001 Volume 42 Issue 5 Pages 814-819
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    This paper introduces the partial melting soldering process as a new soldering technology and investigates the mechanism, mechanical properties, and interface reactions between Cu and off eutectic binary alloys: Sn–7 mass%Ag, Sn–3 mass%Cu, and Sn–2 mass%Ni. In order to show the possibility of the partial melting soldering, reflow was conducted under various temperature conditions between the eutectic and the liquidus temperatures. As a result, intermetallic compound layers were formed at the joint in their partial melting zone. In cases of the Sn–7Ag and Sn–3Cu solders, Cu6Sn5 intermetallic layers were formed, while the intermetallic layer formed between Sn–2Ni and Cu plate was a ternary compound of Sn–Ni–Cu. Hardness test showed that partial melted joint was harder than the full melted joint, indicating that the solid phase in the mushy zone had an effect of reinforcement. According to the aging experiments, the growth rate of intermetallics of partial melted joint was lower than that of full melted one. This indicates that the solid phase in mushy zone also prevented excessive growth of the intermetallics. Finally, shear test had been made to evaluate joint strength, showing that the strength of the partial melted joint was almost 80% that of a full melted joint.
    Download PDF (2854K)
  • Chang-Bae Park, Soon-Min Hong, Jae-Pil Jung, Choon-Sik Kang, Yong-Eui ...
    2001 Volume 42 Issue 5 Pages 820-824
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    UBM-coated Si-wafer was fluxlessly soldered with glass substrate in N2 atmosphere using plasma cleaning method. The bulk Sn-37 mass%Pb and Sn–3.5 mass%Ag solders were rolled to the sheet of 100 \\micron thickness in order to achieve bonding to Si-wafer by fluxless 1st reflow process. The oxide layer on the solder surface was analyzed by AES (Auger Electron Spectroscopy). After 1st reflow the Si-wafer with a solder disk was plasma-cleaned, and soldered to glass by 2nd reflow soldering process without flux in N2 atmosphere. The thickness of oxide layer decreased with increasing plasma power and cleaning time. The optimum plasma treatment condition in this study was 500 W for 12 min and at this condition, 100% bonding ratio for Sn–3.5 Ag and over 80% bonding ratio for Sn–37Pb solder were achieved. The intermetallic compound of continuous Cu6Sn5 was observed along the Si-wafer/solder interface but discrete Cu6Sn5 along the glass/solder interface and the different shapes of Cu6Sn5 were caused by different thickness of Cu as a pad. The fracture of the tensile test specimen occurred at not only solder/UBM and solder/TSM interface but also in Si-wafer and glass substrate.
    Download PDF (2133K)
  • Takahiro Sekigawa, Kazuyuki Oguri, Jiro Kochiyama, Kazuyuki Miho
    2001 Volume 42 Issue 5 Pages 825-828
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    C/C composites with an oxidation protection SiC coating have been evaluated by using high frequency air plasma flow at temperatures between 1554 and 1945K, and at pressures between 1.0 and 5.0 kPa. The oxidation behavior of the coatings after heating tests was investigated. The thickness reduction of the coating was 5.5×10−2 \\micron s−1 at 1945 K and 2.5 kPa. The coating recession was larger at higher temperatures and lower pressures. X-ray photoelectron spectroscopy of the coating surface indicated that SiO2 film was formed on the coating surface below 1856 K, and that high SiC recession rates were observed above 1884 K . At 2.5 kPa under subsonic plasma flow conditions, the active oxidation occurred above 1884 K and gradually changed from the passive oxidation with increasing temperature.
    Download PDF (1637K)
  • Pavlin Dakev Mitev, Masatoshi Saito, Yoshio Waseda
    2001 Volume 42 Issue 5 Pages 829-832
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    Three effective pair potentials of Ag–Br, Br–Br and Ag–Ag pairs in molten AgBr have been calculated from measured partial structural data by applying the modified hypernetted-chain equation coupled with a predictor-corrector method. The present results of pair potentials were also compared with model potentials originally proposed by Vashishta and Rahman. The resultant potentials suggest two characteristic features. One is the relatively shallow minimum in the interaction of Ag–Br pair in the first nearest neighbor region at 0.27 nm, in comparison with the model potential case. The other is both positive and negative deviation from the simple asymptotic form detected in the interaction of Br–Br pair in the range of 0.3∼1.1 nm. These results may be attributed to a particular structural feature of molten AgBr, suggesting not only the purely ionic character.
    Download PDF (661K)
  • Byoung Hee Kim, Dong Soo Suhr
    2001 Volume 42 Issue 5 Pages 833-837
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The effect of WC particle size and fuel/oxygen (F/O) ratio on the characteristics of HVOF-sprayed WC–17 mass%Co coatings was investigated. The microstructure of the coatings was dense regardless of F/O ratio, but became slightly inhomogeneous due to incomplete formation of splats with coarsening WC particle size. The unmelted WC particle size in the coatings manufactured with 1.53 \\micron mean particle size varied from 0.82 to 0.89 \\micron depending on F/O ratio, and the unmelted WC particle size in the coatings manufactured with 6.3 \\micron mean particle size ranged from 2.87 to 3.03 \\micron depending on F/O ratio. The sintered WC–17 mass%Co powder for thermal spraying consisted of WC, fcc-Co and Co3W3C phases, but the HVOF-sprayed coatings consisted of WC, W2C and W phases. The rate of phase change during spraying was significantly influenced by both raw particle size of WC and F/O ratio. As the amount of the decomposed W2C phase was more, the hardness of the coatings was higher. On the contrary, as the amount of the decomposed W phase was more, the hardness of the coatings was rather lower. Consequently, it was revealed that the W2C phase was preferable but the W phase was detrimental to the hardness of the coatings.
    Download PDF (1802K)
  • Ken Unno, Hidehiro Kudo, Atsunori Kamegawa, Hitoshi Takamura, Masuo Ok ...
    2001 Volume 42 Issue 5 Pages 838-841
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The synthesis of vapor-grown carbon fibers (VGCFs) using the nanocrystalline Fe75Si15B10 alloy as a novel catalyst has been investigated, and the growth process of VGCFs was discussed. The nanocrystalline Fe75Si15B10 alloy with fine bcc-Fe grains less than about 20 nm was prepared by the melt spinning technique followed by annealing. VGCFs with a diameter of less than about 40 nm were successfully prepared using the nanocrystalline Fe75Si15B10 catalyst at 773 K under 80%CO–20%H2 atmosphere. High-resolution transmission electron microscopic characterization indicated that the VGCFs catalytically generated over the Fe75Si15B10 nanocrystal surfaces at both 773 K and 873 K most likely have a tubular-type structure, in which the graphite sheets were stacked parallel to the fiber axis. The Fe75Si15B10 catalyst particles appear to be made of the single-phase Fe3C grains, which is in contrast to the Fe91Zr7B2 catalyst particles previously studied by us. The formation of Fe3C leads to the brittleness of catalyst particles and results in a growth of filamentous carbon.
    Download PDF (1697K)
  • Takeshi Yamada, Takayuki Tsuzuku, Hiroaki Sato
    2001 Volume 42 Issue 5 Pages 842-849
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    Continuous fiber reinforced titanium matrix composites (TMCs) are attractive as potential structural materials for aerospace applications, because of their high specific strength and stiffness. However, TMC parts have not yet been put into practical use due to their limited damage tolerance and the enormous production cost resulting from preliminary forming and elaborate tooling for consolidation. In order to reduce the production cost, superplastic TMC sheets (SiC/Ti–4.5Al–3V–2Mo–2Fe) were developed, and the deformation characteristics and cavitation behavior were investigated. In this study, plasma-sprayed preforms, which replace conventional woven preforms, were newly developed to improve the cavitation resistance. The superplastic deformation characteristics and the cavitation behavior of SiC/Ti–4.5Al–3V–2Mo–2Fe composites made of sprayed preforms (spray composites) were investigated. The fiber spacing in the spray composites was controlled more uniformly than that in composites fabricated from woven preforms (woven composites), and two closely spaced fibers were rarely found in the spray composites. The spray composites showed cavitation-induced superplasticity with the strain rate sensitivity exponent (m) being 0.58 at 1048 K, over the strain rate range of 5×10−5 s−1 to 1×10−3 s−1. This indicated greater elongation than that in the case of the woven composites, which may be due to the improved uniformity of fiber distribution. Most of the defects observed in the deformed spray composites were defects due to fiber/matrix interface separation (type-1). On the other hand, those observed in the deformed woven composites were defects caused by weaving ribbons (type-2) or narrow fiber spacing (type-3). They were less related to the deformation conditions. This suggested that the sprayed preforms were effective in preventing the occurrence of type-2 and type-3 defects. The occurrence of type-1 defects was strongly related to the maximum flow stress during deformation, and the occurrence of such defects rapidly increased with the maximum flow stress exceeding the threshold stress of 18 MPa.
    Download PDF (3388K)
  • Il Park, Toru H. Okabe, Yoshio Waseda, Hyo Shin Yu, Oh  ...
    2001 Volume 42 Issue 5 Pages 850-855
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    An electronically mediated reaction (EMR) has been explored to produce a niobium powder semi-continuously by the magnesiothermic reduction of niobium oxide. Nb2O5 feedstock and reductant magnesium alloy were isolated electronically in a molten CaCl2 bath at 1173 K . The feed was freshly recharged twice, but the bath was reused without replenishment during three repeated tests, providing an external path for electron flow. The current flow through the external path between the feed (cathode) and the reductant (anode) was monitored. A small current between 0.05 and 0.1 A was measured in the external circuit. A niobium powder was readily obtained, but its purity decreased from 93 to 82% Nb as the reduction number increased. This, however, demonstrated the possibility of semi-continuous production of niobium powders by the EMR without the direct physical contact between the reactants. The mechanism of the magnesiothermic reduction of Nb2O5 in the molten salt was discussed, using an isothermal chemical potential diagram.
    Download PDF (1618K)
  • Kazuyuki Oguri, Takahiro Sekigawa, Jiro Kochiyama, Kazuyuki Miho
    2001 Volume 42 Issue 5 Pages 856-861
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    Catalycity (kw) of CVD-SiC coated C/C is measured by plasma test using a Russian Plasmatron. The catalycity increased with increasing temperature and with decreasing pressure. The catalycity in Air (kwAir) was ranged between 1.0 and 4.5 m s−1. This shows the applicability of the CVD-SiC coating to the re-entry space vehicles of HOPE-X . The surface of the specimen was covered with SiO2 after the tests. The low kw was caused of the property of SiO2.
    Download PDF (991K)
  • Kiyoshi Nogi, Takuji Kita
    2001 Volume 42 Issue 5 Pages 862-869
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    Thermoelectric materials of α-Fe0.91Mn0.09Si2.5+x mass%Cu (0≤x≤2) were produced by spark-plasma sintering (SPS) and their thermoelectric properties were investigated. The densification was completed in sixty seconds after the temperature reached 1173 K . Liquid phase sintering most likely occurred during the heating because the addition of copper promoted the densification. The relative density of the sintered specimen was between 88 and 90%. When (α+ε)-phase powder, which corresponds to the β-phase composition, was used as the starting material, the eutectoid reaction (α→β+Si) mainly occurs at 943 K . However, there was no β-phase in the α-phase specimen annealed under the same conditions. It was clear that the ε-phase accelerates the eutectoid reaction. Copper accelerates transformation to the β-phase. The eutectoid reaction was completed within only 300 s at 943 or 973 K, and also occurred at 1003 K but the rate was very low. Compared with the (α+ε)-phase specimen, the power factor of the α-phase specimen was very low, because the electrical resistivity of the α-phase specimen was larger than that of the (α+ε)-phase specimen. The silicon grains dispersed in the matrix should be an obstacle to the electrical conduction. The annealing conditions were controlled so that the coarsening of the silicon particles would improve the electrical resistivity. The electrical resistivity was not improved, though the silicon grains could be coarsened. An excess amount of copper and an excess annealing time had a negative influence on the Seebeck coefficient and the power factor. This phenomenon was also observed in the (α+ε)-phase specimen. Fe0.91Mn0.09Si2.5+20 mass%Cu alloy was produced to analyze the reaction between copper and Fe0.91Mn0.09Si2.5 in detail. It was revealed from EPMA results that the composition of the β-phase was shifted from the nominal composition. The copper-rich phase is considered to a eutectic composition of Cu–30 at%Si, and it has little thermoelectric properties. The optimum composition was Fe0.91Mn0.09Si2.5+0.5%Cu, and the optimum annealing conditions were 300 s at 973 K . The power factor of the optimum specimen was about 74% of that of the optimum (α+ε)-phase specimen.
    Download PDF (3046K)
  • Byoung Hee Kim, Dong Soo Suhr
    2001 Volume 42 Issue 5 Pages 870-875
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    This study was performed to investigate the influence of hydrogen-to-oxygen ratio (F/O=3.2, 3.0, 2.8, 2.6) and heat treatment on the characteristics (microstructure, decomposition rate, hardness, and mass loss by erosion) of HVOF (high velocity oxy-fuel)-sprayed coatings using NiCr-coated Cr3C2 composite powder. Decomposition of the Cr3C2 occurred regardless of F/O ratio during spraying. The degree of decomposition of the coatings was increased with decreasing the F/O ratio and with increasing heat-treating temperature. Consequently, after heat treatment at 1000°C about 90% of the coatings adjacent to the surface consisted of Cr7C3 phase and also about 70% of the inner coatings was this phase. Microhardness of the as-sprayed coatings was varied from 987 to 1037 kgf/mm2 with F/O ratio, and the coatings sprayed with F/O=3.0 was the highest (Hv300=1037), and the hardness was increased to 1340 kgf/mm2 after heat treatment at 600°C but it was suddenly decreased after heat treatment from 800 to 1000°C. Erosion resistance was the highest at the coatings sprayed with F/O=3.0. And erosion resistance was improved about 40% than that of as-sprayed coatings after heat treatment at 600°C.
    Download PDF (3720K)
  • Yoshihiko Yokoyama, Nobuyuki Nishiyama, Kenzo Fukaura, Hisakichi Sunad ...
    2001 Volume 42 Issue 5 Pages 876-880
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    High-cycle fatigue strength of a Pd40Cu30Ni10P20 \\citerf1 bulk glassy alloy was examined using the rotating-beam fatigue test. The Pd40Cu30Ni10P20 bulk glassy alloy showed a distinct knee at an applied stress of 0.15 σBB: tensile strength of 1600 MPa) and a cycle number N=4×107. However, in the equal applied stress of 0.15 σB, some samples fractured over the knee cycle numbers. It is likely that the origin of the fracture is to the partial embrittled hardening resulting from crystallization or structural relaxation, which is probably caused by the thermal effect from plastic deformation region. Such a partial hardening by the thermal effect characterizes the fatigue feature of the bulk glassy alloy. It is revealed that no fatigue limit is seen on the Wöhler curve of the Pd40Cu30Ni10P20 bulk glassy alloy.
    Download PDF (2702K)
  • Wataru Takakura, Shoji Ikeda, Yuji Ueda
    2001 Volume 42 Issue 5 Pages 881-885
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The processing speed of the computer needs speeding up because of an increase in the packing density, and the frequency for recording should shift to the higher side. The eddy current loss is brought about by the use of the high frequency, and it can be reduced by increasing the electrical resistivity of the material. The multilayers produced by alternate combination of the metal and the insulator is thought to be as one of the effective methods for preparing the high resistivity material. In this work, the electrical resistivity of the Fe/Al2O3 multilayers produced by varying the thickness of the Fe and Al2O3 films have been discussed. Moreover, in order to produce the magnetic anisotropy the film was prepared by varying the incidence angle of the vapor deposition. The resistivity increases with increasing Al2O3 thickness. The resistivity shows a tendency of having a maximum value in the vicinity of angle θ=45° with further falling and rising behavior. On the other hand, the ratio of the resistivity for the current along the projectional direction of oblique incidence to that perpendicular to the projectional direction has a maximum value around the angle of 45° and afterwards decreases gradually. Theoretical values obtained by considering atomic arrangement are in agreement with experimental results.
    Download PDF (610K)
  • Limin Wang, Chunfei Li, Akihisa Inoue
    2001 Volume 42 Issue 5 Pages 886-889
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The glass-forming ability in Ni57−xTi23Zr15Si5Pdx (x=0, 5, 10, 12.5, 15 and 20) alloys as a function of Pd concentration was examined by melt-spinning and copper mold casting methods. All the melt-spun ribbons are in a glassy state. ΔTx(ΔTx=TxTg) and TgTm were used to evaluate the glass-forming ability, where Tg,Tx and Tm represent the glass transition, crystallization and melting temperatures, respectively. The observation of large ΔTx and TgTm values in the 12.5 at%Pd alloy implies a high glass-forming ability. At this composition, a cylindrical bulk alloy with a diameter of 2 mm was verified to be in a single glassy state. For the bulk cylinder of 3 mm in diameter, the volume fraction of glassy phase is approximately 92%. The compression fracture strength (σ) and Vickers hardness (Hv) of the bulk glassy alloy are 2100 MPa and 620, respectively. These results demonstrate the good mechanical performance of the present Ni-based bulk glassy alloy, which is prominent for future application.
    Download PDF (593K)
  • Jin Onuki, Yasunori Chonan, Takao Komiyama, Masayasu Nihei, Masateru S ...
    2001 Volume 42 Issue 5 Pages 890-893
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    This paper examines microstructures and the reliability of solder joints made by a new void-free process and compares them to the conventional H2 process. A new two step void-free process was recently developed for enhancing the reliability of large area, high power IGBT modules. First, Ar+ were used to clean the surface of plated Ni films on a substrates followed by coating with 0.5 \\micron Ag film, and second, 50 mass% Pb–Sn solder sandwiched between the two substrates was heated at 503 K in vacuum for 5 min. and then cooled in a N2 atmosphere. The fatigue life time of the solder joints formed by the new process was found to be 3 times longer than those made the conventional process. We also found the solder joints from the new process to have considerably smaller amounts of micro and interfacial voids in comparison by those made with the conventional method.
    Download PDF (2240K)
  • Kenji Hiraga, Tetsu Ohsuna
    2001 Volume 42 Issue 5 Pages 894-896
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The structure of an Al–Ni–Fe decagonal phase, which was found as a highly-ordered decagonal quasicrystal with periodicity of 0.4 nm in an Al71.6Ni23.7Fe4.7 alloy, has been studied by atomic-scale observations of high-angle annular detector dark-field scanning transmission electron microscopy (HAADF-STEM). It has been found that a large columnar cluster of atoms with a decagonal section of about 3.2 nm in diameter exists as a basic structural unit. A structural model of the cluster is proposed from HAADF-STEM images.
    Download PDF (1585K)
  • Kenji Hiraga, Tetsu Ohsuna, Kunio Yubuta, Sinya Nishimura
    2001 Volume 42 Issue 5 Pages 897-900
    Published: 2001
    Released on J-STAGE: September 06, 2005
    JOURNAL FREE ACCESS
    The structure of a crystalline approximant (called “PD3c phase”) in an Al71.5Co16Ni12.5 alloy annealed at 900°C for 120 h has been studied by high-angle annular detector scanning transmission electron microscopy (HAADF-STEM) and high-resolution transmission electron microscopy (HREM). The PD3c phase has an orthorhombic structure with lattice parameters a=5.2 nm, b=0.4 nm and c=3.7 nm, and the space group of Pmn21 (No. 31). Its structure is characterized as an ordered arrangement of columnar atom clusters with pentagonal symmetry, and can be interpreted as a crystalline approximant related to an ordered decagonal quasicrystal formed by the projection of a NaCl-type hypercubic lattice.
    Download PDF (1623K)
feedback
Top