In order to estimate molar volume for multicomponent silicate melts, expanded approximation rules was proposed in view of practical use, on the basis of the regular solution approximation rules of exess molar quantities for binary system melts. Necessary parameters for the above rules have been derived using the previous data in binaries and ternaries, etc.. Using these parameters, it was found that the expansivity was related to anion-cation attraction parameter, and the interaction parameter was related to the distance between each network modifier ion arranged in SiO4 tetragonal structure. From the collation with the data of multicomponent silicate melts, it was confirmed that the above expanded approximation rules were applicable to predict the molar volume of melts over the wide range of composition and temperature discussed.
In order to estimate surface tension for multicomponent silicate melts, expanded approximation rules was proposed in view of practical use, on the basis of the regular solution approximation rules of excess molar quantities for binary system melts. Necessary parameters for the above rules have been derived using the previous data in binaries and ternaries, etc.. Using these parameters, it was found that near the melting point, surface tensions of pure oxides and halides 0σi were related to (Ti)mp/0Vi2/3, where (Ti)mp and 0Vi are the melting point and molar volume of pure oxides and halides, respectively. From the collation with the data of multicomponent silicate melts, it was confirmed that the above expanded approximation rules were applicable to predict the surface tension of melts over the wide range of composition and temperature discussed.
Deionized water was injected through a single-hole nozzle at bottom center into a mercury bath in a cylindrical vessel. Water droplets were generated at the nozzle exit and they rose in the bath spreading in the radial direction. The holdup, frequency, mean rising velocity, and mean diameter of water droplets were measured with a two-contact electroresistivity probe and a laser void meter. The laser void meter can detect the arrival of water droplets and the rising velocity of them by analyzing the reflection of a laser beam at the interface between the two liquids. Both measured values of the water holdup α and water droplet frequency fw followed the Gaussian distribution. Consequently, empirical correlations for α and fw were derived in the momentum region near the nozzle exit and in the buoyancy region far from the nozzle exit by referring to those derived for bubbling jets in a cylindrical vessel. The mean rising velocity of water droplets uw was calculated from the mass balance of rising water, and the mean diameter was calculated from the relation dw=1.5α uw/fw. Empirical correlations for α, fw, uwand dw were almost the same as their individual correlations for the bubble characteristics. Therefore there is no essential difference between a liquid-gas jet and a liquid-liquid jet.
For verifying the feasibility of a new gas blowing method, using a perforated plate, the effects of the above gas blowing method on the floating removal of inclusions in a tundish was studied in actual casting experiments for hydrogen induced cracking resistance steel. The validity of a mathematical model, previously reported, which predicts the inclusion removal efficiency was confirmed by the results of actual casting experiments. It was found by the above model that in the state of uniform bubbling flow occurred in the low superficial flow rate of gas, the faster the superficial flow rate, the higher the efficiency became. It was also found that the efficiency was independent of inclusion size.
Suppression of wave motion of molten metal surface is one of the most important technology for controlling initial solidification in various casting process, e.g. continuous casting, twin roll casting, etc, where the suppression of wave motion can improve surface quality of cast metal. The wave motion is motivated by gravity force, so that the diminution of apparent gravity force by imposing vertical electromagnetic force must be successful for wave suppression. In this paper, at first, based on this principle, frequency changes of wave motion were measured in two imposition types of both stationally magnetic field and direct electric current. In case current was parallel to the wave vector, the frequency did not change. On the other hand, in case magnetic field was parallel to the wave vector, the frequency did change according to the dispersion relation modyfied by apparent gravity force. Next, from the standpoint of the process control in actual applications, another experiment was conducted where the imposition area of electric current is limited to around meniscus. As the results, it was found that this technique for wave suppression is still effective for deep molten metal pool.
The morphology of MnS formed by solidification in steels and its modification by alloying of Al, Si, C and Ti have been investigated by means of optical and scanning electron microscopy. The morphology of MnS formed after the primary crystallization of Fe phase can be classified into (i) droplet MnS formed by monotectic reaction, (ii) rod-like MnS formed by eutectic reaction and (iii) fish-born type MnS formed by irregular eutectic reaction. The morphology of primarily formed MnS in high sulfur concentration has been observed to be (iv) spherical, (v) dendritic and (vi) angular depending on the additional elements and melting atmosphere. The mechanism of MnS formation has been discussed on the basis of the phase diagram information. It has been shown that the spherical shape of the primary and secondary solidified MnS is formed through the metastable reactions. The eutectic, dendritic and angular MnS are formed by the stable reactions, where the dispersed particles like TiN and Al2O3 with high melting temperature act as nucleants of MnS crystals. The addition of C and Si promotes the stable reactions by enlarging the temperature difference between the eutectic and monotectic points.
In order to clarify the solidification behavior of semi-solid metals during the initial stage of solidification, a chill block dipping test and numerical analysis have been carried out. The heat transfer between metals and mold and the micro structure near the surface are studied for semi-solid metals of Sn-Pb alloy, Al-Cu alloy, 0.4%C steel and stainless steel. The results obtained in this work are as followes; 1) Thermal resistance between the chill block and the solidified shell of semi-solid metals are the same as that for molten metals. Thus the cooling rates evaluated from the secondary arm spacing are the same in both cases. 2) The shell growth rate of semi-solid metals is larger than that of molten metals because of the decrease of latent heat of fusion. This growth rate is numerically predicted by considering the fraction of solid and the corresponding released latent heat of fusion. 3) The increasing rate of the chill block temperature is small for the casting of semi-solid metals, because the total heat removed from the shell is decreased.
In the previous paper, the author has reported that the surface brightness is expressed as three functions of the inlet oil film thickness and the surface brightnesses of roll and sheet before rolling, and a system for estimating and controlling the surface brightness before rolling can be constructed by using this relation. In this paper, a new system for estimating the surface brightness after each pass in actual mill for cold rolling of stainless steel is proposed by developing this relation. In the new system, the estimation is carried out as follows: (1) Preparation of as hot rolled specimen, (2) Input of rolling conditions at each pass in actual mill, (3) Calculation of the inlet oil film thickness at each pass, (4) Determination of rolling conditions at each pass in laboratory mill, (5) Rolling in laboratory mill and measurement of surface properties of rolled specimen, (6) Surface brightness after final pass. By using the new estimating system, the surface properties after each pass are measured and it is seen that the system is able to estimate the surface brightness after each pass in actual mill.
This paper reports on the investigation we conducted into the mechanical cause of the pinch mark defect called heat buckling which can occur during the processing of thin steel strip in continuous annealing and processing lines. The major cause of heat buckling is the convex crown of hearth rolls. A revolving roll generates a force which compresses strip width. This in turn causes the strip to buckle and produce wrinkles. When wrinkles are crushed on the roll, heat buckling results. We obtained a formula which shows that the critical tension at which heat buckling starts to develop is proportional to the square of the ratio between (strip thickness·yield stress) and (roll crown·friction coefficient·yield stress). Transverse temperature deviation is a cause of strip buckling, ie, center wave or edge wave. Center wave reduces the critical tension of pinch mark defect.
We investigated on shear adhesion strength of galvannealed steel (GA), focusing on interface structure. Galvanizing was performed at the several dipping conditions which was varied in the range between 0.05 and 0.18 mass % aluminium in molten zinc bath. The interface structure was evaluated employing scanning electron microscopy (SEM) and electron channeling pattern (ECP). High aluminium added to molten zinc raised the minimum interface adhesion strength of the coating around 9 mass % iron. Moreover, ruggedness with pitch of α-Fe grain size was remarkably formed at coating/substrate interface, only when the substrate sheets were galvanized in high aluminium-containing zinc-bath. The ruggedness formation seems to be caused by the difference of Γphase growth rate between on (111)-(113)-(313) α and on the plane of  α zone. Γphase grew slower on (111)-(113)-(313) α compared to on the plane of  α zone. The dependence of Γphase growth behavior on iron orientation was promoted by aluminium addition to molten zinc bath. From above mentioned results, the model that GA coating/substrate interface adhesion strength was varied with geometrical shape of the interface, which was reported previously, can be expanded to explain not only effect of iron in coating but one of aluminium in molten zinc.
A computer model was developed to simulate the diffusion-controlled growth of planar boundaries during continuous cooling or heating using the Green's function method. Simulations of proeutectoid ferrite transformation from austenite and precipitation of cementite from both austenite and ferrite were conducted and the results were compared with calculation assuming the additivity rule which is based on the solution to the diffusion equation under isothermal condition. The additivity rule caused quite a large error in all the transformations studied. From theoretical considerations by other authors and present simulation results, it was concluded that the error occurred primarily due to the temperature variation of the (local equilibrium) solute concentration at interphase boundaries. In contrast to the usual expectation the amount of error was insensitive to the cooling rate and/or the magnitude of solute diffusivity.
The effect of Al contents ranging from 0.002% to 0.025% on secondary recrystallization behavior of ultra-low carbon 2.2%Si-1.5%Mn steels was investigated. The secondary recrystallization at relatively low annealing temperature of 1148K was achieved only in the steel with 0.007% sol.Al. The development of Goss texture was affected by annealing atmosphere, and the Goss texture as same as that of conventional grain oriented electrical steels was obtained by annealing in N2 containing atmosphere. The Goss texture of 0.007% sol. Al steel would be developed due to the optimum inhibitor effect of fine (Al, Si, Mn) nitrides whose crystal structure was similar to AlN.
Effect of microstructural changes on the crack generation and propagation during low cycle fatigue was investigated on hot working tool steel SKD6, having the hardness of 48HRC, at 873K in air. The stress amplitude vs. number of cycles relationship under constant strain condition consisted of three stages, as follows. The first stage is a short period of work hardening, and the second stage is dominated by cyclic softening due to dynamic recovery of matrix structure and coarsening of carbides. At the final stage, stress amplitude is drastically decreased by crack propagation, which could result from micro-cracks initiated at scales near surface.
The fracture toughness must be evaluated under the dynamic loading condition, when the materials are used for structures which are desired to a higher safety and must be taken into account of the effect of dynamic loading. Moreover, in order to obtain the valid JIC value, a criterion about both specimen thickness B and ligament width b0 must be satisfied, which is called as “valid condition” and is described in the following equation. B, b0≥α(Jin/σfs), (α=25) However, this equation is presented in the case of static loading and there is no assurance to be able to apply in the case of dynamic loading. Therefore, under static and dynamic loading conditions, the effects of both B and b0 on Jin value and the flow stress σfs are investigated and the valid conditions are discussed. The results are presented that the factor α of the above equation is reduced as α = 20 about b0 and is increased as α=28 about B under dynamic loading condition.