Phase decompositions in several Fe-Al (or Si)-X (X=V, Co, Ge, Ni) ordering alloy systems were investigated mainly by means of transmission electron microscopy and the phase diagrams were thereby constructed. Many types of phase diagram were obtained. In Fe-Si-V system a wide composition range of the phase separation into disordered A2 and ordered D0₃ phases is observed at temperatures lower than Curie temperature, while at the higher temperatures the A2+D0₃ two-phase field becomes narrow. In Fe-Si-Co alloy, the phase separation into B2 and D0₃ occurs in two separate composition regions; one is the region connecting with the B2+D0₃ region in the Fe-Si binary system, and the other exists separately near the Fe-14.3 at%Si-28.3 at%Co alloy. In Fe-Al-V system the phase separation of A2+D0₃ occurs in a large region of Fe-rich corner. For Fe-Al-Ge and Fe-Si-Ge systems the experimental phase diagrams are also proposed.
The free energies of A2 phase and B2 and D0₃ ordered phases in the ternary systems are evaluated by a statistical approach employing the Bragg-Williams-Gorsky approximation: the magnetic interactions are taken into account up to the second nearest neighbour spins of atoms as well as the usual atomic interactions up to the second nearest neighbours. The equilibrium phase diagrams and the Curie-temperature theoretically given by the present calcuation are almost consistent with the diagrams experimentally obtained. The influences of the magnetic energy on the structural stability of ordered phases are also demonstrated. The theoretical description proposed here is useful for the thermodynamic implication of the equilibrium phase diagram.

View full abstract

A two phase, combined Lagrangian-Eulerian approach has been adopted to model submerged gas injection phenomena during ladle refining operations. In this, the motion of the gas phase has been computed numerically in a Lagrangian field in the form of trajectories of a steady stream of bubbles, while the liquid flow fields together with the associated level of fluid turbulence were calculated in a routine manner via the Eulerian scheme. Furthermore, to solve the Eulerian and Lagrangian field equations numerically, a sequential, iterative computational procedure, incorporating the SIMPLE procedure was applied.
The internal consistency of the two phase calculation procedures, the scope of convergence, as well as the influence of grid distribution, on predicted results have been evaluated computationally. Subsequently, model predictions on velocity and turbulence fields, particulaly within the two phase plume region of aqueous gas bubble driven systems, have been compared against corresponding experimental measurements reported in the literature. It is shown that while flow and turbulence parameters in the bulk liquid phase are predicted with reasonable accuracy, some uncertainties remain in the prediction of bubble rise velocity, turbulence kinetic energy, etc. in the two phase region, particularly in the vicinity of the gas injection nozzle and the free surface.

View full abstract

Hot and cold experiments were carried out as a part of fundamental study for the development of a smelting reduction process. Three kinds of ore with different particle sizes and were used as the circulation materials. The circulation rate of ore was controlled to keep a constant solid holdup in the cold experiments. On the other hand, for the case of hot experiments the circulation rate of ore was not controlled, and the solid holdup was kept constant by adjusting the whole inventory in the apparatus. The results obtained are as follows:
(1) The circulation rate of iron ore increased as the particle size decreased and the average solid holdup in the riser increased. The circulation rate also increased as the gas velocity in the riser increased.
(2) The gas velocity and solid holdup were considered to affect the carrying over of particles as driving force and probability, respectively. The ore circulation rate, Gs, was proportional to the product of Fr number and the average solid holdup, (1-ε), in the riser.
(3) The rate controlling step in the reduction of very fine ore was considered to be chemical reaction. On the other hand, the reduction of larger ore was affected by both chemical reaction and diffusion in the particles.

View full abstract

Dephosphorization tests of chromium containing molten iron by CaO-based flux have been carried out by an induction furnace. The following results were obtained:
(1) When the carbon content in metal was high enough, chromium containing molten iron was able to be dephosphorized by CaO-CaF₂-FeO flux. In this case, oxygen partial pressure at the slag and metal inferface was calculated to be about 10^-12 Pa.
(2) The higher ratio of CaF₂/(CaO+CaF₂) and of CaO/O in flux were effective for the dephosphorization without the oxidization of chromium.
Based on these results, 200 ton scale industrial tests have been carried out by the hot metal pretreatment station at Yawata Works. By the combination of the flux injection method and the oxygen top blowing method, it was clarified that the dephosphorization without chromium oxidation and the suppression of temperature drop were able to be achieved simultaneously.

View full abstract

Removal of FeO from CaO-SiO₂-Al₂O₃-FeO slags by low carbon-iron melts has been investigated experimentally at 1723 K. The rates of FeO removal were estimated using experimentally measured rate of pressure rise in the reactor due to evolution of CO (g). Several flash X-ray images were obtained during the course of these experiments. The experiments were conducted by dropping a small amount (10-90g) of slag over an Fe-C melt weighing 400g. Based on the rates of reduction of FeO dissolved in CaO-SiO₂-Al₂O₃ slags by two types of Fe-C (1.3 and 2.2%) melts, it was determined that the reactions were independent of carbon concentration and followed first order rate kinetics with respect to FeO concentration. The reactions did not stop until metal and slag phases reached thermodynamic equilibrium with respect to the iron oxide in the slag. The calculated value for the rate constant was 7.34×10^-4m/s and it was independent of slag weight or foam volume. Therefore, it was concluded that the presence of foam on the matal bath did not influence the mechanism of FeO removal from slag by Fe-C melts.
Maximum height, h_f, of the foam generated was established to be a linear function of slag weight, w_s, present on the Fe-C melt surface and the relationship was given by
h_f=11.69+1.993w_s

View full abstract

Effect of cooling rate on oxide precipitation during solidification was investigated using Ti Al-Ca deoxidized low-carbon steels. The number and size distribution of oxides in the continuously cast steels have been observed and theoretically analyzed. The results obtained are as follows.
The Ti deoxidized steel has more soluble oxygen before solidification than the Al-Ca deoxidized steel. With the increase in the cooling rate during solidification, the number of oxides increases and the size becomes smaller in the Ti deoxidized steel.
During the solidification, segregation makes oxygen content increase above the equilibrium value with oxide in Ti deoxidized steel. The difference between the oxygen content increased by segregation and that in equilibrium is a driving force for the oxide growth. With increasing cooling rate, time for oxide growth decreases and supersaturation of oxygen for oxide precipitation increases. Consequently, the number of the oxide increases by the increase in the frequency of precipitates during solidification for high cooling rate.

View full abstract

In an attempt to overcome the surface defect originated from the intersection of melt, refractory and mould in a tundish and mould connected type caster, a billet was continuously cast under applying the electromagnetic field. To intensify the magnetic flux density at the intersection, specially designed mould and coil, distance between which becomes small with getting close to the billet surface, were adopted.
For the better understanding about the intial solidification, a numerical model of fluid flow coupled with heat transfer under electromagnetic field was developed, taking account of the free shape of the melt formed by the electromagnetic pressure. Validity of the model was confirmed through the fundamental measurement.
Depth of the wrinkles, which were the disconnected surface shell defect formed during continuous withdrawal, was reduced and the pitch of them became narrow by the application of the electromagnetic field. Cooling rate of the shell was decreased and the same result was obtained through the numerical model.
Even in the presence of electromagnetic field, it was not expected from the model that a vacancy was formed at the intersection by the removal of a liquid metal. Inprovement of surface quality is considered to be responsible for the decrease in the initial shell thickness by the induction heating.

View full abstract

The crystallization process of amorphous Ti-52at%Al alloy was studied using sputter deposited materials. The crystallization occurred at about 800K accompanying the abrupt drop in an electrical resistivity vs. temperature curve. From changes in X-ray and electron diffraction patterns due to annealing at 800K, the crystallization sequence was represented as
Amorphous(Ti–52at%Al)–α(hcp)–α₂(DO₁₉)–γ(L1₀).
From the analyses of lattice parameter, the crystallization was considered to be an isoconcentrational process. The formation of metastable phases in preference to the stable γ(L1₀) one was well predicted by the kinetic analysis in terms of relative nucleation rate assuming that the free energy of amorphous phase was lower than that of undercooled liquid by 3.5 kJ/mol.

View full abstract

The texture changes taking place during the recrystallization of a Ti and Nb bearing interstitial free (IF) steel were investigated. Samples taken from the hot band of a 0.08%Ti-0.01%Nb IF steel were subjected to cold rolling reductions of 70 and 85%, and annealed at 720°C to investigate the effect of cold reduction on the recrystallization texture. The main components observed after cold-rolling were the {001}<110> and {112}<110>, together with the ND<111> fiber increased, especially between {111}<112> and {111}<123>. Higher cold reductions led to stronger intensities of the latter components.
The formation of recrystallization textures in IF steels is discussed in terms of a selective growth theory in which nuclei possessing favorable <110> axis rotation relationships with respect to the deformed matrix undergo preferential growth. By adopting a suitable <110> axis variant selection rule and employing it in conjunction with a nucleus availability factor, it is shown that recrystallization textures can be successfully predicted from experimental rolling textures.

View full abstract

The effect of temperature and specimen orientation on the tearing of type 430 stainless-steel sheet was examined. Tests were conducted at temperatures between –50 and +150°C (223 to 423K) on compact tension and conventional tensile specimens manufactured from 0.82mm thick sheet. Tensile properties and K_R-a R-curves were determined. Both the strength and the toughness of the material decreased as the temperature increased. This result for toughness is opposite in tendency to that found in plate and impact toughness specimens, but is explained by the observation that the sheet-metal specimens were in a state of plane stress. Under these conditions the toughness of the material is strongly influenced by its tensile properties, which declined as the temperature increased. Examination of the specimens in the scanning electron microscope indicated that the mechanism of tearing was ductile at all temperatures. The behaviour of blunt-notched and fatigue pre-cracked specimens was similar, confirming the sensitivity of these materials to notches.

View full abstract

A study was made to explain why the fatigue crack growth (FCG) in the intermediate ΔK regime is accelerated in seawater and other liquid environments, compared with in air. Fatigue crack growth rates (FCGRs) were measured under various environmental conditions using compact type specimens cut from a 667 MPa-tensil-strength steel plate.
Anodic and cathodic polarization at -0.6 and -0.8V_SCE, and removal of dissolved oxygen and salts exerted minimal influence on the FCGR in synthetic seawater (SSW). The FCG-acceleration was observed also in several nonaqueous liquids, though the FCGR in a liquid containing no hydrogen atom in its molucular structure was almost the same as the FCGR in air. The set of results showed that the FCGR was increased by hydrogen formed in a process other than a corrosion reaction.
At an overprotecting potential of -1.2V_SCE in SSW, the amount of hydrogen absorbed into the steel was measured as 0.06ppm or less, and hydrogen-embrittlement type stress-corrosion-cracking (HE-SCC) did not take place at a stress intensity factor or 117MPa √m in a sustained load test. In such an environment, a marked FCG-acceleration was observed, and the FCGR was not influenced by stress ratio, if the crack closure effect was compensated. The observation of fracture surfaces showed little effect of environments. These results indicated that the FCG-acceleration was not caused by the superimposition of HE-SCC.
The hydrogen-assisted fatigue-crack-growth (HA-FCG) model was proposed: Atomic hydrogen is dissociated from water or other environmental substances adsorbed onto steel surface, and then is absorbed into the steel to activate the FCG process.

View full abstract