The electrical conductivity of the molten slags of the CaF₂+Al₂O₃ and CaF₂+Al₂O₃+CaO systems for the practical ESR was determined, using the constant-current single pulse technique with three lead-electrodes. All the measurements were carried out under high-purity argon atmosphere, employing a high-purity molybdenum metal crucible with pure iron wire electrodes. The effects of FeO, MnO, MgO, Cr₂O₃, TiO₂, SiO₂ and other oxide components and temperature on the conductivity of the slags were examined. The results indicated that in the common concentration ranges for the ESR practice, the specific conductivity values of the slags in these two systems are monotonously increasing with the FeO and MnO contents in the slags following an essentially similar pattern; the additions of MgO, Cr₂O₃ and TiO₂ change also evidently the conductivity but in another roughly similar mode and make it have a maximum value, whilst the influence due to a small amount of SiO₂ (≤1.5 mass%) is relatively not large; the conduction of CaF₂-based multi-component and complicate liquid slags may be treated as a rate process.

View full abstract

The deoxidation of liquid nickel-copper and nickel-cobalt binary alloys with silicon has been measured at temperatures ranging from 1 450 to 1 650°C by the use of quartz crucibles.
On the nickel basis in nickel-copper binary, the deoxidation product, log K'_Si(Ni-Cu), which increased with the addition of copper up to 40 mass% containing 0.5 mass% silicon, was expressed to be:
log K'_Si(Ni-Cu)=log K_Si(Ni)+0.0342[%Cu]–0.0004[%Cu]²
The effect of temperature on the equilibrium constants of the deoxidation reaction in copper, obtained by the extrapolation, was estimated as follows:
log K_Si(Cu)=–15 680/T+2.16
while log K'_Si(Ni-Cu) on the copper basis was given by the expression:
log K'_Si(Ni-Cu)=log K'_Si(Cu)+0.012[%Ni]–0.0001[%Ni]²
up to 40 mass% of nickel.
On the nickel basis in nickel-cobalt binary containing 0.55 mass% silicon, log K'_Si(Ni-Co) up to 20 mass% of cobalt was represented to be as follows:
log K'_Si(Ni-Co)=log K_Si(Ni)+0.005[%Co]
The temperature dependence of the equilibrium constants at temperatures between 1 550 and 1 650°C for the deoxidation of pure cobalt was expressed to be:
log K'_Si(Co)=-15 270/T+1.81
whereas log K_Si(Ni-Co) on the cobalt basis was found to be:
log K'_Si(Ni-Co)=log K'_Si(Co)-0.001[%Ni]
up to 60 mass% of nickel.
Using these values of the deoxidation product, the interaction parameters for the effect of these elements on the activity coefficient of oxygen in metals have been established.

View full abstract

The effects of composition and temperature on viscosities of liquid metals have been studied by systematical analysis of data found in the literature to estimate the viscosities of liquid alloys. These viscosities are expressed by the following equations:
η = A exp(B/RT)
A=(1.7×10⁻⁷ρ^2/3T_m^1/2M^−1/6)/[exp(B/RT_m)]    [Pa · s]
B= 2.65T_m^1.27    [J · mol^-1]
where ρ is the density of alloy [kg·m^-3], T_m is the absolute temperature of liquidus [K], M is atomic weight [kg·mol^-1], and R is gas constant [8.3144 J·mol^-1·K^-1]. Values calculated from the above equations for the viscosities of various liquid binary alloys were in good agreement with the experimental data from the literature. Using these equations, it is possible to estimate viscosity for any liquid alloy at any temperature.

View full abstract

High-carbon ferromanganese has been produced by electric arc furnace in Japan. But its production cost has increased because of a speedy rising of electricity cost and Japanese ferromanganese has lost competitiveness in the international market. So, for the purpose of decreasing its production cost, high-carbon ferromanganese production tests (production rate=1.9-4.3 t/d) were performed using a melting test furnace with coke packed bed injected with highly oxygen enriched air and a large quantity of pulverized coal. Moreover the operation indices of a commercial plant of 170 t/d in production capacity were estimated by heat and mass balance model, and following results were obtained.
(1) In this test, high-carbon ferromanganese of [Mn]=75% was produced stably using the coke blended with 56% non-coking coal, with coal rate of 1 502 kg/t, coke rate of 1 087 kg/t and productivity of 3.11 t/(d·m³).
(2) The total coal consumption of the commercial plant of this method is estimated to be less than that of the blast furnace in spite of higher fuel rate because a large quantity of pulverized coal is used in this process. Through these investigations, this process seems to have a prospect of being available as a ferromanganese production process instead of electric furnace method.

View full abstract

There are electric furnace (EAF) method and basic oxygen converter (BOF) method as a large quantity of scrap melting method. These methods have problems respectively as follows: high electricity cost in Japan for EAF method, and low heat efficiency and refractory life for BOF method. To investigate the method to meet these problems, scrap melting test was performed using a melting test furnace with coke packed bed injected with highly oxygen enriched air and a large quantity of pulverized coal and the following results were obtained.
(1) The pig iron whose quality is equal to that of the blast furnace can be produced with high carburization and high desulfurization using scrap (100% of material) and coke for blast furnace.
(2) The fuel rate of 240 to 290 kg/t and productivity of 14.7 t/d/m³ were obtained.
(3) The fuel rate can be decreased and productivity can be increased by post combustion with air injection from the shaft wall.
Based on the above mentioned results, the comparison of energy consumption of this method, EAF and BOF was made and a possibility was confirmed that the energy consumption of this method was the least among three.

View full abstract

Since particle size of deadman coke greatly affects the permeability of gas and liquid in the lower part of a blast turnace, elucidation of the behavior of the replacement of deadman coke is an important subject. This paper describes the degradation of deadman coke by t he reaction with molten FeO. With use of a one-dimensional reaction model in which two kinetic parameters were determined from a fundamental experiment, the degradation of the deadman coke in an actual blast furnace was simulated. The effect of FeO in molten slag on the degradation of deadman coke concerns reduction ratio on the melting down of ore, temperature, and productivity. It takes about 2 weeks that vertical size distribution of degrading deadman coke becomes stationary pattern.

View full abstract

By investigating the behavior of spinel ferrite particles in EAF dust, an insight into the mechanism of spinel ferrite dust generation was examined. The results are summarized as follows: (1) the chemical formula of spinel ferrite particles in EAF dust is (Mn_xZn_yFe_1-x-y)Fe₂O₄; (2) when oxygen is transferred at the slag-metal interface, a series of oxide layers form on the molten steel (going from the surface to the interior of the liquid, one progresses through (Mn, Zn)Fe₂O₄, (Mn, Zn)Fe₂O₄-Fe₃O₄, Fe₃O₄-(Mn, Fe)O and (Mn, Fe)O, respectively); (3) some of the coarse spinel ferrite powders are often hollow contain surface blowholes suggesting that CO gas atomizes liquid iron oxides; and (4) CO gas evolution atomizes various liquid oxides. These droplets may oxidize to varying degrees as they solidify and cool producing various compositions of (Mn, Zn)Fe₂O₄ and/or (Mn, Zn)Fe₂O₄-Fe₃O₄ powders.

View full abstract

The formation of clogging materials in an immersed nozzle during the continuous casting of Ti-stabilized stainless steel was studied by plant and laboratory experiments. The major materials deposited were globular Ti oxide as curved chain and Al₂O₃ as large, black agglomerates. On the contrary, TiN and Al₂O₃ were observed in the steel in the tundish, and Ti oxide was scarcely found. The curved chain of globular Ti oxide could only have been formed by reoxidation after nitrogen absorption in the laboratory experiments. The formation mechanism for Ti oxide from TiN decomposition is proposed for the clogging behaviour of Ti-stabilized stainless steel.

View full abstract

The mechanism by which ODS alloys are provided with excellent oxidation resistance remains still ambiguous. The ambiguity originates from a wide variety of scale features: pores, composition, stress etc. We used Al₂O₃ coating film prepared under a same condition in order to avoid the ambiguity and in addition to obtain fundamental information for the development of ceramic coating technique. This work was made on the basis of the following five models. I. Dispersed oxide suppresses the spalling of Al₂O₃ film through trapping a detrimental element S. II. ODS alloys form a diffusion barrier enriched in Y at the Al₂O₃ film/alloy interface and suppress the penetration of alloying elements into the film. III. The above barrier acts as an adhesive of Al₂O₃ to the substrate alloy. IV. Incorporation of Y into the Al₂O₃ film retards the diffusion of alloying elements within the film. V. The above incorporation of Y suppresses the spalling of Al₂O₃ film by increasing the plasticity of the film. We examined these models directly or by simulation experiment using Y₂O₃ undercoat as a barrier and an Al₂O₃ coating film doped with Y₂O₃. The following conclusions are reached: (1) Model I is the most important for suppressing the spalling of Al₂O₃ film. (2) Models II and IV have also remarkable importance for diffusion retardation. (3) Models III and V have minor importance.

View full abstract

The structure and composition of the intermetallic dross phases formed in zinc baths containing (0-0.1) wt% nickel at 450°C have been investigated by X-ray diffraction and energy dispersive spectroscopy. Two intermetallic phases were identified and the presence of each of these phases depends on the nickel content of the bath. ζ (isomorphous with FeZn₁₃) phase, which probably contains less than 0.5 wt% nickel, was the intermetallic phase present in baths with bath nickel contents of less than 0.06 wt%. In baths containing (0.06-0.09) wt% nickel both ζ and Γ₂ (isomorphous with FeZn₄) phases were present. When the bath nickel content was above 0.06 wt%, the nickel content of the ζ phase was found to increase significantly whilst the nickel content of the Γ₂ phase only increased slightly with increasing bath nickel content. When the bath nickel content was above 0.09 wt%, the intermetallic phase was found to be mainly Γ₂. Based on the experimental data, a more precise verision has been proposed for the zinc-rich corner of the Zn-Fe-Ni phase diagram at 450°C.

View full abstract

A Ti-49at%Al gamma-base titanium alloy was deformed at high temperature (T=1150-1200°C) and low strain rates as well as at lower temperature (T=1000°C) and higher strain rates. Textures with distinctly different strength occurred depending on the deformation conditions. Subsequent heat treatments were applied in order to investigate the development of textures and microstructures. Textures and microstructures after deformation at 1150°C<T<1200°C are stabilized by α₂-Ti₃Al-precipitates at the γ-TiAl grain boundaries which occur after the first annealing at 1050°C.

View full abstract

The austempering treatment for spheroidal graphite cast iron is to produce a residual compressive stress at the surface of specimen and some retained austenite corresponding to the austempering temperature. This study was carried out to investigate the effect of the residual stress and the retained austenite on the fatigue behavior. The degree of the residual stress relaxation due to cyclic load which varies with the applied stress ratio (σ_max/σ_B, σ_max=cyclic max. stress, σ_B=tensile strength): at the higher stress ratio the change of the residual stress is dominated at the first loading. At the lower one the extent of the relaxation of residual stress depends on the magnitude of cyclic load and the number of cycles. While the amount of the retained austenite decreases progressively during fatigue process regardless of the applied stress ratio.
These results suggest that a measurement of the residual stress and/or the retained austenite is available to predict the degree of the fatigue damage for spheroidal graphite cast iron.

View full abstract

The equilibrium shape of aliquid meniscus at a liquid-gas phase boundary of a system in which solid, liquid, and gaseous phases coexist is given by a balance of forces acting on the system. It is often greatly desirable in many materials processing applications to know and be able to predict this meniscus shape, because controlling it often affects the structure or properties of the resulting material. A method of calculation is introduced along with the Surface Evolver computer program, and solutions for simple geometries obtained using this program are compared with existing analytical and simple numerical solutions.

View full abstract

The equilibrium shape of a liquid meniscus at a liquid-gas phase boundary of a system in which solid, liquid, and gaseous phases coexist is given by a balance of forces acting on the system. A method of calculation was introduced in Part I of this series for determining this equilibrium shape. Here, this method is used to calculate the equilibrium shapes of solder joints surrounding gullwing and jbend type lead wires present in surface mounted integrated circuits. A study is also done to determine the optimal volume needed to produce a good solder joint. Criteria are established, dimensionless calculations are done for gullwing and jbend leads, and a method is outlined for applying these results to lead wires with different geometries.

View full abstract

Mechanism of vortex formation in continuous casting moulds was investigated using a full scale water model. It was found that vortices occurring near the SEN was due to interaction between the two surface streams flowing inwards. Biased flow as such does not generate vortices in the mould, but increases its depth and frequency. Vortex depth increases with increasing casting speed. Vortex formation in the mould can be reduced by diminishing biased flow and optimising the SEN design. When casting wide slab at a shallow SEN immersion depth, vortex formation takes place near the narrow faces.

View full abstract