During annealing of steels different kinds of enrichments are observed on steel surfaces. Equilibrium segregation of a dissolved element leads to enrichment in the range of a monolayer, two interacting elements can form two-dimensional compounds by cosegregation and also chemisorption induced segregation leads to monolayer surface compounds. If the concentrations or pressures are exceeded which are necessary for the formation of three-dimensional bulk compounds, enrichments occur due to surface precipitation resp. oxidation. On the other hand, oxidation can cause segregation or accumulations of dissolved elements beneath the oxide scale.
These different phenomena are fundamentally discussed and examples are presented, which are important in the steel production and for the steel properties.

View full abstract

The plasma arc melting (PAM) process has attracted keen attention in melting titanium, because it has many advantages such as melting under the atmospheric pressure and refining by utilizing ultra-high temperature and plasma gas.
In particular, in melting titanium alloy, the plasma arc melting process produces less evaporation loss of alloying elements, enables comparatively easy composition control, and provides less restrictions in material form.
It is very important to raise the melting efficiency in plasma arc melting from the commercial point of view. So, basic investigation was carried out on arc voltage and heat efficiency when titanium-based materials were melted with plasma arc. The results are summarized as follows:
(1) When argon, neon, and helium are used for plasma torch gas, the plasma arc voltage becomes higher in that order, and the same tendency is found in the fraction of heat transferred to the molten pool. By substituting helium for argon as torch gas, the melting rate can become double because of the increase of arc voltage and heat efficiency to the melting material.
(2) When titanium sponge is melted, arc voltage increases and greatly fluctuates by the vapor of chlorides contained. Alkali atoms are detected in the plasma arc column by spectrometry analysis. The temperature of plasma arc column is estimated.
(3) Plasma arc voltage is shown as a linear equation of the arc length. The value of the intercept of the line is greater when helium is used as torch gas than as atmosphere gas.

View full abstract

A thermal stress analysis within a coal briquette is performed with various heating patterns in order to simultaneously make the raeduction of heat consumption rate and the enhancement of formed coke quality. The single briquette is charged into the small oven, which is purged of air and filled with nitrogen gas. The oven is then placed in the X-ray CT equipment and heated by the prescribed heating pattern to continuously oberve the crack formation behavior of the briquette. The thermal stress analysis model which previously developed by authors is used to estimate the stress distribution within the briquette in this study. The effects of creep characteristics of the plastic and semi-coke layer, dilatation of the plastic layer and the pyrolytic reaction fraction dependence of thermophysical properties are considered in this model. The calculated stress distribution and deformation behavior in the briquette are compared with experimental observation results by X-ray CT. The results show that the model used in this study can quantitatively predict the stress distribution within the briquette and the model is an effective tool to investigate the high performance formed coke making process.

View full abstract

To clarify the refining ability of the CaO-CaCl₂-CaF₂ systems and the effect of replacement of CaF₂ with CaCl₂ on the CaO activity, the activities of CaO in the CaO-CaCl₂ and CaO-CaCl₂-CaF₂ systems were measured at 1673 K by empolying a chemical equilibration technique. The activity of CaO relative to liquid CaO in the CaO-CaCl₂ system was observed to have negative deviation from ideal behavior with the solubility of CaO being X_CaO =0.32. The activity of CaO in the CaO-CaCl₂-CaF₂ system at 1673 K decreases by replacing CaF₂ with CaCl₂ at constant CaO content. The refining ability of the CaO-CaCl₂-CaF₂ fluxes is discussed.
The activity coefficient of calcium in silver was found to be constant up to the calcium content of 0.625 mass%. The standard Gibbs energy of the formation of CaO was determined from 1573 to 1773 K.

View full abstract

The effects of a horizontal and a vertical DC magnetic field on the Rayleigh-Taylor instability between two inviscid liquid metals were studied by linear analysis. Since the dispersion relation in the vertical field is in the same form as that in the horizontal field, the stability characteristics of the interface are the same in both cases. In a case where the upper fluid has a higher density than the lower one, no magnetic field can support the upper fluid, and any small fluctuation at the interface always arises in a non-oscillatory mode. The growth rate of the instability is proportional to –2/3 power of the field intensity in the high intensity region. On the other hand, when the density of the lower fluid is higher, the fluctuation decays in an oscillatory mode. There is an optimum field intensity at which the fluctuation decays most rapidly. The intensity is not affected strongly by the conductivity of the liquid and value is less than 1 Tesla for ordinary metals. The decay rate increases with increase of the field intensity below the optimum value, but decreases proportionally to –2/3 power of the field intensity above the optimum value. The convective vortex of the fluid induced by the instability is elongated in the direction of the applied magnetic field.

View full abstract

This work aims at optimising the mineral phase composition of iron ore sinter to improve its properties. At first, the conditions that favour the slag formation in iron ore sintering were studied by measuring the conventional melting temperature. The ease of liquid phase formation was estimated in relation with the nature of the ores and with the amount of additions.
In a second step, the reactions between slag and the iron ore nuclei in the mix granules were studied. For this purpose, a laboratory furnace reproducing approximately the atmosphere and thermal profile observed in the industrial conditions was conceived. These experiments enabled to relate sinter mineral phase composition to the process conditions and to the type of iron ore.
At last, the results obtained in the laboratory experiments were confirmed by pilot plant trials. The relationns between sinter properties and mineral phase composition were then establishes.

View full abstract

With reference to results of off-line model experiments and actual furnace operation tests performed beforehand, a high-rate pulverized coal injection operation test was carried out at Kimitsu No. 3 blast furnace and the following conclusions have been derived:
(1) A stable operation has been achieved at the pulverized coal rate of 190 kg/t-pig by the control of heat flow ratio and the burden distribution control of peripheral gas flow suppression, with the production of low Al₂O₃, high-reducible sinter and the raise of cold and hot strengths of coke.
(2) With an operation performed at the pulverized coal rate over 200 kg/t-pig and the coke rate below 300 kg/t-pig, it has become clear that a technical subject posed at this operation is how to improve the reduction-meltdown behavior in the lower part of furnace.
(3) It has become clear that a factor controlling the pulverized coal injection at the actual furnace is the discharge of unburnt char from the furnace top. Setting the upper limit of injection rate certainly is a subject to be worked out hereafter.

View full abstract

The formation mechanism of ZnO-containing dusts was studied using an apparatus where the evaporation and formation of dust from liquid iron was controlled by the composition of an Ar-O₂ gas stream. The investigations were made at 1600°C and oxygen partial pressure of 0 to 1 bar in the gas phase. It was shown that he total amount of the dust increased with increasing oxygen partial pressure until a maximum value was reached. Futher increase of p_O2 resulted in a reduction of dust formation and its complete prevention at p_O2–0.8 bar. The rate of zinc evaporation also attained a maximum with increasing oxygen partial pressure, while evaporation of iron slightly increased up to an oxygen partial pressure of 0.5 bar. The observed experimental results are described with the aid of a kinetic model.
Analysis of size distribution and structure of oxide particles indicates that 20% of particles are smaller than 1 μm and 50% smaller than 4 μm.
It is concluded that evaporation and dust formation are controlled and can be optimized by changing process parameters scuch as composition of the Ar-O₂ gas. Further decisive parameters such as gas flowrate, temperature and stirring intensity of the melt were kept constant in the present work.

View full abstract

Mathematical and physical modeling have been extensively used to study the fluid flow phenomena taking place in continuous casting tundishes. Tracer dispersion studies are commonly carried out to characterize the overall fluid flow behavior within the tundish. It is shown in this paper that the buoyancy forces introduced by using a tracer whose density is different from that of the fluid in the tundish may have a significant effect on the fluid flow behavior within the tundish. A three-dimensional mathematical model for solving fluid flow equations incorporating the buoyancy force has been developed to study the fluid flow behavior in continuous casting tundishes. Water modeling experiments have also been carried out to verify the mathematical model.

View full abstract

How to start up the casting operation is one of the serious problems encounted in twin-roll strip casting process. During the initial pouring stage, the metal pool should be progressively developed to reach the optimal pool level and the solidified metal layers formed on the surfaces of two opposite-rotating rolls should be welded together before the dummy sheet is withdrawn from the minimum roll gap. Improper operation with inaccurate withdrawal time often results in unstable process and products as well as the breakup of cast strip, which will be detrimental to the formability and subsequent coling of cast strip. Therefore, the proper withdrawal time of dummy sheet plays a very important role in the initial stage of a vertical twin-roll casting process. The accurate withdrawal time is, however, very sensitive to the variations of thermal distribution, flow conditions and metal-roll interactions.
In order to elucidate the fundamental transport phenomena in twin-roll casting, a commerical software called ProCAST was employed to simulate the transient fluid flow, heat transfer and solidification behaviors during the early stage of the process in this study. The coupled set of governing differential equations for mass, momentum and energy blance were solved with the finite element method and the transient free surface problem was treated with a Volume of Fluids (VOF) approach. An enthalpy method was employed the handle the phase change during solidification. The advantage of this model is the great capability to treat the problems of moving boundary and free surface of fluid. With this mathematical model, the filling sequences, flow patterns and corresponding temperature profiles in the metal pool under an actual casting condition of a vertical twin-roll strip casting process were simulated in this study. From the calculated results, how the metal pool progressively develops between the rolls and how the solidified layer grows in thickness with time on the rotating rolls are possibly revealed. Moreover, the metal pool level, the solidification front as well as the position of solidification end could be obtained. With these available informations, the optimal withdrawal time of the strip casting process can be thoughtfully determined.

View full abstract

The dendrite growth rates for Ag-5mass%Cu and Cu-10mass%Ni alloys are measured as a function of the degree of bath undercooling. The growth rates for pure silver and copper in the literature are compared with the model and the characteristic value of interface kinetics is determined by fitting the prediction curves the the data. The determined values of the upper limit or interface advance, V₀, are 500 ms^-1 for silver and 600 ms^-1 for copper, which correspond to the values of the linear kinetic coefficient of 0.5 ms^-1K^-1 for silver and 0.6 ms^-1K^-1 for copper, respectively. With the determined values of V₀ the data of dendrite growth rates for alloys are analyzed using the LKT model. The data are in good agreement with the estimation by the model only when nonequilibrium effects are assumed. The estimated values of growth rate coefficient, β₀ (=a₀/D_L) are about 1.0 m^-1 s for both alloys. The result shows that the nonequilibrium effect proposed by Aziz and by Boettinger and Coriell are quite applicable.

View full abstract

A model experiment was made for study on the effects of temperature on the reaction rate in isothermal peritectic reaction in iron-carbon system by using a soliid/liquid diffusion couple method. Measurements of the variation in carbon concentration over δ, γ, and liquid phases show a decrease in the carbon concentration from the γ/liquid interface to the δ/γ interface and an equilibrium partition of carbon at both the interfaces. The γ-phase grows into both δ and liquid phases. A regression analysis of the thickness of the γ-phase, x (μm), with the reaction time, t (s), results in the following relations.
x=36.6t^1/2 at 1755 K
x=54.7t^1/2 at 1722 K
x=71.0t^1/2 at 1689 K
These relationships imply that the peritectic reaction follows the parabolic law and that the reaction rate is higher at lower temperatures. It was found that about 90% of the γ-phase formed through the peritectic reaction is the transformation product from the δ-phase and about 10% the solidification product from the liquid phase.

View full abstract

The Rotary Reduction Mill, called the RRM, has been developed to reduce solid metallic materials efficiently in a one-pass operation. This RRM is a rotary mill, in which three rolls are arranged around the pass line, and the roll-axes are inclined or are inclinable to the pass line. This paper reports the effect of rotary rolling conditions on rolling load, rolling torque and power consumption in the RRM. To investigate the effect, heavy-reduction rolling was carried out by hot-rolling solid materials, such as C-steel (S45C), stainless steels and high-Ni alloys, in the RRM. In this paper, first, the effect of face angle of rolling (i.e. "roll face angle minus cross angle") on feed efficiency will be discussed. Next, the effects of various other rotary rolling conditions (including cross and feed angles, face angle of rolling, elongation ratio, rolling temperature, roll surface conditions and roll diameter) on rolling load, rolling torque and power consumption in rotary rolling will be described in detail.

View full abstract

The dissolution behaviour of a 22% chromium duplex stainless steel with very low nitrogen content and containing various amounts of ruthenium was investigated. It was found that the ruthenium additions inhibits the anodic dissolution behaviour when active corrosion took place. Furthermore it was established that only the ferrite phase corrodes preferentially at various potentials in the active and passive regions, which is contrary to the behaviour of similar commercial duplex stainless steels with significant nitrogen contents, and which do not contain ruthenium.

View full abstract

Barium Carbonate (BaCo₃) is commonly used as an energizer in pack-carburization. Investigation was therefore carried out to establish the suitability of sea shell and coconut shell as energizers in pack-carburization of mild steel. Samples were carburized at 950°C in carburization compounds containing charcoal and energizing aterials at predetermined proportions. The results obtained showed that sea shell was a more effective energizer than the coconut shell and compared favourably with barium carbonate. Sea shell showed 82.5% efficiency compared with BaCO₃ and 134.7% with limestone (CaCO₃) at 950° carburizing temperature. It was found that 30% sea shell addition to charcoal gave optimal carburization case depth which was 2.8 times the values obtained from 100% charcoal. However, there was no significant difference between 100% charcoal and compounds containing various proportions of coconut shell in the carburizing kinetics.

View full abstract

Effect of carbon content on the static softening taking place in hot deformed austenite was studied by means of interrupted tension and compression tests at a temperature of 1133 K and at a strain rate of 2×10^-3s^-1 for plain carbon steels. The softening curve after dynamic recovery (DRV) consists of two stages, followed by complete softening. In contrast, the softening curve after dynamic recrystallization (DRX) consists of three distinct stages accompanied with three plateaus, followed by incomplete softening. These can be controlled by metadynamic recovery (MDRV) and recrystallization (MDRX) in addition to classical static recovery (SRV) and recrystallization (SRX). The static softening taking place after DRX sensitively depends on the volume fractions of DRX nuclei, growing DRX grains and fully work heardened DRX grains. The increasing in the fractions of DRX nuclei and growing DRX grains causes the amount of rapid softening in stage I and incomplete softening in stage III to increase, respectively. The static softening rates on DRV and DRX matrices increase with carbon content because of the increased diffusivity of vacancies, which can promote the operation of static restoration mechanisms, i.e. SRV, SRX, MDRV and MDRX.

View full abstract

Microstuctural modifications of bonding interface in explosively welded Ti/SUS 304 and Ti/SS 41 steels clads were studied by transmission electron microscopy. Metastable phases such as amorphous and supersaturated solid solution of β-Ti were observed at the interface. These were considered to be the trace of melting and subsequently rapid solidification of thin layers along the contact surface of both the parent materials. It is concluded that the melting layer is responsible for the bonding of explosively welded Ti/steel clads.

View full abstract