Possibilities for control of slag foam height by sound waves were investigated using cold models. The foaming and defoaming rates were measured under various frequencies and intensities of sound. Nitrogen gas was injected into a liquid containing a surfactant to produce the foam layer. Simultaneously, the foam surface was exposed to sound waves propagated through the gas phase to the surface. The sound waves were generated by a loudspeaker. Two liquids of different viscosities were used: water and a water-glycerin solution. It was found that sound waves can control the defoaming rate, while the foaming rate does not vary practically under the application of sound. The latter finding is associated with the fact that the height of foam, produced in the experiments, was much smaller than the steady height. The defoaming rate is essentially reduced by the sound of lower frequency (200-1000 Hz). This effect is explained in terms of a radiation force arising when the sound waves are propagated inside the foam layer. Sound waves of intermediate frequencies (1000-8000 Hz) cause the defoaming rate to increase compared with that in the absence of a sound field. The defoaming mechanism is assumed to be associated with a rupturing of the foam film accelerated by the sound waves. On the whole, the effect of sound on the defoaming rate increases as sound intensity and liquid viscosity become higher.
The decarburisation of liquid Fe-C-S drops was studied at 1723 K employing the electromagnetic levitation technique with two different types of gas mixtures, namely, mixtures of O2 and CO2, and mixtures of O2 and H2O. For the decarburisation with O2 and CO2, the rate is limited by the O2 and CO2 transport in the gas phase with additional resistance arising from the interfacial reaction kinetics of CO2. Sulphur significantly reduces the surface reaction rate with CO2, and the retarding effect appears to be enhanced by existence of the simultaneous oxygen decarburisation. For the simultaneous decarburisation by O2 and H2O, the basic mechanism of decarburisation is similar to that of O2 and CO2, but there exists evidence that the homogeneous gas phase flame reaction of O2 and CO to produce CO2 occurs in the vicinity of the interface. This results in a much lower rate of decarburisation than the expected one, since portion of the incoming O2 gas is converted to CO2 and the rate of decarburisation with CO2 is slower to a great extent than that with O2. The conversion rate is increased as the sulphur content in the melt is increased, and this is attributed to the increase in H2O in the vicinity of the interface with increase in sulphur. Mixed control models are developed and found to adequately explain the observed results.
The free surface motion of a liquid metal has been investigated by changing the wave patterns of a quasi-sinusoidal magnetic field and a modulated quasi-sinusoidal one. The theoretical expression of Lorentz force induced by the quasi-sinusoidal magnetic field applied on an electrically conductive cylinder has been derived. The Lorentz force consists of a time-averaged part Fave and oscillating parts Fω and F2ω with angular frequencies ω and 2ω, respectively. The surface disturbance of a liquid gallium was observed under the imposition of the magnetic fields. It was found that the oscillating part of Fω dominants the surface wave motion in the low frequency range and the substantial frequency of the surface wave is the same as that of a coil current. This result agrees with the theoretical derivation. While in the high frequency range, the induced recirculating flow causes an irregular fluctuation on a free surface. It is noticed that the disturbance on the free surface is intensified at the resonance frequency and the around of 1000 Hz.
Isothermal phase relations for the MnO-Al2O3-TiO2 system have been investigated under moderately reducing conditions, PO2=2.29×10-11 atm and 1.91×10-10 atm at 1673 K and 1873 K, respectively, equilibrated with a saturating oxide tablet. Activities of MnO in the MnO-Al2O3-TiO2 system have been measured in equilibrium with silver at 1673 K and copper at 1873 K, respectively. Iso-activity curves for MnO have been drawn from the experimental results of the activities as well as those for Al2O3 and TiO2 calculated by Schuhmann's tangent intercept procedure, which provide the deoxidation equilibrium of molten steels. The sulfide capacity significantly depends on MnO content in this ternary system. It slightly increases when Al2O3 replaces TiO2 at a constant MnO content in weight percent, whereas γMnS considerably decreases.
The influence of the raw material bed condition on pore formation in the sinter cake was investigated by on-line observation of the sintering process in the hot stage and quantification of the pore combination behavior. Using a new sintering test device which employs an X-ray computerized tomographic scanner, on-line observation of cross-sectional images of the sinter cake in the sintering process has become possible. When the bulk density of the raw material bed was low, the combination of +5 mm pores was remarkable and large pores formed with comparative uniformity in the sintering bed. In contrast, when the bulk density was high, the combination of +5 mm pores was slight, and combination of -5 mm pores with +5 mm pores was remarkable. Then, large pores formed along local +5 mm pores. It is considered that this difference in the morphology of pore formation effects the permeability and uniformity of burning in the sintering process.
In order to increase an interfacial area for the promotion of mass transfer in refining processes, a new method exciting surface wave on a molten metal has been proposed in the way introducing an intermittent alternating magnetic field with the stepwise change of amplitude. For estimating the mechanism of the wave excitation on a metal surface and the value of an excited interfacial area, the surface wave behavior on a liquid gallium was measured by use of a laser level sensor by changing intermittent frequency, fi, and duty, D. At lower intermittent frequency compared with the natural frequency of a liquid metal, the change in the magnetic amplitude excited surface motion on an equilibrium surface which could have been formed by the continuous imposition of each magnetic amplitude as a result of the balance between an electromagnetic pressure and a static liquid pressure. When the intermittent frequency is adjusted to the neighborhood of the natural frequency, descending of free surface in the wall vicinity takes place accompanying with increase in the magnetic amplitude and excites surface motion. When the intermittent frequency coincides with the natural one, a free surface area increases remarkably.
To see the effect of filling on the solidification process, an attempt is made to study numerically the fluid flow and heat transfer characteristics of pure metal during mold filling and subsequent solidification process simultaneously in a rectangular mold. To analyze the fluid flow and heat transfer with free surface motion, the MAC method is employed in the SIMPLER algorithm. Thermal stresses are also calculated based on temperature distributions through finite element analysis. Results show the distinct difference between present analysis and conventional analysis which just deals with solidification process only. Present analysis makes it possible to predict the defects occurring during filling and solidification processes, since any section of a mold at any time can be examined without difficulty, which is very difficult in an experimental approach.
The cooling curves at initial stage of solidification of carbon steel sample were measured using a new temperature measurement system that consisted of a two-dimensional optical pyrometer. Undercooling and recalescence phenomena were observed on the measured cooling curve of ultra low, low and middle carbon steel samples, and the solidification sequence at initial stage of solidification was shown to be different from normal solidification. A numerical analysis employing the dendrite tip growth model has been constructed, and fitted to a measured cooling curve by the parameter of interfacial heat transfer coefficient, and then the interfacial heat transfer coefficient between sample and chill plate which made of a transparent sapphire glass can be predicted accurately. The evenness of solidified shell is made clear to be influenced mainly by thermal deformation.
Directional solidification and laser experiments have been performed in Fe-18.0%Cr-Ni alloys to evaluate the effects of alloy composition and growth velocity on microstructure and phase selection. Solidification microstructure selection maps are described by microstructure (dendrite/cell, plane front and eutectic) models and compared with experimental results. Theoretical predictions are in good agreement with experimental results. It demonstrates the potential of this approach to analyze the microstructure and phase formation.
The present contribution focusses on the characterisation of the shear strength of galvannealed on Ti and TiNb interstitial-free steels currently used in the automotive industry. A modified lap shear testing method was developed which gave the shear strength of the substrate/coating interface independently of the mechanical charateristics of the substrate. The criteria for adhesive selection and the required sample geometry are reviewed in detail. An in-depth electron microscopic study of the fracture plane structure was also carried out. It revealed, among other things, a quantitative relation between the shear strength of the coating and the area of the steel/Γ phase fracture plane. It is shown that the shear strength of galvannealed coatings is influenced by (1) the Fe content of the galvannealed layer, (2) the type of steel substrate (Ti versus TiNb interstitial-free steel, drawing IF steels versus high strength IF steels) and (3) the effect of certain alloying elements.
The effect of the deoxidation and desulfurization products of Ce2O3 (the mean planar diameter, dA=1.6 to 2.1 μm) and CeS (dA=2.2 to 2.8 μm) particles on the austenite grain growth was studied in Fe-0.20 (0.09 and 0.53) mass%C-0.02 mass%P alloys. The melts at 1873 K were continuously cooled to 1673, 1573 and 1473 K and then were held for 0 to 180 min in the austenite single phase region. Grain growth in the presence of particles was strongly inhibited by the pinning effect and the mean grain size was independent of the carbon content for a given holding time, holding temperature, and volume fraction of Ce2O3 particles ( fV=0.08 to 0.12 %). The planar limiting grain diameter, DA, obtained at holding times of more than 60 min, was found to be significantly smaller than that predicted from the Zener relation, and the deviation from this relation increased with increasing the dA/fV value. The fraction of particles at the grain boundaries, ΦA=0.07 to 0.23, was about 20 to 60 times higher than that estimated from the random distribution from the Zener limit. Based on these values for ΦA, the limiting grain size was discussed as a function of fV value, and the present results for Ce2O3 and CeS particles were compared with those obtained for the MgO and ZrO2 particles in Fe-0.20 mass%C-0.02 mass%P and Fe-10 mass%Ni alloys.
The effects of the primary inclusion of Ce2O3 (the mean planar diameter, dA=1.6 to 2.1μm) or CeS (dA=2.2 to 2.8μm) particles on austenite grain size and grain shape distributions were studied in Fe-0.09 to 0.53 mass%C-0.02 mass%P alloys, which were continuously cooled from 1873 to 1673 K, followed by holding for 0 to 180 min. It was found that the shape of the relative spatial grain size distribution, DV/DV, mean (the mean grain diameter), was invariant with respect to the content of carbon, the number of Ce2O3 or CeS particles (the volume fraction of particle, fV=0.02 to 0.19%) and holding time. The spatial grain size distribution could be best described by the log-normal distribution function. The heterogeneity parameter, Dmax (the maximum grain diameter)/DV, mean ratio in three dimensions, increased with an increase in the fV value. The distribution of the number of grain sides and the grain size distribution for each grain side follow approximately the log-normal distribution function, and these results were consistent with those observed in previous experimental results for normal grain growth, in which a second-phase particle was not present.
The simulation method applied to determine no-recrystallization temperature (Tnr) has also been used to simultaneously determine phase transformation temperatures on cooling (Ar3, Ar1) in a set of six steels with different Ti and N contents. A study has been made of the influence of Ti and N contents on these parameters for different strain conditions, and the results permit an estimation of what would be the most appropriate contents depending on the type of rolling (hot strip mill, near net shape casting). Austenite grain size was also determined at 1100°C and 1300°C for all the steels, it being found that a Ti/N ratio close to two gave the finest grain.