Small-angle neutron scattering measurements were carried out on liquid Sn-0.4mol%SnO₂ (solid spherical particle)-1.1at%S, Sn-0.4mol%SnO₂ and Sn-1.1at%S systems at 723 K. The scattering intensity for Sn-SnO₂ system is proportional to the parameter, Q^-4, while the scattering intensity for Sn-SnO₂-S system is lower, showing nearly Q^-3 dependence. To analyze them, the scattering intensity has been calculated for Sn-SnO₂ and Sn-SnO₂-S systems with various particle sizes and thicknesses of the surface layer. For Sn-SnO₂-S system, the best fit with the experimental data has been obtained for a particle size of 4μm and the surface layer thickness of 100 Å, consisting of S.

View full abstract

Flux-metal interfacial tension has been examined by X-ray photography of a metal drop melted in a molten flux. The experiments were carried out at 1 580°C using a zirconia crucible with low solubility in the fluxes. The systems used in the investigation were as follows:
Metal phase: Fe-O alloy, oxygen content in the alloys varied from 0.002 to 0.070 mass%.
Flux phase: Ternary system: CaO-SiO₂-M_x(F, O)_y, where M_x(F, O)_y represents NaF, CaF₂, Al₂O₃, Na₂O, MgO, Li₂O or ZrO₂ and concentrations varied from 2 to 40 mass%. Multicomponent system: CaO-SiO₂-Al₂O₃-Fe₂O₃-Na₂O-NaF-CaF₂-MgO-Li₂O-ZrO₂.
The interfacial tension was observed to increase with increasing NaF, CaF₂, Na₂O, MgO or Li₂O contents, while only minor changes in the values were observed with varying Al₂O₃ or ZrO₂ contents in CaO-SiO₂ or CaO-SiO₂-Al₂O₃ systems. The interfacial tensions between flux and molten iron decreased with increasing oxygen activity in iron,

γ=γ₀-0.324 ln(1+116a_O)

View full abstract

Effect of titanium addition on the size and dispersion of MnS inclusions in Fe-0.1C-1Mn-0.02S(mass%) base alloys has been investigated by microscopy. Titanium addition brings about a marked decrease in the size of MnS inclusions which are otherwise large and globular in Ti-free specimens. The formation mechanism of MnS inclusions is discussed on the basis of phase diagram information and thermodynamics of particle engulfment during solidification. It is shown that the reduction in size of the MnS particles consequent on Ti addition is directly the result of successive nucleation events involving (Ti, Mn)O and MnS at the solid/liquid interface of iron.

View full abstract

Three-dimensional structure of iron-ore sintercakes, which are prepared by pot test (90 mm-dia.; 400 mm-height) with varying the location in bed and Al₂O₃ content, has been investigated with the use of an X-ray CT and a three-dimensional image analyzing system. It has revealed that sintercake basically consists of a continuous plate-shaped matrix including numerous closed pores of under-2 mm diameter and a continuous intercommunicating pore. Moreover, the shape of the matrix/pore in the lower part of the bed becomes round, though few changes by Al₂O₃ can be detected.

View full abstract

Particles of hematite ores in a porcelain boat were reacted at 873∼1273 K by H₂-CH₄ mixtures having low sulfur activities a_s based on Fe/FeS equilibrium where metallic iron remained stable. In the tests without sulfur, much free carbon appeared beside metallic iron and iron carbide. The free carbon increased with temperatures. In the tests with sulfur, nearly one hour reaction could convert their particles mostly to iron carbide such as Fe₃C, especially favorable in a_s=0.1∼1.0 at 1173∼1223 K. The conversion yield was sensitive to inlet flow volume ratio H₂/CH₄ and ore type, while the authors' previous test in H₂-CO gas mixtures had already indicated their insensitivity.

View full abstract

The traditional de-NOx processes in iron ore sinter plants, power plants or in refineries is to treat the waste gas by the Selective Catalytic Reduction (SCR) process or Selective Non-Catalytic Reduction (SNCR) process which removes the nitrogen oxides from the flue gas. Space and huge capital requirements are needed for installing of such de-NOx facilities. This study takes a different approach, aiming at reducing the formation of NOx during sintering rather than treating the waste gases. Our preliminary experimental findings showed that NO was drastically decreased by introducing hydrocarbons into the sinter mix. Rice husk and sugar had the potential to improve the sinter bed permeability and shorten the high temperature duration time in the combustion zone. The encouraging results showed that the NOx concentration was decreased from 223 to 160 ppm, or emission was decreased from 533.8 g/t sinter to 283.3 g/t sinter by 1 % sugar addition and the productivity was increased from 37.5 to 45.4 t/m²/24 h. The sinter qualities of Tumbler Index (ISO+6.3 mm%) was kept the same, the RDI was improved from 31.5 to 27.3 %.

View full abstract

A new technique using a static magnetic field (DC field) for the modification of macroscopic segregations and porosities in continuously cast billets or blooms has been investigated. If the DC field is applied to the crater end area in a strand, numbers of the V-segregations and porosities could be expected to result in a decrease, since the molten steel flow in the semi-solid region seems to be modified by the DC field. This phenomenon was observed when the magnetic flux density of the DC field of 0.2 to 0.42 Tesla was applied during the solidification of the center of the cast ingots, where the solid fraction had values of 0 to 1.0, which is in the final stage of solidification. In addition, the application of the DC field to the center part of the ingot with a solid fraction range, 0 to 0.3, showed similar effects. Hence, the application of a higher magnetic flux density, when the ingot contains a low solid fraction at the center part, was found to result in a decrease of the V-segregation and porosity. It is suggested that the DC field homogenized the fluid flow velocity in the semi-solid region at the center part of the ingot.

View full abstract

A new continuous rheocasting process for steels and high temperature alloys has been developed by Inland Steel incorporating a dual chamber casting machine utilizing separate electromagnetic stirrers around each chamber. Initial casts on the rheocaster were plagued by breakouts and very rough as-cast surface morphology. The mechanism for solidification and formation of cast surface morphologies on as-rheocast bars has been determined and verified by comparing predicted surface morphologies to actual surface morphologies. The mechanism was then used to develop new withdrawal patterns to eliminate sticker breakouts and obtain good surface quality.

View full abstract

Preventing crack formation of solidifying steel shell at increasing speed in continuous slab casting operation is a prerequisite to increase the ratio of hot charge rolling and hot direct rolling of cast slabs. Crack formation is influenced by mechanical properties of the steel shell near its solidus temperature. Thus, fracture strength of solidifying steel shells containing 0.004 to 0.70 mass% C has been measured in-situ by using a method similar to "Submerged Split Chill Tensile (SSCT)" test by Ackermann, Kurz and Heinemann. The fracture strength at an early stage of solidification, when the shell thickness is 2 to 5 mm, has been found to be 1 to 3.5 MPa which is lower than that determined by conventional hot tensile tests, since the crack initiates at and propagates in interdendritic region in the shell in the in-situ SSCT test. Comparing the fracture strength of the shell with the frictional force at the shell/mold interface, upper limit of casting speed of slab caster is calculated to be around 8.5 m/min.

View full abstract

An ultrawide range viscometer which is based on a combination of the indentation, parallel plate and rotating plate methods has been developed. A cylindrical glass sample is placed between spiked and smooth discs and pressed by a suitable load and its deformation rate is measured under an ascending temperature process. At first, the spike penetrates into a glass and then the surface of spiked disc causes a bulky creep of the sample. On melting of the sample, the concentric rotation method is used in which a disc is rotated at adequate speed and the torque exerted on the disc is measured. Thus, a wide range viscosity, extending from 10¹² to 10¹Pa·s, can be sequentially determined by the measurements of indentation rate, creep rate and rotational torque. Since the principles and a mechanism of this viscometer are rather simple, the absolute value of the sample viscosity can be determined, at least to the right order, without any calibration using standard reference materials.
The reproducibility of the results was demonstrated by the measurements, covering the glassy to molten states, of polymer glasses up to 500 K, and of boron oxide and pyrex glasses up to 1270 and 1470 K, respectively.

View full abstract

The sludge produced from the pickling of stainless bands settles to the bottom of the pickling tank and becomes an impediment to the operation. In this research, a very high concentrations of Cr, Ni, and Fe were found in the sludge. In particular, the concentration of Cr was more than 40 mass%. It was also found that the particle size of the sludge is as small as 9.2 μm. The diameter increases by coagulation and the sedimentation velocity of the sludge is 1.4 m/h.
Based on these results, a study was made on an ejector capable of preventing the sludge precipitation in the pickling tank and moving the sludge, and a sedimentation tank where the sludge would precipitate.
First, 14 groups of ejectors were installed on the two sides of the pickling tank. Two circulation tanks connect with the pickling tank and each group of ejectors was installed in different circulation tanks. Continuous action of the ejectors causes the sludge to remain afloat and prevents it from settling out in the pickling tank. The sludge is moved to the circulation tank, which is also equipped with the same type of ejectors as the pickling tank to prevent the sludge from settling.
Next, precipitation and separation of the sludge are performed in this sedimentation tank. After separation in the sedimentation tank, the sludge was neutralized by mixing with hydrated lime using kneading device. The neutralized sludge is loaded into a lugger bag and can be used as a raw material for steel making.

View full abstract

The normal spectral emissivity of copper in solid and liquid states was measured at 1273 to 1673 K with wavelength during 400 to 850 nm. The ratio of radiation intensity to that at a reference wavelength of 650 nm was measured in order to avoid the difficulty of simultaneous measurement of the radiation intensity from a black body. The emissivity of copper increases in solid state and decreases in liquid state with increasing temperature and decreases with increasing wavelength in both states. At the melting point of 1358 K, the emissivity at 650 nm discontinuously increases from 0.123 in solid to 0.143 in liquid.

View full abstract

Zn/Fe alloy electrodeposition on a low carbon steel sheet has been examined mainly by means of scanning electron microscopy and X-ray diffraction. The results obtained are as follows:
The crystal structure and the morphology of the film deposited in a low Fe bath changes from granular η/Γ mixed phases, to pyramidal η-phase crystals and then to layered hexagonal η-phase plates with increasing current density. The initial deposits on the substrates, however, are layered hexagonal η-phase plates at all the current densities examined. The final morphology is determined by the deposits on these η-phase crystals. The c/a ratio of η-phase varies from 1.87 to 1.63 with increasing Fe concentration in the crystals. The morphological change of η-phase crystals can be explained in terms of the atomic configurations on both (0001)_η and {1100}_η planes and the c/a ratio.
In the case of a high Fe bath, the deposited film is varied from granular η/Γ to cuboidal Γ/Γ₁ mixed phases with increasing current density.

View full abstract

Low alloy Cr and CrMo steels have been studied to determine their susceptibility to hydrogen embrittlement. The steels were quenched at a constant cooling rate of 30 K s^-1 and tempered at 673 and 973 K. Hydrogen charging of steels was carried out in 0.5 M H₂SO₄ solution without and with addition of propargylic alcohol as corrosion inhibitor. In spite of the current density of 2-3 mA cm^-2 and the applied load of 40 % yield strength was found that the steels tempered at 673 K were highly sensitive to hydrogen embrittlement. The steels tempered at 973 K showed significantly higher resistance to hydrogen embrittlement. In spite of higher strength, the CrMo steel showed higher resistance to hydrogen embrittlement than the Cr steel. The fracture of steels was initiated at the second phase particles followed by transgranular fracture from iron carbide or iron-manganese sulphide inclusions. Since the inclusion distributions are similar in both steels, the enhanced resistance of CrMo steel to hydrogen embrittlement is due to finer prior austenite grain and presence of Mo₂C in the ferrite matrix. With the addition of propargylic alcohol in 0.5 M H₂SO₄ solution the resistance of steels to hydrogen embrittlement was increased and fracture surface showed mostly dimple fracture.

View full abstract

Omega method has been proposed for predicting rupture life from creep strain ε-time t data. The method is based on a linear relation between logarithm of creep rate ε and strain, namely a creep curve without the primary and secondary creep stages. Such a linear relation, however, seldom holds in real creep data. In this paper, the original equation is modified to the following form:

ε =ε₀+_Ω¹ {ln (1+ζt)-ln (1-ηt)}

where ε₀, Ω, ζ and η are parameters characterizing a creep curve. The first and second terms in the parentheses describe primary creep and tertiary creep, respectively. The equation is applied to ferritic steels. The four parameters of the equation can uniquely be determined for a creep curve with a curved ln ε-ε relation. The equation can well reproduce creep curves with a prominent primary creep stage. Because of the high reproducibility, the modified equation can predict rupture life with higher accuracy at an earlier stage of creep than the original equation.

View full abstract