日本金属学会誌
Online ISSN : 1880-6880
Print ISSN : 0021-4876
ISSN-L : 0021-4876
32 巻, 9 号
選択された号の論文の22件中1~22を表示しています
  • 日野 光久, 江島 辰彦, 亀田 満雄
    1968 年 32 巻 9 号 p. 809-814
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The surface tension, density and viscosity of PbO-Na2O-SiO2 melt have been measured over the temperature range 700° to 1200°C by the dipping cylinder method, the Archimedean method and the counter-balanced-sphere method, respectively.
    The surface tension decreases with the addition of Na2O to (PbO·SiO2), while it increases with increasing Na2O in the Na2O-SiO2 binary system.
    The negative deviation from the additivity in molar volume indicates the bond breaking of SiO2 and production of discrete anions. Enlargement in the expansion of melt, (∂V⁄∂T)P, with increasing Na2O indicates that the melt becomes more ionic with the addition of Na2O.
    Logarithm of viscosity, logη, shows a linear relation with 1⁄T2 rather than 1⁄T. The apparent activation energy of viscous flow decreases with increasing Na2O for the (PbO·SiO2)-Na2O system. For the (PbO·SiO2)-(Na2O·SiO2) system, the curve of activation energy vs composition shows a minimum at about 10 mol%Na2O.
  • 平野 賢一, 一法師 康武
    1968 年 32 巻 9 号 p. 815-821
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Interdiffusion in Fe-Ti alloys was studied in the temperature range between 700° and 1300°C, using diffusion couples of pure metals or pure metal and alloys. The Balluffi method was used to determine the interdiffusion coefficient \ ildeD in both the Fe-rich and Ti-rich solid solutions, taking into account the variations of the atomic volume with composition.
    From the Arrhenius relationship, the activation energy \ ildeQ and the frequency factor \ ildeD0 for the interdiffusion were calculated. The results showed that \ ildeQ was considerably low in the βTi phase, especially at the Ti-rich side, in agreement with the case of tracer diffusion.
    Measurements of the Kirkendall effect suggested that diffusion in this alloy was controlled by the vacancy mechanism.
    Formation of the intermetallic phases was not found in the interdiffusion couple of pure metals and the possible reasons for this were discussed.
  • 高橋 忠義, 鈴木 敏夫, 市川 洌
    1968 年 32 巻 9 号 p. 821-825
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Undercooling phenomena were studied in the bulk specimens of an Al-4 wt%Cu alloy cooled by spraying water on a graphite crucible.
    (1) The solute concentration of primary crystal solidified by undercooling was nearly consistent with that of the solidus line in equilibrium phase diagram when undercooled below 15°C. However, when the degree of undercooling increased over 15°C, the solute concentration was increased to a saturated value of approximately 2 wt%Cu deviated from that of the equilibrium solidus line. Such a deviation of solute concentration from the solidus line was expected to be caused by pricipitation of the θ phase.
    (2) The dendritic structure of the specimen solidified by undercooling became finer with increasing degree of undecooling. When the primary crystal was undercooled below the equilibrium eutectic temperature, the microstructure obtained was not dendritic. The structure showed clear grain boundaries and spherical θ phases which were homogeneously distributed along the grain boundaries as well as within grains.
    (3) It was suggested from the changes of nonequilibrium liquidus and eutectic temperatures with increasing cooling rate that the maximum degree of undercooling in Al-4 wt%Cu alloy could be about 198°C.
    (4) When melting and solidification were alternately repeated in the same specimen, the degree of undercooling was increased stepwise with increasing number of cycles of melting and solidification. Also it was found that the oxidation of the specimen surface before melting increased the degree of undercooling.
  • 高橋 忠義, 工藤 昌行, 萩原 巌
    1968 年 32 巻 9 号 p. 826-831
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    It has been observed in an Al-Cu alloy that the partial remelting of dendrites occurs under normal conditions of crystal growth.
    In order to study the relation between the morphology of crystal and the condition under which the crystals were formed during solidification, the upper portion of a molten metal was stirred along the mould wall during cooling from just above liquidous temperature to an appropriate temperature in the range of solidification.
    As a result, it was concluded that the dendrites crystallized during solidification could easily be broken into pieces by the resistance of flow fluid of the residual molten metal, because the very fine and sphere-like crystals were observed after solidification.
    Furthermore, in order to obtain an ingot with the fine crystals described above, the upper portion of the molten metal was continuously stirred during solidification of Al-Cu alloy or carbon steel.
    The results obtained were as follows:
    The fine crystals were continuously accumulated upward from the bottom of the ingot, and the upper portion retained a molten state till final solidification. The ingot obtained showed a flat top surface and a homogeneous distribution of solute concentration.
  • 和田 春枝, 郡司 好喜, 和田 次康
    1968 年 32 巻 9 号 p. 831-836
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The solubility of nitrogen in liquid iron has been measured over the temperature range from 1800° to 2100°C using irons with various oxygen contents. The measurement was made by a levitating melting method. The solubility increases with increasing oxygen concentration as well as with increasing temperature. The dependence of solubility on the temperature is consistent with the results obtained by Pehlke et al. in the lower temperature range, if the partial molar heat capacity of nitrogen in liquid iron is regarded as about the same as that of iron, i.e. about 11 cal/mol deg. The entropy of solution of nitrogen in liquid iron is compared with those in γ- and α(δ)- iron.
    The interaction coefficient of nitrogen with oxygen, eN(0), was obtained as −0.14 on the average in that temperature range, which is close to the value obtained by Fuwa et al. in the lower temperature range.
  • 依田 連平, 新井 隆
    1968 年 32 巻 9 号 p. 836-841
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Molybdenum, coated with nickel-base, iron-base and cobalt-base heat-resistant alloys by casting, was hot-worked for its high temperature uses under the stressed condition in air, and the results of the creep-rupture test were compared. When S-816 cobalt-base heat-resistant alloy was used as coating materials, the bonding by fusion between molybdenum and S-816 is considerably inferior to the others. It was known that one of the reasons is that the coefficient of melting rates of molybdenum into molten S-816 alloy is especially small. Molybdenum coated with any of the heat-resistant alloys can be best swaged at 1150°C, and cracking on the molybdenum part and the diffusion zone did rarely occur. In the creep-rupture strength at 1000°C in air, molybdenum coated with S-816 was strongest for the large stress and short time, while molybdenum coated with Nimonic 90 nickel-base alloy was strongest for the small stress and long time.
  • 杉本 克久, 沢田 可信, 森岡 進
    1968 年 32 巻 9 号 p. 842-848
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Behaviours of pitting dissolution of aluminium in various halide solutions have been investigated by anodic dissolution at a definite potential which is more noble than the pitting potential in the solutions. Effects of the matrix purity of aluminium and environment factors on the anodic pitting dissolution have also been discussed in the current investigation. The results obtained are summarised as follows:
    (1) The rate of pitting dissolution degreases with the increasing matrix purity of aluminium and increases with increasing concentration of halogen ions in the solution ranging from 0.1 to 5 N. The effect of halogen species on the rate of pitting dissolution is not different from each other.
    (2) The number of pits is changed by the matrix purity and depends on the concentration and pH of NaCl solution. The maximum pit number is observed at 5 N NaCl and the minimum at pH=10.6.
    (3) The growth rate of pit radius is larger than that of pit depth.
  • 片桐 敏夫, 藤井 京子, 高本 信次郎
    1968 年 32 巻 9 号 p. 848-851
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Effect of diffusion annealing on the boride layers of iron and steel was investigated. From microscopic examination and measurements of microhardness, the following results were obtained.
    (1) In argon atmosphere the Fe2B layer grew up with the disappearance of the FeB layer by diffusion annealing, and the depth or the microstructure of the Fe2B layer was not varied after the FeB layer had been disappeared.
    (2) The single layer of Fe2B formed on S 55 C steel was not cracked by water quenching from 900°C.
    (3) It was observed that the Fe2B layer grew up easily along the grain-boundaries rather than through the lattice.
    (4) By diffusion annealing, the gradient of microhardness in both the boride layer and the matrix disappeared.
    (5) The behavior of the boride layer subjected to the diffusion annealing in vacuum was much the same as in argon atmosphere, but continued annealing of the single boride layer of Fe2B caused the appearance of the deboronized layer.
  • 増本 量, 中村 直司
    1968 年 32 巻 9 号 p. 852-855
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The present investigators already measured the strain gauge factor K and electrical properties on the wires of Fe-Cr alloys containing less than 50% chromium, and obtained the largest value of 6.2 in K for the alloy containing 20% chromium. So they have further investigated the influence of the addition of less than 70% cobalt or 65% nickel on their properties. Although K of Fe-Cr alloys shows some decrease with increasing cobalt or nickel content, the electrical resistivity ρ increases considerably, while the mean temperature coefficient of electrical resistivity Cf and the mean thermo-electromotive force Emf relative to copper decrease and reach comparatively small values. The results for Fe-15%Cr-35%Co and Fe-10%Cr-20%Ni alloys show that K is 3.9 and 3.7, ρ is 88.6 and 86.0 μΩ-cm, Cf is 2.6×10−4 and 3.0×10−4, and Emf is −0.1 and −4.7 μV/°C, respectively.
  • 増本 量, 村上 雄悦, 比内 正勝
    1968 年 32 巻 9 号 p. 856-860
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Kobayashi and two of the present authors discovered in the previous work high permeability alloys “Nimalloy” in the nickel and manganese system. Much subsequent work has been carried out on the effects of additions of various elements on characteristic properties of nickel-manganese alloys, through which the highest initial permeability of 76000 and the highest maximum permeability of 441000 have been obtained. The results obtained by antimony additions to nickel-manganese alloys show that the initial and maximum permeabilities are in general increased with increasing antimony content; the alloy containing 76.43%Ni, 16.27%Mn and 7.30%Sb exhibits the highest initial permeability of 10870 when cooled at a rate of 2800°C/hr from 900°C followed by reheating 420°C for 4.5 hr, while the alloy containing 76.17%Ni, 18.53%Mn and 5.30%Sb shows the maximum permeability of 80700 when cooled at a rate of 240°C/hr from 900°C and reheated at 460°C for 2 hr. The latter alloy shows a magnetic hysteresis loss of 45.46 erg/cm3/cycle and a coercive force of 0.0219 Oe, respectively, for the maximum magnetic induction of 5000 G, a magnetic induction of 5390 G at a field of 50 Oe and an electrical resistivity of 57.1 μΩ-cm at 20°C.
  • 松本 昇, 三沢 輝起, 久保 真一, 音谷 登平
    1968 年 32 巻 9 号 p. 860-866
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    When Ni-C melt containing a small amount of sulphur was unidirectionally solidified, layers with spheroidal graphite and with lamellar graphite appeared alternately, which were both parallel to the solid-liquid interface. The condition for the formation of this periodic structure was investigated.
    Ni-C alloys 150 to 300 g in weight were melted in a graphite crucible, and surphur of 0 to 300 ppm was added to melts. These melts were cast into the graphite tube mould whose bottom end was closed by a water-cooled copper chill plate and then solidified unidirectionally. Thermocouples were inserted into the mould for thermal analysis. The results were as follows; the ingot of 0 ppm S became a wholly spheroidal graphite zone, and that of 300 ppm S, an almost lamellar graphite zone except for the region near the chill plate. When S was 50∼150 ppm, the ingot had a periodic structure and its rate of solidification was 0.03∼0.05 cm/sec. The thicknesses of the lamellar and spheroidal layers in that structure were 1∼3 mm and 10∼2 mm, respectively, and the maximum number of the periods reached more than 10 in the 80 mm length ingot. The interface from the spheroidal to the lamellar layer was rather smooth, but the reverse was considerably rough and jaggy. There were many dendrites parallel to the growth direction in lamellar zones.
    The mechanism of formation of this periodic stcture could not be explained by the ordinary concept of solidification.
  • 菊池 武昭, 黒沢 利夫, 柳橋 哲夫
    1968 年 32 巻 9 号 p. 866-872
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The reaction between silica and carbon was investigated in argon and carbon monoxide atmospheres as a fundamental research of the direct reduction of aluminum, because silica is involved in the source materials. When the pellet consisting of SiO2/C=1/2 was heated in argon, SiO gas was evolved above 1300°C and only silicon carbide was formed by the following reactions, SiO2 (s)+2C (s)=SiO (g)+CO (g) and SiO (g)+2C (s)=SiC (s)+CO (g). The above reaction occurred at higher temperatures in the CO atmosphere.
    Metallic silicon was found by the reaction between silicon carbide and SiO gas evolved from the SiO2-C mixture at 1900°C, and this reaction was assumed to be SiC (s)+SiO (g)=2Si (l)+CO (g).
    Metallic silicon was also formed when the pellet consisting of a SiO2-SiC mixture was heated owing to the reaction of SiO2 (l)+2SiC (s)=3Si (l)+2CO (g), and so the above two processes were estimated to be the main mechanism for the production of silicon.
    The fractions of SiC, Si, SiO and SiO2 formed in the cases of SiO2-C and SiO2-SiC mixtures were determined up to 2000°C, and their reaction processes were considered thermodynamically.
  • 小田 仲彬, 久保 正二, 山岸 忠彦
    1968 年 32 巻 9 号 p. 873-876
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    On the basis of a generalized method for the determination of C, H, N, S and X presented in the preceding reports (Japan Analyst, 11 (1962), 406, 411), the generalized method for fluorine has been studied, and satisfactory results were obtained except for fluoroborates.
    For the determination of fluorine, the sample (20∼30 mg as F) mixed with 500 mg of WO3 and 500 mg of SiO2 was heated at 800°∼1300°C in the current of oxygen, and the resulting SiF4 was absorbed in NaOH solution. The absorbed solution was titrated by using 0.1 N Th(NO3)4 and the Alizarin S indicator.
    The error in the proposed method for fluorine was within ±0.5%, and the time required for analysis was 80 minutes per sample.
  • 斎藤 利生, 村上 昇一
    1968 年 32 巻 9 号 p. 876-881
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    In order to shorten the annealing time for three kinds of hardenable low alloy structural steels and tool steels, the isothermal anneal treatments that comprise each proper working in various annealing processes were stadied.
    Isothermally annealed steels were affected clearly by working, when deformed at the holding temperature or the austenitizing temperature: the steels deformed at the holding temperature were remarkably softened. When the steels were held at an isothermally heating temperature after working at that temperature, the steels in which the transformation begins with a considerable delay were remarkably softened. On the other hand, the steels whose transformation takes place very fast at the holding temperature were somewhat strengthened by working at that temperature, because the steels containing large quantities of already transformed preducts were deformed.
    Both the steels, deformed at the austenitizing temperature and the metastable austenitic temperature, followed by isothermal annealing, showed a uniform and fine spheroidal pearlitic structure.
  • 西田 義則, 鈴木 鎮夫
    1968 年 32 巻 9 号 p. 882-886
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The condition at mold-ingot interface has a great influence on the cooling of ingot. But the influence appears not to be fully analyzed.
    In this paper, it is shown that the cooling of ingot before solidification can be analyzed by considering the heat resistance R at mold-ingot interface and approximate solutions are taken by using the heat balance integral.
    The results are as follows,
    (1) In the case of semi-infinite mold and ingot, eqs. (23), (24), (25), (26) are obtained for a constant R, eqs. (28), (29), (30) for linear variation of R with surface temperature of ingot, and eqs. (32), (33), (34) for second-degree polynominal variation of R with the surface temperature.
    (2) When the thickness of an ingot with constant R is 2a cm and mold is semi-infinite, eqs. (35), (36), (37), (38) can obtained.
  • 土井 俊雄, 石田 文彦, 川辺 潮, 北田 正弘
    1968 年 32 巻 9 号 p. 886-892
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The superconducting properties were examined on Nb-Zr-Ti alloys with an extensive range of composition, and the effect of precipitation on the critical current density Jc of Nb-40 at%Zr-10 at%Ti alloy in magnetic field was investigated in relation with the microstructural changes due to the transformation of the β to the β′ and β″ phases at 700°C. The main results are as follows: (1) The composition range with high values of Jc was found to expand over the line binding the compositions Nb-47Zr and Nb-67Ti in Nb-Zr-Ti alloys. Above all, Nb-40Zr-10Ti alloy has the highest Jc below 80 kOe. (2) For Nb-40Zr-10Ti alloy aged intermediately at 700°C for 3 hr, the critical field was 105 kOe at 4.2°K; the value of Jc was 3.6×105 A/cm2 at 40 kOe and 5.0×104 A/cm2 at 80 kOe; and the critical temperature was 10.3°K. (3) It may be due to the flux-pinning of the β′ phase, the weaker superconductor in the β′ and β″ phases, that the Jc value of Nb-40Zr-10Ti alloy increases by aging at 700°C. (4) The Jc value reaches a maximum value as the size of the β′ phase and the spacing between them approach the order of 10∼102 Å.
  • 石田 文彦, 土井 俊雄, 梅沢 正
    1968 年 32 巻 9 号 p. 893-897
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The critical current density Jc vs. applied magnetic field H characteristics were examined on the Nb-Ti side alloys in the Nb-Zr-Ti system, and the effect of precipitation on the Jc vs. H characteristics was investigated in relation to the microstructural changes due to the transformation of the β to the α and β phases at 400°∼600°C in Nb-5 at%Zr-60 at%Ti alloy. The main results are as follows: (1) The composition with the high Jc at 70∼90 kOe was the Nb-5Zr-60Ti alloy. For the Nb-5Zr-60Ti alloy aged at 400°C for 100 hr, the value of Jc was 6.4×104 A/cm2 at 70 kOe and 4.4×104 A/cm2 at 80 kOe. (2) It may be due to the flux-pinning of the α phase, the normal conductor at 4.2°K in the α and β phases, that the Jc value of Nb-5Zr-60Ti alloy increases by aging at 500°C. (3) Flux-pinning by α precipitates was most effective in the early stages of precipitation and the growth of precipitates resulted in reduced values of Jc. (4) The peak effect of Jc may be due to the interaction between flux lines at high magnetic fild near critical field HC2 in the structure with random distribution of pinning strength.
  • 三浦 維四, 浜中 人士
    1968 年 32 巻 9 号 p. 898-902
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    The authors have studied the method of preparation of Al-Si eutectic fibrous composites by unidirectional solidification, of which their microstructures are already reported. In this paper, the mechanical properties of unidirectionally solidified Al-Si eutectic composites were compared with those of as-cast Al-Si eutectics over a range of temperature, and the strengthening mechanism of Al-Si eutectic fibrous composites were discussed.
    The main results are as follows:
    (1) Unidirectionally solidified eutectic composites exhibited a mechanical reinforcing behaviour. Compared with the as-cast structures, the proportional limit showed a considerable increase, but the ultimate tensile strength was hardly changed.
    (2) The strengthening mechanism of Al-Si eutectic fibrous composites was well explained by the theory proposed by Kelly and Davie. The observed and calculated values of strength and critical aspect ratio agreed fairly well with each other.
    (3) Fracture of the unidirectionally solidified specimen occurred after silicon colonies had broken to a critical length in the tensile direction.
  • 椙山 正孝, 福迫 達一, 佐藤 彰
    1968 年 32 巻 9 号 p. 902-907
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Regarding the process of direct-rolling of molten metals for the manufacture of finished products from molten metals as a developed process of continuous casting, a study was made to produce half-made products by the direct-rolling of molten metals. Using our laboratory machine (with Bessemer’s twin-roll system), the direct-rolling of molten metals was carried out with pure Al which has high plasticity and solidifies in a manner of the skin formation type. Surface conditions and structures of direct-rolled plates were examined. The main results obtained were as follows:
    (1) Surface conditions of the direct-rolled plates were known to be improved by employing the lower pouring temperature, slower direct-rolling speed and secondary water spray cooling, and applying no front-tension.
    (2) The formation of ripple marks on the surface of the direct-rolled plates, which was promoted by the front-tention, was considered as due to a mechanism of adhering and yielding of the metal upon solidification.
    (3) From observations of the macro-and microstructures it was concluded that at a slower direct-rolling speed the metal as solidified was rolled. And because of the high rolling temperature, the polygonization on the hot-rolled Al was observed.
    (4) Sumps were observed by means of pouring molten lead and it was found that the shape of sumps was very complicated at the lower pouring temperature and very simple at the higher pouring temperature.
  • 椙山 正孝, 福迫 達一, 佐藤 彰
    1968 年 32 巻 9 号 p. 908-913
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Regarding the process of direct-rolling of molten metals for directly manufacturing finished products from molten metals as the developed one of the continuous casting, production of half-made products by the process of direct-rolling of molten metals was studied. Using our laboratory machine No.2 (with Bessemer’s twin-roll system), the direct-rolling of molten metals was carried out with Al-Cu alloys which have a wide solidification range and solidify in a pasty manner. The rolling conditions and properties of the direct-rolled plates were examined.
    The main results obtained were as follows:
    (1) The copper content and the solidification range of Al-Cu alloys have little effect on the upper limit of direct-rollable speed related with the pouring temperature of molten Al-Cu alloys, but its upper limit is lower than that of the pure Al.
    (2) The slower the direct-rolling speed, the more strongly segregate copper atoms in the middle of the thickness of plates of Al-5.0%Cu alloys where solidification takes place at last.
    (3) Al-Cu alloys are inferior in surface conditions to the pure Al.
    (4) The low copper alloys (1.0%Cu, 2.0%Cu) have better mechanical properties at the slower direct-rolling speed, but for the highest Cu alloys (5.0%Cu) it is reverse.
    (5) Dendrite subgrain boundaries tend to disappear more completely at a slower direct-rolling speed after the heat treatment (515°C 5 hr WQ, 150°C 5 hr AQ) of the direct-rolled plates.
  • 榊 隆
    1968 年 32 巻 9 号 p. 913-918
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    An experiment was carried out to establish a simple and accurate method for the determination of microamounts of niobium in iron and steel.
    The niobium was coprecipitated as pentaoxide with tungstic acid by the hydrolysis method with sodium sulfite and tannic acid in hydrochloric and perchloric acids to which 5 mL of sodium tungstate solution (1 mL=1 mgW) was added as coprecipitating reagent and was separated from other elements. By means of this separation method, the microamounts of niobium can be separated accurately from iron and others.
    The separated niobium was determined by the spectrophotometric method with xylenol orange.
    In this spectrophotometric method, there is no significant effect even in case 0.1 mg of alminum, 1 mg each of antimony, bismuth, iron, manganese, tin and zirconium, 0.02 mg of chromium, 0.5 mg each of cobalt, copper, nickel, phosphorus and tantalum, 20 mg of molybdenum, 5 mg of titanium, 0.2 mg of vanadium and 50 mg of tungsten are contained in the separated sample solution.
    As a result of the experiment, the author succeeded in establishing a method in which 0.005 to 0.05% of niobium in iron and steel can precisely be measured without difficulty.
    The niobium contents in synthetic and actual samples were measured by this method with satisfactory results.
  • 賀集 誠一郎, 長瀬 道夫, 林 主税, 中島 篤之助, 高島 教一郎
    1968 年 32 巻 9 号 p. 919-925
    発行日: 1968年
    公開日: 2008/04/04
    ジャーナル フリー
    Preparation of Zircaloy standard samples for spectrochemical analysis by using a 120 kW electron beam skull furnace is described. Major elements (Sn, Fe, Ni and Cr) and impurity elements, such as Al, Cu, Ti, Si, Co, B and Hf, were observed before and after the skull casting.
    Results of the experiment are summarized in the following:
    (1) Three major elements (Sn, Fe and Ni) in Zircaloy can be controlled to within ±10% of the respective predicted preset level, provided each content in the skull being previously controlled to within ±50% of a predetermined value. Chromium showed a low and scattered yield.
    (2) Composition analysis of major elements in the upper, middle and lower parts of an ingot showed less than 4% deviation in Sn, 8% in Fe and Ni and 11.1% in Cr from the average contents in the ingot.
    (3) Impurity elements of B, Ti, Si, and Al in Zircaloy can be controlled to within ±15% of the preset level. Yields of Cu and Co are less than 20% and widely scattered results are obtained, even if the 120 kW rotated electron beam is used.
    (4) Chemical analysis of impurity elements in the upper, middle and lower parts of the ingot showed less than 5.2% deviation in Ti and less than 10% in B, Si, Co, Cu and Al from the average contents of the respective elements.
    (5) In skull melting of the alloy, there is an active zone near the surface of the skull which reacts with the molten pool. For the control of the Al or Ti content in Zircaloy the surface layer of the skull up to 5 mm in depth is estimated to be reactive.
feedback
Top