For thermodynamic prediction, the deoxidation equilibrium of aluminum in Fe-36%Ni alloy was investigated by employing a cold crucible under an Ar gas atmosphere at 1 773 K. The interaction parameters between aluminum and oxygen by taking liquid Fe-36%Ni alloy as the basis (i.e., solvent) are evaluated as follows:

e_O^Al = -3.8, r_O^Al = -0.9, e_Al^O = -6.4, r_Al^O = 700, r_Al^{Al, O} = 3.1, r_O^{Al, O} = 833

within the composition range of [Al]<1 mass%.
The equilibrium constant for reaction 2Al_{(in Fe-36%Ni)}+3O_{(in Fe-36%Ni)} = Al₂O_3(s) was obtained in the temperature range of 1773 to 1973 K by using data from the present study and from previous works reported:

log K_AO = 0.58−24460/T

The deoxidation equilibrium of aluminum in Fe-36%Ni can thus be thermodynamically described in the range of [Al]<1 mass% using both the first and second order interaction parameters as well as the equilibrium constant determined in this study.

View full abstract

The behavior of magnesium in the desulfurization process of molten iron with magnesium vapor produced in-situ by the aluminothermic reduction of magnesium oxide has been investigated.
The magnesium concentration first increased rapidly, reaching the maximum, and then decreased gradually to a very low level. The magnesium concentration of the molten iron was mainly that of the dissolved magnesium and the following decrease in the magnesium concentration was due to the evaporation from the melt surface and the mass transfer of the dissolved magnesium to the bubble surface. Under the present experimental conditions, the magnesium concentration increased with increasing temperature, pellet mass, carrier gas flow rate and decreasing initial sulfur concentration.
Decreasing the pellet mass and increasing initial sulfur concentration made the desulfurization efficiency higher and decreased the amounts of magnesium dissolving into the melt and leaving the melt. The equilibrium relation between [ppmMg] and [ppmS] did not conflict with the present experimental results at temperatures from 1553 to 1673 K.
A mathematical model for analyzing the behavior of magnesium in the present desulfurization process has been developed. The calculated magnesium and sulfur concentrations are well consistent with the experimental results. The calculated results demonstrate that the existence of the peak of magnesium concentration is reasonable. The present mathematical model can also explain the effects of pellet mass and initial sulfur concentration on the behavior of magnesium injected into the melt.

View full abstract

Recently, the use of the elliptic cross section cylinder for bath smelting operations was adopted by the AusIron process. The high gas flow rates used have been found to generate wave motion in the bath that can enhance refractory wear. The standing wave modes were found for the bath. Two distinct sets of eigenvalues were obtained, one in the major axis and the other in the minor axis. The two sets converged when the ellipticity of the cylinder is zero, which is the shape of a circle. The calculated wave frequency was found to agree well with experimental data. A complete set of eigenvalues have been calculated for elliptic cylinders and fitted to Chebyshev polynomials, enabling quick estimation of the standing wave frequencies.The study showed that the ellipticity of the vessel for the AusIron process needs to be carefully chosen as the standing wave modes that can be generated may result in beat frequencies that can interact with the natural frequencies of the vessel support and auxilliary equipment.

View full abstract

The formation and evolution of the microstructure of casting are important research areas in the field of material science and engineering. The solidified microstructure of aluminum alloy was simulated by combining the CA (Cellular Automaton) model with macro heat transfer. A modified CA (MCA) model, which uses a more similar shape to the actual dendrite to describe the growth grain, was proposed and studied. Because of the huge computational capacity to simulate the microstructure of casting, a relevant parallel computing technique based on the serial arithmetic was developed, which can greatly improve the computing scale and efficiency and can also ensure the computing accuracy as well. The simulation results are compared with the experimental results and agreed quite well.

View full abstract

In this study, the surface crack sensitivities in samples of a Nb-bearing steel and a high Ni grade steel from continuous casting slabs have been examined by simple compression tests at various temperatures between 700-1100°C using a metallography method to evaluate the severity of surface cracks, and corresponding microstructure in the two grades of steel has been investigated. The results show that most specimen have cracks on the hoop surface after 50% compression in height in the temperature range of interest, and the critical hoop strain obtained from the metallography examination gives a clear cracking tendency for the two grades of steel. Microstructure observation revealed that the static precipitation of TiNb(CN)before deformation and thin ferrite film along grain boundaries are important for controlling surface crack sensitivity for the Nb-bearing steel, while coarse grains with flat boundaries, and grain-boundary precipitation of Cu₂S and flake-like Ti(CN) at lower temperature, is responsible for the high crack sensitivity in high Nisteel between 700-1100°C, which implies Cu, S, Ti and N content should be kept as low as possible in thisgrade of steel, and surface temperatures of continuous casting slabs at the straightening point should be above 970°C for Nb-bearing steel while 980°C for high Ni steel to avoid transverse cracking.

View full abstract

The three-dimensional fluid flow in a vertical bending continuous caster was numerically studied. Three dimensionless collision numbers were introduced to analyze the inclusion collision mechanism. The analysis showed that turbulent collisions were the major factor causing inclusions to collide with each other in the continuous caster. Stokes collisions had a minor effect and Brownian collisions were negligible. A mathematical model was then developed to study the inclusion collisions in the continuous caster. The mathematical model considered the inclusion mass transfer and expressed the radius and population of new inclusions after coalescence relative to the mass and population conservation. Since the motion of clustershaped inclusions differs from that of spherical inclusions, the inclusion physical parameters were modified.The results showed that the inclusions congregated approximately one fourth of the face width from the slab edge so that the characteristic radius distribution of the inclusions had a ‘W’ shape, while the inclusion concentration and number density had an inverse ‘W’ shape in the longitudinal direction. More inclusions were trapped near the inner arc and they had larger characteristic radii than those near the outer arc. The concentration and inclusion number density decreased with the distance from the free surface, but the inclusion radius increased.

View full abstract

In this paper, the model⁹⁾ that predicts the surface profile and area of exit cross section of workpiece in oval-round (or round-oval) pass rolling sequence has been improved. Afterward, the generality and robustness of the model was studied to assess the potential that finite element method generally used for predicting it might be replaced by the model. Since only the shape of the inlet cross-section of workpiece and geometry of the roll groove are considered in the model, the problem of obtaining the final rolled shape is greatly simplified and subsequently the computational time required for whole rolling process is a few seconds.
Extensive hot bar rolling experiments at different temperatures (800-1100°C) was carried out to investigate the effect of the change of rolling conditions and material parameters, such as the ratio of the specimen diameter to roll diameter, roll gap (i.e., pass height), roll groove design, steel grades and temperature of material on the model. This model has then been applied to a rod mill to extend its application coverage.
It was shown that the predicted surface profile and area of exit cross section are in good agreement with those experimentally measured for the variation of rolling conditions and material parameters. It was found that if we are interested in the capability for predicting the surface profile and area of exit cross section of workpiece for the entire rod (or bar) rolling line within a very short time, the proposed model might be an alternative which can replace the three-dimensional finite element method usually used in the analysis of rod(or bar) rolling analysis.

View full abstract

In order to perform accurate three-dimensional rolling analysis of sheet deformation, it is required to predict the contact length accurately in consideration of the surface waviness due to the flattening deformation of work roll surface. Flattening deformation analysis based on theory of elasticity and three-dimensional elastic FEM have been developed, and their accuracy in predicting flattening deformation and contact length are validated in this report. As the flattening deformation of the work roll surface changes sharply at the boundary of contact area, the contact length is expected to change according to the number of work roll surface division, especially in circumferential direction, for calculation. Then, the influences of the number of division on roll profile and contact length used for numerical analysis are investigated. The following results are obtained; 1) The differences between the results obtained by three-dimensional elastic FEM and formula for roll flattening of Nakajima and Matsumoto is not significant. 2) When the number of division of work roll surface is increased, the end point of contact region moves toward the downstream side, therefore, the contact length becomes longer and it is necessary to divide the work roll surface into enough much elements. 3) As long as enough much elements are used for the calculation, solutions, such as rolling pressure distribution or thickness distribution of rolled strip, obtained by three-dimensional analysis reveal little difference from those obtained using formula for roll flattening given by Hitchcock.

View full abstract

Ultrafine grain structure in the Type 304 austenitic stainless steel are pursed through dynamic recrystallization.The recrystallization behaviors are studied at various combinations of deformation temperatures and strain rates accompanying the higher strain under a plain strain compression. The effects of the strain, the strain rate and the deformation temperature are investigated, and the relationship between the deformation conditions and the dynamic recrystallized grain size is analyzed. The critical strain needed for the initiation of recrystallization increases with the Z-H parameter. Empirical equations concerning the critical strain and the dynamic recrystallized grain size are discussed, and processing parameter maps are proposed for the complete dynamic recrystallization.

View full abstract

The deformation microstructure of various warm (ferritic) rolled steels was characterized and its influence upon the subsequent annealing behavior determined. The materials investigated included three interstitialfree (IF) steels (stabilized with either titanium or niobium), an extra low carbon (ELC) steel, and four experimental low carbon chromium steels with varying levels of boron, nitrogen and phosphorus. Single pass rolling experiments were conducted in a pilot mill at temperatures between 440 and 850°C and the asrolled microstructures were examined using optical microscopy. Particular attention was paid to the nature and intensity of the in-grain shear bands produced. Partial annealing was conducted to examine the nucleation of recrystallization in the deformed microstructure. Shear bands of moderate intensity were usually formed in the IF steels, which tended to be insensitive to rolling temperature. For the ELC steel, intense shear bands were formed at low rolling temperatures, but at higher temperatures this intensity was found to be drastically reduced. The development of shear bands at the higher rolling temperatures was significantly enhanced by alloying with chromium. The differences in shear band frequency and intensity are explained in terms of the dynamic strain aging behaviors of the various materials. Recrystallized grains were found to nucleate preferentially on the shear bands during annealing, regardless of their morphology or intensity.

View full abstract

A method has recently been developed to estimate the speed with which precipitation reactions occur in power plant steels. It is based on Avrami theory but with an adaptation that allows the treatment of simultaneous reactions. In the present work, a number of approximations and inconsistencies in the theory have been eliminated and this kinetic theory for simultaneous reactions has been modified with the treatments of both diffusion-controlled growth and capillarity effect in multicomponent systems. The modified model can predict not only volume fraction changes of each carbide but also particle sizes. New experimental results on alloy carbide in 3Cr1.5Mo and 2¼Cr1Mo steels are reported and shown to be comparable to the modified theory.

View full abstract

Hot band Fe-3%Si steel (CRGO or cold rolled grain oriented) was cold rolled with different reductions.The main objective of this study was an overall understanding of deformation texture and microstructuredevelopment. Hot band CRGO had a strong α-fiber (RD//<110>) texture. Cold reduction strengthened the α and γ (ND//<111>) fibers, but weakened θ (ND//<100>). All Taylor type deformation texture models were reasonably successful in predicting these bulk texture developments, and the Lamel model seems to be the 'best-fit' model, both in terms of a 'deviation' parameter (indicating differences between experimental and simulated values of idealized texture components) and a 'trend' parameter (indicating the relative change(s) in texture components with strain). The striking feature of the microstructure was the 'selective' appearance of grain interior strain localization's. These appeared at approximately 37° with the rolling direction (RD). Though 37° bands appeared only in orientations with high Taylor factor (M), the absolute value of the Taylor factor alone, was not enough for the appearance of such bands. Negative textural softening or (dM/dε) values, on the other hand, were always associated with the appearance of 37° bands, justifying or explaining their formation on the basis of a macroscopic plastic instability theory.

View full abstract

Precipitation strengthening is important to improve creep strength of heat resisting steels at elevated temperatures. Especially, in the high-Cr ferritic steels recently developed for Ultra Super Critical Power Plants, precipitation behavior is complicated and should be clarified because the correlation between various strengthening factors are still not well understood. In this study, we focus on the change in precipitation behavior of MX type carbonitrides in model steels with different heat treatment conditions.
It is found that dominant precipitation behavior has been changed with cooling rate from normalization temperature before tempering and better creep properties have been obtained in the steel quenched rather than in the one air-cooled. It is confirmed that precipitation of M₂₃C₆ is suppressed and total fraction of MX is relatively increased by the quenching procedure. It is thought that better creep resistance is mainly due to fine distribution of MX that is achieved by suppressing typical co-precipitation of MX such as Nb(C, N)-VN during cooling after normalization and tempering.

View full abstract

The minerals containing fluorine in hot metal pretreatment, BOF and secondary refining slags are identified by using X-ray diffraction method and electron microprobe analysis. From the area fractions of these minerals on a polished cross-section estimated by using an image processor, it was found that a main mineral in hot metal pretreatment and BOF slags is 2CaO · SiO₂ containing fluorine and that in secondary refining slag is 11CaO · 7Al₂ O₃ · CaF₂ or 12CaO · 7Al₂ O₃ containing fluorine. The CaF₂ phase is observed in hot metal pretreatment and BOF slags with the fluorine content above 2.3 mass%. The dissolution behavior of synthetic F-containing minerals such as CaF₂, Ca₅ F(PO₄)₃, 3CaO · 2SiO₂ · CaF₂, 11CaO · 7Al₂ O₃ · CaF₂ and 2CaO· SiO₂-1 (or 2) mass%CaF₂ solid solution, has been investigated in shaking test and the solubilities ofthese minerals are studied as a function of pH value of aqueous solution.

View full abstract

The immobilization of fluorine by calcium aluminum ferrites has been studied using 20 mass ppm F-containing aqueous solution. It is found that the rate of the F immobilization by Ca₂(Al, Fe)₂O₅ is initially slower than that by 3CaO · Al₂O₃ at the solid/liquid ratio of 1/200, but a similiar immobilization degree is obtained at later stage. The hydration product contributed to the F immobilization is the Ca₃(Al, Fe)₂(OH, F)₁₂ phase, which is confirmed by the combination of XRD and electron microprobe analysis. For a ferrite phase of Ca₂(Al_0.68Fe_0.32)₂O_5´ the hydration product has a chemical composition of Ca₃Al_1.54Fe_0.46[(OH)_1-xF_x]₁₂ (x=0-0.13). The F immobilization by 2CaO · Fe₂O₃ does not occur to a significant extent, but the F immobilization by 12CaO· 7Al₂O₃ is accelerated with the addition of 2CaO · Fe₂O₃ due to the formation of the Ca₃(Al, Fe)₂(OH, F)₁₂ phase. It is also observed that F can be immobilized by the formation of ettringite (3CaO · Al₂O₃ · 3CaSO₄ · 32H₂O, 3CaO · Fe₂O₃ · 3CaSO₄ · 32H₂O) through the substitution of F ion for the SO₄²⁻ site.

View full abstract