日本金属学会誌
Online ISSN : 1880-6880
Print ISSN : 0021-4876
ISSN-L : 0021-4876
54 巻, 1 号
選択された号の論文の16件中1~16を表示しています
  • 原勢 二郎, 清水 亮
    1990 年 54 巻 1 号 p. 1-8
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Mechanism of secondary recrystallization in grain oriented silicon steel processed both with one-stage and two-stage cold rolling methods has been investigated using materials containing MnS alone or MnS and AlN as inhibitors.
    (1) In the one-stage cold rolling method, the Goss orientation has the specifically high intensity of Σ9 orientations in a primary recrystallized matrix. When AlN and MnS are used as inhibitors, the Goss orientation grows preferentially consuming grains with Σ9 orientation while other grains remain inhibited resulting in the Goss secondary recrystallization.
    When only MnS is used as an inhibitor, no secondary recrystallization occurred due to the poor inhibitor effect of MnS.
    (2) In the two-stage cold rolling method, the Goss secondary recrystallization evolves regardless of the inhibitors used. The required intensity level of the inhibitors for the secondary recrystallization in the two-stage cold rolling method is relatively low compared with that of the one-stage cold rolling method.
    The intensity of Σ9 orientation in relation to the Goss orientation is not the highest among other orientations, but the intensity of the Goss orientation itself has a peak value in a primary recrystallized matrix. Therefore, the PCNΣ9 value, a product of the intensity of Goss orientation and the intensity of Σ9 orientation in the primary matrix, is the highest except for the orientations having very high intensity of Σ1 orientations. The occurrence of the Goss secondary recrystallization in the two-stage cold rolling method is considered to be determined by the PCNΣ9 value as this value is associated with a probability of a Goss grain coming into contact with a grains of Σ9 orientation.
  • 柴柳 敏哉, 吉本 篤人, 堀 茂徳
    1990 年 54 巻 1 号 p. 9-16
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Void formation on grain boundaries in Cu-9.7 mass%Zn alloy has been investigated in connection with the grain boundary character. The alloy was prepared by melting electrolytic copper and pure zinc (99.9%) in high purity argon gas atmosphere. The ingot was hot rolled at 1093 K and then cold rolled to 1 mm in thickness. Tensile tests were performed on an Instron type testing machine in argon gas atmosphere. The initial strain rates were 3.5×10−3, 3.5×10−4 and 3.5×10−5 s−1. Microstructure and fracture surfaces of the specimens were observed with an optical microscope and a scanning electron microscope. Grain boundaries were characterized from their boundary misorientations by analysing ECPs (Electron Channeling Patterns) taken from two grains.
    A ductility minimum appeared at 673 K at any strain rate and the elongation to fracture decreased with decreasing strain rate. The observed ductility loss is attributed to the initiation, growth and coalescence of voids on grain boundaries during deformation. This void formation occurred at smaller strain with increasing test temperature. The minimum sigma value of the cavitated coincidence boundaries was 31 when deformed at 673 K and the initial strain rate of 3.5×10−3 s−1. Meanwhile, the sigma value was small as 9 at the initial strain rate of 3.5×10−4 s−1. Therefore, it was suggested that the strain rate affected the relationship between the propensity to void formation and the grain boundary character.
  • 嶋影 和宜, 遠藤 一彦, 佐藤 忠夫, 戸田 茂雄
    1990 年 54 巻 1 号 p. 17-24
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Reduction reaction and mechanism of samarium, europium, thulium and ytterbium oxides with metallic lanthanum have been investigated under the conditions of high temperature and vacuum atmosphere. Samples were used in the compact state prepared by pressing both rare earth oxide and metallic lanthanum powders.
    Reduction reaction of each rare earth oxide with metallic lanthanum parabolically proceeded with prolonged time. Reduction curve for each oxide could be arranged by using both logarithms of reactivity degree and time. The slope of a straight line obtained was about 0.5 for the reduction of each rare earth oxide. Apparant activation energies of the reduction reaction calculated from Arrhenius plots were 96 kJ/mol for samarium oxide, 236 kJ/mol for europium oxide, 154 kJ/mol for thulium oxide and 243 kJ/mol for ytterbium oxide, respectively.
    An electrochemical reduction model based on Wagner’s theory being used for high temperature oxidation of metal was applied for the reduction mechanism of each rare earth oxide with metallic lanthanum. From this model, a rate-determing step for the reduction reaction of each oxide could be considered to be the transfer process of lanthanum ion or oxygen ion in lanthanum oxide layer of reduction product.
  • 村山 順一郎, 薄木 智亮
    1990 年 54 巻 1 号 p. 25-32
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    The effect of ion bombardment conditions on depth resolution in a Cr/Ni thin multilayer sample has been studied using the secondary ion mass spectrometer (SIMS).
    The depth resolution is improved either with increasing mass or with decreasing velocity of primary ions. The resolution is mainly related to surface roughness which is induced by ion bombardment. The surface roughness was reduced in the case of low velocity cesium ion bombardment comparing with other ion species such as argon, nitrogen and oxygen. In oxygen ion bombardment, the resolution was extremely deteriorated at the interface from nickel to chromium, due to the selective sputtering phenomena in the nickel layer.
    Furthermore, the presence of hydrogen compound having high electron affinity at the interface complicates the interpretation of depth profile due to the change of positive secondary ion yield. To prevent the change in the yield, it has been considered that the bombardment of ion species having high electron affinity is very useful.
  • 須佐 匡裕, 永田 和宏, 後藤 和弘
    1990 年 54 巻 1 号 p. 33-40
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    The mechanism of the thermal oxidation of silicon substrates was studied, based on the determination of the oxygen pressure dependence of the solubility of oxygen in SiO2 films by the use of the double oxidation method.
    Silicon substrates were oxidized in a dry oxygen ambient. The results were analyzed, based on the linear-parabolic model. The activation energies were calculated to be 157 and 130 kJ/mol for the linear rate constant, kL, and the parabolic rate constant, kP, respectively. Both kL and kP were proportional to the first power of oxygen pressure, PO2.
    The double oxidation method was employed for determining the oxygen pressure dependence of the solubility of oxygen into the SiO2 film, C*, the interstitial diffusivity of oxygen in the SiO2 film, D, and the rate constant of the chemical reaction at the SiO2/Si interface, k. C* was proportional to the first power of PO2 but D and k were not dependent on PO2. On the basis of this result, it was concluded that oxygen gas dissolves into the SiO2 film as O2 molecules.
    The predominant mechanism of silicon oxidation was estimated to be as follows:
    \ oindentOxygen gas dissolves into the interstitial sites of the SiO2 film as O2 molecules. The O2 molecules diffuse toward the SiO2/Si interface through the interstitial sites. At the interface the reaction of Si+O2→SiO2 occurs.
  • 下尾 聰夫
    1990 年 54 巻 1 号 p. 41-47
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    In relation to the preparation of SiC, the mechanism of reduction of SiO2 by carbon black and activated carbon has been investigated. Using powdered mixtures with a molar ratio C/SiO2 of 3, the reduction rate has been determined by thermogravimetry and chemical analysis at temperatures from 1873 K to 2073 K under an atmosphere of argon.
    The reduction of SiO2 with carbon black is significantly slow. A small quantity of SiO2 evaporates as SiO and a high yield of SiC can be obtained. Jander’s rate equation is applicable to the rate of reduction. This result suggests that the dense particles of carbon black are continuously surrounded by SiC at the early stage of reduction. The activation energy is 544 kJ/mol. The rate-determining step is considered to be the diffusion of carbon in SiC.
    Activated carbon reduces SiO2 rapidly, because of its high porosity and reactivity. As the evolution of SiO is serious, a yield of SiC is low. The initial rate of reduction is controlled by the gas diffusion through the porous layer of carbon particles. The activation energy is 244 kJ/mol. At a late stage of reaction, the gas diffusion becomes difficult as a consequence of decrease in the population and the size of pores. The resistance of the diffusion in SiC layer formed around carbon to the overall rate of reduction cannot ignore.
  • 木村 直樹, 岩崎 智, 中村 泰
    1990 年 54 巻 1 号 p. 48-53
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Diffusion experiments were made by using diffusion couples between Sm and Co metals and Co metal and Co-Sm alloy with liquidus composition to clarify the reaction steps and mechanism occurring in the Reduction-Diffusion Process for making Sm-Co magnet alloys. Microscopic observation and EPMA analysis showed that Sm reacted easily with Co metal to produce liquid Sm-Co alloy, and subsequently diffused into Co metal to form multi-layers of intermetallic compounds such as Sm2Co17, SmCo5 and Sm2Co7. The thickness of the layer (W) was found to be proportional to the square root of the reaction time (t); W=(kt)1⁄2, where k is the rate constant of layer growth. The diffusion constant (D in m2 s−1) of Sm in a single phase was calculated from the observed values of k to be expressed by D=1.17×10−3exp(−265000⁄RT) and D=6.90×10−5exp(−220000⁄RT) for Sm2Co17 and SmCo5, respectively (the activation energy in J/mol).
  • 亀田 和夫
    1990 年 54 巻 1 号 p. 54-60
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    The activity of the liquid Cu-In system has been determined by the emf measurements of the following galvanic cell:
    ⊕Pt/Ni(s), NiO(s)/ZrO2(+CaO)/Cu-In(l), In2O3(s)/Ir/Pt\ominus (for NIn=0.20−0.85 at 937-1270 K).
    Deviations of the activities of indium and coppor from Raoult’s law are positive for the indium rich region and are negative for the copper rich side of the system. The present result agreed approximately with the data measured by Azakami and Yazawa.
    The α functions of indium for the system show the curves against NIn, suggesting that the alloys can not be considered as the regular solution.
    Heats of mixing are negative for copper rich alloys and slightly positive for indium rich alloys. The maximum and minimum values of the heats of mixing for the system are 0.40 kJ/mol at about NIn=0.76, −2.40 kJ/mol at about NIn=0.24, respectively.
    The thermodynamic properties of liquid alloys composed of IB metals (Cu, Ag and Au) and indium have been discussed in terms of the alloy solution theory of Engel. Namely, the activities and the heats of mixing for liquid alloys seem to be consistently explained on the assumption that the filling of the d shell is incomplete for the IB metals as well as the transition metals, whereas it is complete for indium.
    The activities and the heats of mixing of liquid Cu-In and Ag-In alloys are quite different from those of the liquid Au-In system, in which the activities show considerably negative deviations and the heats of mixing are exothermic, in the whole concentration range, owing to the larger differences of electronegativity factor in the gold alloys.
  • 平岡 裕, 入江 宏定, 岡田 雅年, 安永 暢男, 峰田 進栄, 池田 正幸
    1990 年 54 巻 1 号 p. 61-66
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    It is now well accepted that narrow bead welding, as in the case of the electron-beam application, and production of weldments with high precision and high quality can be achieved by the use of high-power laser beams. In this work, a continuous wave high-power CO2 laser was applied for molybdenum. First, effects of welding conditions such as laser power, welding speed and kind of shield gas were investigated. Secondly, ductile-to-brittle tansition behaviour of the laser-beam-welded joint obtained in the most optimum condition was evaluated and was compared with that of the electron-beam-welded joint.
    The material used for welding was sintered molybdenum sheets having lower levels of impurities such as oxygen. The laser source was a continuous wave CO2 laser with a nominal power of 2.5-7 kW. The welding was performed by a melt-run technique with the welding speed of 1.7-83 mm/s and the shield gases of He and Ar. Ductile-to-brittle transition behaviour, which represents the ductility of a material, was evaluated by a three-point bend test. After the test, the fracture surfaces were examined by a scanning electron microscopy.
    The results are summarized as follows:
    (1) Sound and full penetration of a molybdenum sheet of 1 mm thickness was achieved under limited welding conditions (power: 7 kW, welding speeds: 42-50 mm/s and shield gas: He) as compared with electron-beam welding.
    (2) Transition of the penetration mechanism from a thermal conducting mode to a key-hole mode occurred under the welding conditions in which the power was 7 kW and the welding speed was lower than 50 mm/s.
    (3) Ductile-to-brittle transition behaviour of the laser-beam-welded joint in an optimum condition was approximately the same as that of the electron-beam-welded joint.
  • 高橋 輝男, 橋本 雍彦
    1990 年 54 巻 1 号 p. 67-75
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    To prepare a dispersion-strengthened copper, mechanical alloying was carried out using pure Cu, Nb, Ta and graphite powders. After the mechanical alloying for 144.0 ks, the Cu-Nb-C solid solution+NbC mixture powder and the Cu-Ta-C solid solution+Ta+TaC mixture powder were obtained in the Cu-Nb-C and Cu-Ta-C systems, respectively. These powders were heat-treated in a vacuum at various temperatures from 873 to 1323 K. In the Cu-Nb-C and Cu-Ta-C systems, Nb and C or Ta and C reacted, resulting in precipitation of very fine NbC and TaC particles in copper. In the Cu-Nb-Ta-C system, very fine (Nb, Ta) C particles were precipitated. The mean particle sizes of the NbC and TaC particles were about 3.0 nm by the heat-treatment at 873 K for 3.6 ks and 15.0-20.0 nm at 1273 K for 3.6 ks. The mean particle sizes of these carbide particles have a tendency to decrease with decreasing heat-treatment temperature. The hardnesses of NbC- and TaC-dispersion-strengthened coppers were very high; for example, the room temperature hardnesses of Cu-2.50 vol%NbC and Cu-10.0 vol%NbC alloys were Hv 280 and 350, respectively, after the mechanical alloying for 72.0 ks followed by the heat-treatment at 873-973 K for 3.6 ks.
  • 小橋 真, 長 隆郎
    1990 年 54 巻 1 号 p. 76-83
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    The effect of anoying elements on dispersion of SiC particles into molten aluminum was investigated. The time required for incorporation of SiC particle into molten aluminum was defined as an incorporation time. The incorporation time of SiC was shortened by alloying magnesium, or by alloying titanium together with tin. These alloying elements have strong affinity for SiC and are able to form silicide as a result of reaction with SiC. Magnesium silicide was detected by using EPMA, although the existence of titanium silicide couldn’t be confirmed. The incorporation time of SiC was prolonged by adding zinc, copper or tin which has no affinity for SiC. Moreover, particulate dispersion is considerably influenced by surface-active elements. The incorporation time of SiC particle was shortened by adding lithium, that was a surface-active element with strong affinity for SiC, whereas it was prolonged by adding the surface-active elements such as lead or bismuth, having weak affinity for SiC. This is probably because lead or bismuth hinders the wetting between SiC particle and molten aluminum, adsorbing at the interface. The addition of silicon into aluminum prolonged the incorporation time of SiC, because the dissociation of SiC, that might be necessary for good wetting, was controlled by the presence of silicon.
  • 太田 芳雄, 中川 幸也, 辻 純二
    1990 年 54 巻 1 号 p. 84-92
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Single crystal turbin blades made of nickel-base superalloys hawe been frequently employed for modern air-craft engines. It was reported that recrystallization could take place during solution heat-treatment at highly strained parts of castings due to solidification constractions or at parts where various cold works were given. It was also observed that there were in differences susceptibility to the recrystallization among single crystal alloys. In the present study, the reduction of creep strength due to recrystallization was measured using single crystal samples (Alloy 454, TMS26) with intentionally induced recrystallized layers to various depths, and, the susceptibility dependence on the alloys (CMSX-2, TMS12-2, TMS26) was examined by reheating samples after tensile tests. The following conclusions were obtained.
    (1) The recrystallization took place in all of the alloys at temperatures above 1273 K.
    (2) The degree of the reduction of the creep rupture strength was related to the reduction of the load-bearing cross-sectional area due to recrystallization.
    (3) When the reheating temperature was between 1273 K and γ′ solvus of the alloy, the grain size of the recrystallized parts was small (<0.1 mm). However, when the temperature exceeded the γ′ solvus, the abnormal grain growth took place and the sample ruptured instanteneously.
  • 福田 正, 形浦 安治, 音谷 登平
    1990 年 54 巻 1 号 p. 93-100
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Effects of carbide and nitride precipitates on the mechanical properties of Fe-30%Cr alloys with C from 0.001 to 0.12% and of those with N from 0.004 to 0.13% were investigated. Charpy impact tests at various temperatures and tensile tests at room temperature were carried out for the specimens water-quenched from 1123 K.
    The results obtained are summarized as follows.
    (1) The ductile-brittle transition temperature determined by the impact test is well-correlated with the volume fraction of carbide and nitride precipitates, and not with the chemical composition of C, N, or (C+N). The transition temperature first increases remarkably with the volume fraction of these precipitates up to 0.4 vol%, and then remains almost constant for 0.6 vol% and above.
    (2) The temperature range of ductile to brittle transition in the absorption energy and temperature curve is narrow for the specimens containing carbide, and wide for the specimens containing nitride.
    (3) The strain at fracture in the tensile test decreases with increasing the volume fractions of carbide and nitride.
  • 佐藤 博, 梅田 高照, 木村 康夫
    1990 年 54 巻 1 号 p. 101-110
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Amorphous formation regions for the pseudo-ternary SrO-Fe2O3-B2O3 system were investigated using melt spinning and water quenching techniques. The specimens were heat treated, and then examined to determine the composition and temperature ranges where the M-type ferrites precipitated, and the relationship between crystalline phases and magnetic properties was discussed.
    The amorphous formation region by melt spinning is restricted within the composition range of less than 60 mol%Fe2O3 and 60-70 mol%SrO, and that by water quenching within the composition range of less than 25 mol%Fe2O3 and more than 30 mol%B2O3.
    The region, where M-type ferrites are crystallized by heating, is divided into two: one is the area where the particle size and coercive force of M-type ferrite depend on cooling rate, and the other is the area where they are independent of cooling rate. In the former, the composition ranges are determined where fine M-type ferrite particles of high coercive force (higher than 320 kA/m (4 kOe)) precipitate from the heat-treated amorphous phases. Although the maximum coercive force of 560 kA/m (7000 Oe) is obtained in the latter, fine particles smaller than a superparamagnetic particle size are also formed.
  • 岩村 栄治, 永山 勝久, 鈴木 俊夫, 梅田 高照
    1990 年 54 巻 1 号 p. 111-116
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    Microstructures and magnetic properties were studied of melt-spun Nd15Fe77B8 ribbons with various thicknesses. High coercive force up to 1824 kA·m−1 (22.8 kOe) took place for ribbons with 45∼100 μm in thickness. It was found that interspaces among Nd2Fe14B grains in these ribbons were filled by a non-magnetic phase. Electron diffraction patterns of the grain boundary phase were analyzed to reveal that this phase is of tetragonal structure with lattice constants of a=0.512 nm and c=0.801 nm. An orientation relationship was found to exist such that [001] of the boundary phase lies pallalel to the [001] direction of the host Nd2Fe14B phase.
    The boundary phase is metastable, since it was decomposed into fcc-Nd and a small amount of NdFe4B4 phase on heating above 690 K.
    Coercive force of these ribbons was discussed in terms of a nucleation controled coercivity mechanism. The smoothness of the interface between Nd2Fe14B and the boundary phase controls the nucleation behavior of reverse domains. Thus as-melt-spun ribbons exhibited smooth boundaries and high coercive force, heat-treated ones exhibiting rough boundaries and low coercive force.
  • 伊東 明俊, 三輪 敬之, 井口 信洋
    1990 年 54 巻 1 号 p. 117-124
    発行日: 1990年
    公開日: 2008/04/04
    ジャーナル フリー
    The shape recovery property of some glass-ceramics and sintered ceramics has been investigated with the high temperature range torsion test apparatus. The results are summarized as follows.
    (1) The shape memory phenomena occurred in both β-spodumene glass-ceramics and 2ZnO-B2O3 glass-ceramics. This shows that the shape memory phenomenon of ceramics is not caused by the peculiar characteristics of mica crystal.
    (2) The sintered ceramics such as mica (KMg3AlSi3O10F2), silicone nitride (Si3N4), silicone carbide (SiC), zirconia (ZrO2) and alumina (Al2O3) brought the similar phenomenon, indicating that the shape memory phenomenon of ceramics is not limited to glass-ceramics generated by the special manufacturing process, but found in the material which contains little glass phase.
    (3) The correspondence between the temperature range for the occurrence of shape memory phenomenon and the stress relaxation temperature range is recognized in all the materials studied in this paper.
    (4) The critical prestrain, within which the 100% shape recovery occurs, is about 0.1% with β-spodumene glass-ceramics, silicone nitride and zirconia. This value is 1/5 that of mica glass-ceramics.
feedback
Top