Various basic data are available for FeOx-SiO2-CaO system in coexistence with metallic iron, which are useful to reductive smelting, but available data are scarce on the effect of minor oxides, especially at higher oxygen potential. In the present paper, the liquidus surface data are reviewed precisely, and recent progress of the experimental and thermodynamic studies is introduced. On the olivine (fayalite) liquidus surface which spreads widely in reductive condition, the addition of CaO or Al2O3 decreases the liquidus temperature, but at higher oxygen pressure more than 10-9 atm, olivine is replaced by spinel (Fe3O4). Above 10-9PO2, the silicate liquid region in the FeOx-SiO2-CaO system is limited by 4 primary surfaces of Fe3O4, SiO2, CaSiO3 and Ca2SiO4, and solid Fe3O4 surface extends substantially by increases in oxygen pressure, Al2O3, and CaO as well as decrease in temperature. However, addition of Al2O3 tends to enlarge the liquid region at SiO2, CaSiO3 and Ca2SiO4 surfaces. The addition of MgO increases the melting temperature substantially regardless the kinds of primary crystal surface. Besides conventional fayalite and calcium ferrite slags, the ferrous calcium silicate (FCS) slag is newly proposed. Instead of triangle FeOx-SiO2-CaO diagram, the phase diagram of CaO plotted against Fe/SiO2 is useful especially for the expression of multi-component system at higher oxygen potential. Also to predict the liquidus surface of these complex systems, the computer aided thermodynamic assessments must be useful method.
In many electromagnetic processings of molten metals, the interfacial motion between a molten metal and a slag plays an important role on the productivity and the quality of products. However the behavior of its motion has not been precisely examined yet because of technical difficulties in the direct observation of it. Here, a new experimental method has been developed to observe the interfacial motion between a liquid metal and a non-conductive transparent liquid. In the method, a laser slit beam which was rayed into the interface from the oblique upper side is so projected on a screen as to show the shape of the interface, which is recorded by a high-speed video camera. The interfacial motion caused by the imposition of a quasi-sinusoidal magnetic field is classified by use of the Shielding parameter Rω such that a periodical oscillation and an irregular motion dominate in the low frequency range (Rω<<1) and the high frequency range (Rω>>1), respectively and the periodical oscillation and the irregular motion coexist around the frequency range of Rω≈1. It is noticed that the increase in viscosity of the liquid imitating a slag acted to suppress the interfacial motion and the wave number appeared larger in the case with the liquid imitating a slag than that in the case without it.
The penetration of a mold flux between a mold and a molten metal which is driven by a mechanical mold oscillation is indispensable for a continuous casting of steel. In this paper a new continuous casting process in which the function of mold oscillation is replaced by an electromagnetic force has been proposed. The basic parameters in the new process were evaluated in a model experimental work using a molten gallium and the penetration behavior of a mold flux was visualized by use of a high-speed video camera. The non-contacting distance caused at meniscus by the electromagnetic force and the consumption of the mold flux in a continuous casting of a molten tin were measured. It is found that the non-contacting distance which is formed by imposing an intermittent alternating magnetic field under no mechanical mold oscillation nearly agrees with the minimum stroke length of the mold oscillation which is required for the success of the continuous casting of a molten tin. Surface quality of cast billets was improved by increasing the consumption of the mold flux and by adjusting the intermittent frequency of the magnetic field to the optimum one existing near the intrinsic frequency of a molten pool.
A new user-friendly three-dimensional simulation system for bar and wire rod rolling processes has been developed. A graphical user interface specially designed for rolling engineers is coupled with strategic FE simulator for rolling process CORMILL. This simulation system enables them to make accurate prediction of load and deformation characteristics of bars and wires under rolling. The effects of roll profiles as well as cross-sectional temperature distributions to width spreads are validated by comparing numerical results obtained by this system and experimental measurement. After four years' developing project organized by ISIJ, this system is now being installed in bar and wire rod rolling industries. A new R&D activities in bar and wire rod rolling will be promoted as the developed system can be easily operated by rolling engineer without special knowledge for non-linear FEM.
Kinetics of austenite to pearlite phase transformation in 0.4 mass%C-X (X=Cr, Ni) steels were investigated. The isothermal transformation curves were analyzed with of the Avrami Equation: X(t)= 1-exp(-Gtn) where X(t) is the fraction transformed and n and G are constants. The exponents n were found to be about 2 except for Cr steels. The value of n for steels containing 0.5 and 1.0 mass% Cr were about 3. This results indicated that the pearlite transformation in 0.4 mass% C steels in present study is controlled by the site saturation mechanism. The parameters G, transformation rate constant of pearlite, were independent of the supercooling from A1 temperature and the driving force of pearlite transformation. These values were dependent on the energy differences of ferrite phase from supercooled austenite phase. This energy difference was equivalent to the driving force of eutectoid ferrite. The predicted transformation rate of pearlite is independent of transformation temperature and containing alloy. The pearlite transformation rate of 0.4 mass% C steels was found to be controlled by eutectoid ferrite growth.
Effects of Mo-C dipole on the development of recrystallization texture are discussed in 0.03 mass% carbon steels containing 0.07 to 1.01 mass% Mo. The hot-rolled steels were annealed at 973K and then cold-rolled 85% in reduction. Decarburized and cold rolled steels were also used. These steel sheets were annealed up to 973, 1023 and 1073K with two kinds of heating rate in argon gas and then quenched. The following results were obtained. (1) The development of γ-fiber in the early stage of recrystallization at 1023K is restrained in all the slowheated carbon steels. But it is accelerated during grain growth up to 1073K in the carbon steels containing 0.51 mass% Mo or more. (2) γ-fiber in all the carbon steels is not developed in quick-heating up to 1023K or 1073K. (3) γ-fiber in the decarburized steels which were slowheated up to 1073K is developed regardless of molybdenum content. (4) Either the increase of Mo-C dipoles themselves or the decrease of single solute carbon atoms as a result of formation of dipoles may contribute to the development of γ-fiber during grain growth by slow-heating.
For the purpose of obtaining refined grains over full cross section for improved ductility in heavy-gage products of a medium carbon 5Cr-Mo-V steel, the effects of pre-transformation microstructures on the grain refining during austenitizing were investigated. Samples were prepared to possess either one of three different microstructures, pearlite, bainite and martensite. These samples were heated slowly to the desired austenitizing temperature followed by rapid cooling, and the formation of new austenite grains was examined. Though austenite grains were not refined in cases of martensite and bainite microstructures, these were remarkably refined in case of pearlite. Portions of pearlite were enriched with carbon, where the transformation temperature into austenite was lower, and this is considered to be the reason that finer austenite grains were formed. Above results show that arranging pearlite microstructure before quenching heat treatment is effective for obtaining refined grains over full cross section in heavy-gage products.
γ grain size refinement is one of the most important factors for microstructural control in heat treatment. Cold or warm deformation prior to austenitization is known to have a beneficial effect on γ grain size reduction, and in this paper, the effect of large-strain deformation at a temperature range just below Ac 1 transformation temperature on γ grain size after austenitization was studied using 0.3%C-9%Ni steel. Cylindrical specimens were deformed up to 70% in a single pass by compression using laboratory deformation simulator. Increase in reduction ratio up to 50% continuously refined γ grain size down to 2 μm after subsequent reheating to austenitization temperature: 700°C. Drastic change in α→γ transformation behavior was found for the reduction above 70%. Spontaneous reverse transformation was induced by such a large strain deformation even without subsequent reheating. It was revealed by TEM observation that γ grain size of the specimen was remarkably refined down to around 0.5 μm. Adiabatic heating due to deformation likely contributes to such a spontaneous transformation. Transformation mechanism for the spontaneous transformation was discussed.
Microstructure and tempering behavior of rapidly solidified layer formed by laser surface melting has been investigated in W-type high speed steels with different Co content. In SKH2 which contains no Co, rapidly solidified layer consists of two regions; a lower region characterized by forming of δ-dendrites and interdendrite compounds of χ-phase, and an upper region in which dendrites have δ-core surrounded by martensite. The addition of Co suppresses the formation of δ-ferrite and leads to primary formation of meta-stable austenite. This results in formation of super-saturated martensite in lower region of SKH3 containing 4 mass% Co and in whole region in SKH4 containing 9 mass% Co. The increase of the amount of Co also changes inter-dendrite compounds to M6C and MC type carbides. This change of microstructure caused by Co addition improves hardness of rapidly solidified layer. Subsequent heat treatment at 823K of rapidly solidified layer of SKH4 produces extremely highly secondary hardening to HV 1000, probably because of the increase of the amount carbides precipitation during tempering.
Fatigue crack propagation behaviors of Ti-5Al-2.5Fe with various microstructure for biomedical use were investigated in air and a simulated body environment, Ringer's solution, in comparison with those of Ti-6Al-4V ELI and SUS 316L stainless steel. The crack propagation rate, da/dN, of equiaxed α structure is nearly the same as that of Widmanstatten α structure in Ti-5Al-2.5Fe in air when da/dN is related to the nominal cyclic stress intensity factor range, ΔK. Ti-5Al-2.5Fe shows nearly same da/dN as Ti-6Al-4V ELI having equiaxed α structure, but shows a greater one than Widmanstatten α Ti-6Al-4V ELI. Without fine precipitated α, the da/dN of Ti-5Al-2.5Fe tested in air and in Paris regime is nearly equal to, but in threshold regime, greater than that of SUS 316L stainlsess steel. Fine precipitated α of Ti-5Al-2.5Fe tested in air makes the da/dN in threshold regime nearly equal to, but in Paris regime greater than that of SUS 316L. When da/dN is related to ΔK, testing in Ringer's solution makes greater the da/dN of both Ti-5Al-2.5Fe and Ti-6Al-4V ELI than that obtained by testing in air. However, when da/dN is related to the effective cyclic stress intensity factor range, ΔKeff, the da/dN of both alloys is nearly the same in air and in Ringer's solution.