All issues
Volume 44, Issue 3
Displaying 1-17 of 17 articles from this issue
- |<
- <
- 1
- >
- >|
-
Toshio Yonezawa, Takayuki Imazu, Tomikane Saida, Hiroshi Susukida1980 Volume 44 Issue 3 Pages 223-229
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSIn order to clarify the effect of metallurgical factors such as surface treatment, cold working and micro structure on the corrosion and mass transfer of 304 stainless steel in sodium, a sodium exposure test was conducted for 10010 h at 360, 380, 530, 550 and 570°C with a sodium loop which was simulated to the secondary coolant system of FBR and the auxiliary sodium loop.
The following results were obtained:
(1) The corrosion and mass transfer rate of electrolytically polished specimens were larger than that of mechanically polished or pickled specimens.
(2) The effects of microstructure and cold working on the weight change and carburization rate were not remarkable.
(3) The effect of downstream on the corrosion and mass transfer rate was found remarkably for the standard 304 stainless steel specimens exposed at 570°C in sodium.
When oxygen content in flowing sodium was 10 to 20 wt ppm, it was also estimated that the corrosion rate of stainless steel was controlled by the following processes:
(a) diffusion of alloying elements in the layer of corrosion product, such as NaCrO2, at the surface of stainless steel
(b) diffusion of oxygen or carbon in the sodium sublayer
(c) diffusion of compounds of chromium and iron with oxygen in the sodium sublayer
(d) reaction between alloying elements and oxygen and carbon in sodium at the surface of stainless steel.View full abstractDownload PDF (549K) -
Masato Murata, Yoshihiko Mukai1980 Volume 44 Issue 3 Pages 230-238
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSStress corrosion cracking (SCC) of austenitic stainless steels does not necessarily show the same characteristics in different environments, although it has been generally attributed to so-called active path corrosion (APC). This means that the mechanism of APC is varied to some extent depending on environments, materials, etc.
In this paper, for a comprehensive treatment of the SCC mechanism of austenitic stainless steels, the possibility of a systematic arrangement of various SCC characteristics was studied. As the result, the characteristics of SCC were divided into the following classes according to their stress dependency; A group (SCC in a narrow sense), transition, B group (corrosion cracking-like). The typical environment of A group is 42%MgCl2 solution (416 K). In this case, the initiation of SCC depended on yield phenomena of materials, and the SCC growth rate da⁄dt on the stress intensity factor K. In transition, the environment was represented by 20%NaCl+1%Na2Cr2O7·2H2O solution (376 K). At the initiation stage of SCC, pitting corrosion preceded SCC. It is rather difficult to put da⁄dt in order by K, because the accuracy of K was lowered with increasing secondary cracks. In B group the environment was represented by 25 kmol/m3 H2SO4+0.5 kmol/m3 NaCl solution (R.T.). The initiation of SCC did not depend on yield phenomena of materials. It is almost impossible to arrange SCC propagation processes by K because of the appearance of SCC in the form of hair cracks.View full abstractDownload PDF (1125K) -
Sachio Oki, Seiroku Ogawa, Kunio Okabayashi, Keiichiro Shoji1980 Volume 44 Issue 3 Pages 239-245
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSA study have been made to clarify the relationship between microstructural parameters and fatigue limit of plain carbon steels having ferrite and martensite mixed structures. Results obtained are summarized as follows:
(1) The fatigue limit of plain specimens, σω0, depends on the microstructures. The relation between σω0 (MPa) and mean free ferrite path or ferrite particle size, λf (m), can be expressed as σω03.7·λf=1.32×104.
(2) The above equation is valid regardless of the hardness and volume fraction of martensite when the hardness of martensite is in the range of Hv 260 to 375.
(3) The equation shows a similar relation to σωcm·l=C, where σωc (MPa) is the alternating stress required to propagate a preexisting crack of length l (m). It can be explained by assuming that a critical crack length corresponding to the plain fatigue limit, σω0, given by l0=C·σω0−m is the same as the maximum length of the stage I slip-band cracks obtained in the plain specimens tested below the fatigue limit, and the length of these cracks is correlated with microstructural parameter, λf.View full abstractDownload PDF (1087K) -
Takaharu Iwadachi, Akihisa Inoue, Tetsuroh Minemura, Tsuyoshi Masumoto1980 Volume 44 Issue 3 Pages 245-254
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSNonequilibrium phases such as amorphous, chi (χ), epsilon (ε) and austenite have been found in Fe-X-C (X=Cr, Mo, W) ternary alloys quenched rapidly from the melts. The completely amorphous phase is formed in the composition ranges of 27∼50 at%Cr and 14∼20 at%C for Fe-Cr-C, of 6∼27 at%Mo and 13∼22 at%C for Fe-Mo-C, and of 10∼14 at%W and 18∼20 at%C for Fe-W-C. The high amorphous-forming tendency of these alloy systems seems to be due mainly to a remarkable increase of the glass transition temperature by the addition of Cr, Mo or W, reflecting in strong-attractive interactions between carbon and alloy elements. χ- and ε-phases are solid solutions containing large amounts of carbon and alloy elements, and their formation regions are located near the alloy compositions where the average group number of metallic atoms is in the vicinity of 7 and 7.7, respectively. The χ-phase has a complicated bcc structure and its lattice parameter increases from 0.892 to 0.905 nm with increasing carbon and alloy contents. The ε-phase is a hcp structure with lattice parameters of about a=0.27 nm and c=0.43 nm. Austenite phase has an ultra-fine grain size of about 0.2 μm and appears in the composition ranges of about 5∼10 at%C and 4∼30 at%Cr, 4∼13 at%Mo or 2∼8 at%W. The formation of this phase seems to become easy by the combination of three factors of the suppression of phase transformations in the solid state, the increase in solid solubility of carbon and alloy elements and the lowering of the Ms point by dissolution of these elements.View full abstractDownload PDF (1420K) -
Hirobumi Okabe, Hideo Ike1980 Volume 44 Issue 3 Pages 254-261
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSA study has been made on the isothermal oxidation behaviour of Fe-18Cr alloys containing 0.1 to 0.9 wt%Ti in 1 atm of oxygen at 1173∼1473 K for 86.4 ks. The weight gain increases with increasing Ti content. The addition of Ti in the range tested, therefore, appears to have no effects on improving the oxidation resistance, but rather to be deleterious. The scales formed have been mainly composed of Cr2O3, with Ti oxides distributing in the outer part of the scale, and in the alloy as internal oxides, respectively. When a small amount of Al, 0.5 wt%, is added to this alloy in addition to 0.5 wt%Ti, the internal oxidation of Ti has been considerably suppressed and has concentrated on the outer surface as well as in the inner part of the scale. The deleterious effects of Ti have also been observed in this case.View full abstractDownload PDF (1673K) -
Kensuke Sassa, Toshihiro Watanabe, Michio Inouye1980 Volume 44 Issue 3 Pages 262-271
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSThe rates of desulphurization from liquid slag to the gas phase have been studied under an oxidizing atmosphere. A series of experiments was carried out by introducing Ar-O2 gas mixtures on to the surfaces of those liquid slags containing sulphur, such as CaO-SiO2, CaO-Al2O3 and some CaO-SiO2-MO ternary slags as well as CaF2 bearing slags.
The results are as follows:
(1) The desulphurization reaction of those slags by oxidizing gases is divided into a rapid rate reaction at the initial stage and a slow rate reaction at the later stage. The former should be controlled by the chemical reaction at the surface of the slags and the latter by the diffusion of sulphur in slags.
(2) The rates of desulphurization of slag are proportional to the square root of partial pressure of oxygen and the rates of desuliphurization of the binary CaO-Al2O3 are higher than those of the CaO-SiO2 system. However, the rates of desulphurization for ESR slags are higher than those for CaO-Al2O3 slag. The rate constants for CaO-SiO2-MO ternary slags can be estimated from the sulphide capacity of the slags.
(3) It was found that the rates of desulphurization increase in proportion to the concentration of basic oxides (CaO, MgO, MnO, FeO) but decrease as acid oxides (SiO2, Al2O3) contents are increased. Particularly, the effects of MnO and FeO are significant.View full abstractDownload PDF (731K) -
Shiro Kohara, Norio Muto1980 Volume 44 Issue 3 Pages 271-276
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSThe compatibility of SiC-coated carbon fiber with aluminum was investigated. The carbon fiber was coated with SiC by chemical vapour deposition. The elastic modulus and the tensile strength of the carbon fiber increased after the SiC coating treatment.
The compatibility of carbon fiber with aluminum was assessed by strength testing after heating the aluminum-coated fiber specimens in an argon atmosphere. The heat treatments were carried out for 90 ks at temperatures of the solid state of aluminum, or for 1.8 ks at temperatures of the liquid state of aluminum. Though the uncoated carbon fibers showed a marked loss in strength after the heat treatments, the SiC coated coarbon fibers showed no loss in strength after the heat treatments at 673 to 873 K for 90 ks and at 973 to 1073 K for 1.8 ks. However, the noticeable degradation in strength of the SiC-coated carbon fibers was observed after the heat treatments at temperatures higher than 1173 K for 1.8 ks. The wettability between carbon fiber and liquid aluminum was improved by the SiC coating.View full abstractDownload PDF (572K) -
Hideo Tai, Yukio Narita, Masahiko Nakamae, Shigenori Hori1980 Volume 44 Issue 3 Pages 276-282
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSEquilibrium Phase diagram for the CdTe-In system has been determined by means of differential thermal analysis, dissolution measurement, X-ray diffraction and microscopic examination. This system can be considered to be a pseudobinary system from results of the X-ray experiment. The phase diagram of this system is a simple eutectic type and the eutectic point is 429.5 K for the composition of 4×10−3 mol% CdTe. The solubility of CdTe in liquid In, x (mole fraction), is represented as
(This article is not displayable. Please see full text pdf.)
Based on the phase diagram, liquid phase epitaxial growth of CdTe was carried out using an In solution containing CdTe. CdTe single crystals of n-type were grown on p-type CdTe substrates and epitaxial diodes were fabricated from these crystals. These epitaxial layers and diodes were characterized with the aid of X-ray, microscopic, and electrical techniques such as Hall effect, I−V, C−V and spectral photoresponse measurements.View full abstractDownload PDF (755K) -
Shinichi Yoda, Reiki Takahashi, Kenji Wakashima, Sokichi Umekawa1980 Volume 44 Issue 3 Pages 283-288
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSA detailed metallographic examination of interphase boundary structures has been performed with reference to the familiar system, tungsten fiber reinforced copper, to clarify the origin of interfacial sliding-induced relaxation of internal stresses which are generated upon temperature cycling due to differential thermal expansion of the constituent phases.
Remarkable void formation at the fiber/matrix interface was observed in specimens subjected to hundreds of thermal cycles in the temperature range between 0.35 and 0.8 of matrix homologous temperature with a hold time of 600 s at the upper cycling temperature. Isolated voids elongated in the fiber axis are formed at an early stage of cycling and increase in number as the cycling proceeds until they coalesce with each other to form irregulary shaped macrocavities, which leads to a decrease in effective bonding area at the interface. To examine the decrease in effective load-transmitting-capacity of the interface due to the void formation, the effects of the voids on the tensile behavior have been studied by means of tensile tests and monitoring acoustic emission as well as fractography. The characteristic features of tensile behavior of thermally cycled composite specimens are summarized as; (1) decrease in elongation, (2) increase in aspect-ratio, and (3) increase of the stress drop at the serration in the stress-strain curve. The results are interpreted as being due to the interface bonding strength decreased with voids increasing, due to a reduction of effective bonding area. It is concluded that the interfacial sliding, causing the thermal ratcheting of the composites, is more easily induced by the interface void formation.View full abstractDownload PDF (911K) -
Yoichi Ishida, Hidenobu Ishida, Hideki Ichinose, Kazutaka Kohra1980 Volume 44 Issue 3 Pages 288-293
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSExact determination of the Burgers vector of the dislocation is performed by the weak-beam “fringe counting” technique. Both the magnitude and the sense of the Burgers vector are determined exactly from the number of the terminating thickness fringes at the exit of the dislocation from the wedge shaped foil. The number is equal to the inner product of the Burgers vector and the diffracting vector. Thickness fringes in the foil of Al-Mg and Fe-Mn alloys were examined by an 1 MV electron microscope for diffractions up to {211} in BCC and {220} in FCC crystals. A complete determination of the Burgers vector is achieved when three weak-beam micrographs with the diffraction vectors not sharing a plane are compared in the same area. When the possibility of the Burgers vector is limited within basic lattice vectors, the identification can be made with little tilting of the specimen. The small tilt angle and the narrowness of the dislocation image are helpful in the analysis of relatively dense dislocation structure.
It is concluded that the present technique may be utilized as the standard technique for the determination of the Burgers vector in metals and alloys, in place of the conventional ones.View full abstractDownload PDF (849K) -
Tsuyoshi Yamamoto, Toshiyuki Takashima, Keizo Nishida1980 Volume 44 Issue 3 Pages 294-299
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSFor the purpose of studying interdiffusion in a ζ-solid solution of the Ni-Al system, the experiments were performed using the method of vapor-solid couple at 1273∼1573 K. As a vapor source, the fine chips of 12 wt%Al-Ni alloy (ζ+ε) were used. The surface concentration of Al in the diffusion annealed test pieces almost coincided with the solubility limit reported in the phase diagram of the Ni-Al system at each experimental temperature.
Fine alumina markers placed on the surface of test pieces prior to diffusion were found inside the test pieces after annealing at each temperature. The composition at the marker position indicated a higher Al concentration with an increase of annealing temperature. The ratio of the intrinsic diffusion coefficients of these two constituents, DAl⁄DNi, was 1.3∼7.0 at the marker position.
The interdiffusion coefficients (\ ildeD) were dependent upon the annealing temperature and Al concentration, and were evaluated to be the orders of 10−15∼10−13 m2/s in this experimental temperature range. Log \ ildeD increased about linearly with increasing Al concentration.
The activation energies for interdiffusion (\ ildeQ) obtained from the temperature dependence of \ ildeD decreased from 282 to 258 kJ/mol with increasing Al concentration. The impurity diffusion coefficient of Al in Ni, DAl*, at each temperature was obtained from extrapolating of \ ildeD to 0 at%Al and its activation energy, QAl*, obtained from the Arrhenius plot of DAl* was evaluated to be 284 kJ/mol.
The relationship between the entropy term, \ ildeD0, and \ ildeQ can be represented by the following equation:
(This article is not displayable. Please see full text pdf.)View full abstractDownload PDF (637K) -
Takao Kozakai, Toru Miyazaki, Hirotaro Mori1980 Volume 44 Issue 3 Pages 300-305
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSA relationship between the morphology of microstructure and the linear expansion-coefficient per unit composition change η was investigated for three Ni-Cu-X spinodal decomposition alloys; Ni-36.1 at%Cu-9.8 at%Sn, Ni-37.0 at%Cu-7.4 at%Nb and Ni-38.0 at%Cu-4.9 at%Ti alloys.
The microstructures in these alloys aged in a temperature range of 673∼823 K are as follows. For the Ni-Cu-Sn and Ni-Cu-Nb alloys with relatively large η; −7.44×10−2 and −6.56×10−2, the microstructures consist of a periodic array of zones along the ⟨100⟩ directions and satellite reflections are observed in the electron diffraction pattern. On the other hand, for the Ni-Cu-Ti alloy with a smaller η, −1.26×10−2, the zones are aligned with weak periodicity along the ⟨100⟩ direction and no satellite is observed. Such a morphological distinction is interpreted in terms of the different values of η in the elastic energy term 2η2Y[hkl] of these alloys.View full abstractDownload PDF (886K) -
Masamichi Shibata, Shigeru Asano1980 Volume 44 Issue 3 Pages 306-311
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSInternal friction was measured as a function of temperature from 130 to 290 K at a frequency of about 500 Hz for hydrogen-charged Fe-Cr alloys containing up to 27% of chromium. In all the alloys, a characteristic internal friction peak always appeared after hydrogen charging of cold rolled specimens. This peak was identified as the Köster-type cold work peak based on the interaction of hydrogen with dislocations. The results obtained are as follows: (1) The peak was found at about 250 K for high chromium alloys (6–27%Cr), while the usual cold work peak was observed at 150∼180 K for pure iron and mild steel. (2) The peak height decreased with increasing the chromium content up to 6%, but it became constant for high chromium alloys (6–27%Cr). (3) The peak shift during room-temperature aging did not occur for high chromium alloys (6–27%Cr), in contrast to the cases of pure iron and low chromium alloys where the cold work peak is known to shift remarkably to lower temperatures. (4) The binding energy between hydrogen atoms and dislocations was estimated from the peak temperature to be about 0.3 eV in pure iron and about 0.5 eV in 27%Cr alloy. The binding energy increased almost linearly with increasing the chromium content. This means that the interaction of hydrogen with dislocations became stronger with the content of chromium in Fe-Cr alloys.View full abstractDownload PDF (496K) -
Yoshiro Ibaraki, Toshimi Sasaki, Hachiro Honma1980 Volume 44 Issue 3 Pages 311-316
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSThe micro-structure of as cast Ti-6Al-4V alloy is acicular alpha or the Widmanstätten structure, and it is believed that this structure cannot be changed to other structures by the conventional heat treatment. The cyclic heating has been developed to give a two-phased structure (alpha+beta) capable of being conventionally heat-treated.
Castings are subjected to two cycles of heating from 673 to 1213 K at a heating rate of 0.067 K/s and then conventionally heat treated.
Mechanical properties of the castings are determined and compared with those of forgings. The cyclic heating followed by the conventional heat treatment is applied to some of the forgings, in order to examine the effect of the cyclic heating.
The following conclusions are obtained.
(1) Tensile properties of the casting, particularly the elongation and reduction in area, are improved by the cyclic heating but are slightly lower than those of the forgings.
(2) Impact energy values of the castings are always higher than those of the forgings, when subjected to the same heat treatment.
(3) As cast and as cyclic heated castings maintain the impact energy values at ambient temperature even at 77 K.View full abstractDownload PDF (654K) -
Tatsuhiko Ejima, Tsutomu Yamamura, Naoto Uchida, Yutaka Matsuzaki, Mas ...1980 Volume 44 Issue 3 Pages 316-323
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSIn order to clarify the relative importance of various factors controlling the diffusion in liquid metals, the measurements of impurity diffusion coefficients have been carried out over the temperature range of 980 to 1320 K. Aluminum which can be regarded as a trivalent simple metal was chosen as a solvent. On the other hand, homovalent elements (Ga, In and Tl) and the fourth period elements (Fe, Co, Ni, Cu and Ga) have been chosen as the solutes. The results obtained can be summarized as follows:
(1) The diffusion coefficients of homovalent solutes are in the order of DGa>DIn>DTl, that is, the diffusion coefficient increases as radius and mass of solute element decrease.
(2) The apparent activation energy for diffusion of the solute belonging to the fourth period in the periodic table increases with decreasing valence of the solute. This may be due to the coulombic interaction between the solvent and the solute atoms caused by the excess charge locating on the solute atom. By assuming the Hartree self-consistent potential for the charge screening potential, the activation energies for diffusion were calculated. The calculated values agreed well with the observed apparent activation energies.
(3) The diffusion coefficients of solutes in molten aluminum calculated on the basis of the hard sphere theory originally derived by Enskog are slightly smaller than those observed. The apparent energy obtained from the temperature dependence of the calculated diffusion coefficients are also smaller than the measured ones. The difference between the calculated and observed apparent activation energies for diffusion can be explained in terms of the coulombic interaction between the solute and the solvent atoms.View full abstractDownload PDF (685K) -
Haruno Okochi, Katsuyuki Takahashi, Shunichi Suzuki, Emiko Sudo1980 Volume 44 Issue 3 Pages 324-330
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSSimultaneous determination of multielements (C, B, Mo, Ta, Co, Fe, Mn, Cr, Nb, Cu, Ti, Zr, and Al) in nickel-base superalloys (Ni: 68∼76%) was performed by emission spectral analysis. At first, samples which had various nickel contents (Ni: 68∼76%) were prepared by using JAERI R9, nickel and other metals (Fe, Co, or Cr). It was confirmed that in the internal standard method (Ni II 227.73 nm), analytical values of all the elements examined decreased with a decrease of the integration time (ca. 3.9∼4.6 s), that is, an increase of the nickel content. On the other hand, according to the fixed time integration method, elements except for C, Mo, and Cr were not interfered within the range of nickel contents examined. A series of nickel-base binary alloys (Al, Si, Ti, Cr, Mn, Fe, Co, Nb, Mo, and W series) were prepared by high frequency induction melting and the centrifugal casting method and formulae for correcting interferences with near spectral lines were obtained. Various synthetic samples were prepared and analysed by this method. The equations of calibration curves were derived from the data for standard samples (JAERI R1∼R6, NBS 1189, 1203∼1205, and B.S. 600B) by curve fitting with orthogonal polynomials using a computer. For the assessment of this method studied, the F-test was performed by comparison of variances of both analytical values of standard and synthetic samples. The surfaces of specimens were polished with a belt grinder using #80 of alumina or silicon carbide endless-paper. The preburn period and integration one were decided at 5 and 6 s respectively. A few standard samples which gave worse reproducibility in emission spectral analysis was investigated with an optical microscope and an electron probe X-ray microanalyser.View full abstractDownload PDF (888K) -
Haruno Okochi, Katsuyuki Takahashi, Shunichi Suzuki, Emiko Sudo1980 Volume 44 Issue 3 Pages 331-337
Published: 1980
Released on J-STAGE: April 04, 2008
JOURNAL FREE ACCESSEmission spectral analysis of high speed steels (N, S, C, Mo, Si, W, Ni, Co, Mn, Cr, V, and Cu) was investigated by the fixed time integration technique. In this method, the readings of intensity were not affected by the amount of iron, while according to internal standard method (Fe II 271.44 nm), the decrease of integration time (ca. 5.0∼4.1 s) and the resultant decrease of reading of the intensity were observed with increasing amounts of iron (Fe: 65∼85%). Spectral interferences with coexisting elements such as V, Mo, W, and Co which change over a wider composition range in the samples were examined. The samples which contained the same amounts of other elements with varying amounts of one of V, Mo, W, and Co were prepared. The formulae for correcting interferences with nearby spectral lines were obtained using those samples. It was confirmed that in the high speed steels containing larger amounts of carbon (ca. 1.3%) the calibration curves had to be made separately. JSS 606∼611, NBS SRM D840 and D841 were used in this study but the contribution of concentrations of V, Mo, W, and Co was not sufficiently distributed for making calibration curves. Then, it was examined that the samples for calibration were prepared by recasting with varying mixing ratios of two kinds of standard samples (for chemical analysis) in which the contents of these elements were largely different from each other by high frequency induction melting and the centrifugal casting method. The satisfactory results were obtained. The standard samples for instrumental analysis (JSS 606∼611) and samples recast from the same standard samples for chemical analysis by the above method were compared by the establised emission spectral analysis. The analytical results of N, C, Mo, Si, W, Ni, Co, Cr, V, and Cu were coinsident, but the analytical values of carbon of high content and manganese in the recast samples showed negative errors. The calibration curve of sulfur (S: 0.004∼0.009%) could not be prepared with the standard samples for instrumental analysis but it could be successfully obtained with the recast samples.View full abstractDownload PDF (669K)
- |<
- <
- 1
- >
- >|