This study is to clarify the flow characteristics and hold up behavior of powders in a packed bed with lateral inlets through which gas and powders are injected. The experiments were carried out for two kinds of the packed particles and powders. Longitudinal and radial distributions of powder hold up as well as longitudinal distribution of pressure in the packed bed were measured. It was found that more powders deposited in the lower part, especially in the lower central part of the packed bed and that the total hold up of powders increased with the increase in the diameter of the packed particles. No blockade occurred for powders of 98 and 175 mm under the experimental conditions.
The gas and powder two-phase flow in the packed bed was simulated by a two-dimensional mathematical model considering gravitational force, the interaction forces between gas and the packed particles, gas and powders, and powders and the packed particles. The effect of static hold up of powders on the voidage of the packed bed was reflected in computations through the trial and error approach according to the experimental data of total hold up of powders. The pressure distribution computed by the mathematical model has good agreement with the observed one, and so that the static and dynamic hold ups of powders obtained are reasonable.

View full abstract

Laboratory size (4.5-5 kg) low carbon steel melts were deoxidized using Al in zirconia crucibles. Some of these heats were treated with different types of synthetic fluxes to evaluate the oxide inclusion removal and desulfurization characteristics of the respective synthetic fluxes. The sulfur, oxygen and aluminum contents in the steel melt were determined as a function of refining time and temperature. It was observed that the reaction between aluminum and oxygen in solution closely follows thermodynamic equilibrium. Reoxidation of the steel melt occurred through oxygen pick-up from the atmosphere. The extent of this reoxidation was found to be a function of both the oxygen content and the exposed surface area of the steel melt. The effectiveness of calcium-aluminate (12CaO·7Al₂O₃) based fluxes in protecting the steel melt from reoxidation and in increasing the rate of removal of inclusions was evaluated. Synthetic fluxes and refining techniques were developed to obtain total oxygen level of 10 ppm within the first 10 min of refining. Sulfur removal followed first order reaction kinetics. The effectiveness of calcia saturated calcium-aluminate (12CaO·7Al₂O₃) based fluxes in desulfurizing the steel melts was evaluated. The effects of adding barium oxide and fluorspar to the flux, the initial sulfur level of the steel melt, and addition of tellurium to the steel melt on the kinetics of desulfurization were also studied.

View full abstract

CO₂ solubilities of Na₂O-SiO₂-B₂O₃ melts were determined by the thermogravimetric method in the overall temperature range of 1000-1250°C after correction for loss of weight due to volatilization from the melts as well as drag force of flowing gas. The melts had mostly 50 mol%. Na₂O, with SiO₂ and B₂O₃ contents varying from 0 to 50–mol%. The values of CO₂ Solubilities for 50Na₂O-50SiO₂ melts were within the ranges reported in literature. For 50 mol% Na₂O, the solubility decreased with increasing B₂O₃ content in the melt and with increase in temperature. Also linear dependence of logarithm of carbonate capacity with theoretical optical basicity were obtained at constant temperatures.

View full abstract

The effects of particle size distribution on the kinetics of hot metal desulphurization were investigated by pilot-scale injection. Three different particle size distributions of calcium carbide were injected into 70 kg heats of carbon-saturated iron. The sulphur contents and oxygen activities were measured during the injection. The reaction in the plume during the injection could be described as a first-order, diffusion-controlled reaction, after an incubation period lasting between 20 and 40 sec. This rate constant was found to increase as the particle size decreased. A kinetic analysis based on mass transfer theory was performed considering the total particle size distribution of the calcium carbide. Through this analysis a new average size, directly related to the mass transfer behaviour, was developed. Comparison of the observed and theoretical dependencies of the first-order rate constant on the powder feed rate suggests that the fraction of particles in contact with the melt decreases as the particle size decreases. Finally, issues of scale-up and economic assessment are discussed.

View full abstract

Cold model experiments were performed to make clear the critical condition causing the entrapment of slag under bottom gas injection. Entrapment of slag was judged from visual observation. The following empirical correction of the critical liquid velocity on the centerline of the vessel. u_{cl, c} was derived.
u_{cl, c}/V=1.2(ν_s/ν_m)^0.068(H_s/D)^-0.11
d_B/D<H_s/D<1/2, 0.6<ρ_s/ρ_m<1, 0.3<ν_s/ν_m<120,
45 mN/m<σ_ms<63 mN/m
u_{cl, c}=1.2 u_rP^-0.28
u_r=(gQ_{a, c}/H_m)^1/3
P={Q²_{a, c}/(gH_m⁵)}^1/5
V=(σ_msg/ρ_s)^1/4
where, g: the acceleration due to gravity,
Q_{a, c}: the critical gas flow rate for the entrapment of slag,
H_m: the thickness of metal layer,
σ_ms: the interfacial tension between slag and metal,
ρ_s', ρ_m: the densities of slag and metal,
ν_s', ν_m: the kinematic viscosities of slag and metal,
H_s: the thickness of slag layer,
D: the bath diameter,
d_B: the bubble diameter.
This correlation could predict the critical gas flow rate Q_{a, c} for previous hot model experiments as well, provided that the flow pattern in the baths were similar to the present case.

View full abstract

In order to provide the fundamental data for the removal of copper from ferrous scrap by metallic bath immersion process, the dissolution rates of stationary copper cylinder into molten Al-Cu and Mg-Cu baths were measured as a function of temperature and the composition of molten bath. Furthermore, kinetic analysis was conducted with respect to mass transfer and diffusion coefficients.
It was confirmed that both the molten baths of Al and Mg systems can dissolve solid copper very rapidly and have much greater mass transfer coefficient than the other metals such as Pb and Bi do so. The dissolution rate and mass transfer coefficient are substantially the same for the both baths and decrease with the increase of Cu concentration of molten baths. Apparent activation energy of mass transfer are in the range of 32 to 40 kJ/mol for Mg-Cu system and about 30 kJ/mol for Al-Cu system.
In the case of Al-Cu system, diffusion, coefficient is almost unchanged by varying Cu concentration of molten bath. In Mg-Cu system, however, it markedly decreases with increasing Cu concentration, particularly in the concentration range more than about 30 mass%Cu and underlies that for Al-Cu system.

View full abstract

The equilibrium constant K_Al for the reaction: 2Al+3O=Al₂O₃(s) and the first-order interaction parameter e_Al^O were estimated at 1873 K by using the values for the contents of oxygen, aluminum and other components obtained in slag-metal equilibrium experiments and those for activity of alumina, along with the respective interaction parameters. The values for log K_Al and e_Al^O were found to be 13.3±0.59 and -6.83±0.07, respectively.
Activities of slag components in the CaO-Al₂O₃-SiO₂ system were also evaluated at 1873 K from the values for nitride and sulfide capacities coupled with nitrogen and sulfur distribution ratios. These results were discussed by comparing with previous values.

View full abstract

We disintegrated an annealed SUS304 stainless steel bar into separate crystal grains, and measured the distribution of the volume-equivalent grain diameter (the three-dimensional grain size distribution) by weighing the individual grains. We also measured the distribution of the area-equivalent grain diameter (the two-dimensional grain size distribution) on a cross section of the steel bar. From this two-dimensional grain size distribution, the three-dimensional grain size distribution was calculated by the method previously proposed by the authors. The measured and the calculated three-dimensional grain size distributions were then compared. The measured three-dimensional grain size distribution was approximately log-normal. The variation coefficient of the distribution (the standard deviation divided by the average value) was 0.4, and the ratio of the maximum grain size to the average size was about 3.0. The calculated three-dimensional grain size distribution was very similar to the measured one, indicating that the method of calculation previously proposed by the authors yields accurate results.

View full abstract

Non-isothermal austenitisation kinetics of three plain carbon steels have been studied with the help of Differential Scanning Calorimeter (DSC) employing modified Kissinger's analysis. Kinetic parameters like activation energy and coefficient of Johnson-Mehl-Avrami (JMA) kinetic equation so obtained, were used to calculate the theoretical intercritical annealing time for intercritical annealing of three plain carbon steels to develop dual-phase structure. Good correlation has been obtained between theoretical and experimentally found intercritical annealing time.

View full abstract

The effect of austenitizing temperature were investigated on the microstructure and mechanical properties of 12%Cr steel. Low-temperture austenitizing below 1000°C induced the carbide coarsening during subsequent tempering at 750°C for 1 hr due to the nucleation effect of undissolved M₂₃C₆. The large and spheroidized carbides enhanced the subgrain growth. On the other hand, the complete dissolution of M₂₃C₆ above 1000°C caused the fine carbide formation on lath boundaries, which retarded the subgrain growth during tempering. Furthermore, the dissolution of Nb(C, N) above 1100°C enhanced the tempering resistance through increasing the stability of lath morphology and reducing the growth rate of M₂₃C₆. The increase in strength with increasing austenitizing temperature was attributed to the fine carbide distribution and the high dislocation density. Further, as the austenitizing temperature increased, the impact energy markedly reduced, due to the large proir austenite grain size and the high strength.

View full abstract

The wear resistance of an Fe-12.2wt%Cr-0.84wt%Mo-0.43wt%V-1.44wt%C alloy tool steel after cold treatment at 223K (subzero treatment) and after cryogenic treatment 93K (ultra-subzero treatment) has been investigated. The wear resistance of steels after cryogenic treatment is superior to that after cold treatment. The effects of cryogenic treatment on the microstructure were also studied by means of X-ray diffraction and transmission electron microscopy methods. Unlike cold treatment, cryogenic treatment improves the preferential precipitation of fine η-carbides instead of ε-carbides. These fine carbide particles enhance the strength and thoughness of the martensite matrix and then increase the wear resistance. The formation mechanism of fine η-carbide is discussed.

View full abstract

In this investigation a stable high manganese austenitic steel, 0.45C-17Mn-2.8Al, has been studied for hydrogen embrittlement using cathodically precharged specimens. Tensile testing of axisymmetric and plane strain specimens precharged with hydrogen show an appreciable loss of 8-10% reduction in area (RA) whereas the loss in % elongation is lesser. The true fracture strain decreased from 0.88 to 0.73 for axisymmetric and from 0.79 to 0.60 for plane strain specimens. Hydrogen precharging is observed to result in decrease of CTOD at crack initiation by about 0.07 mm and a decrease in crack tip fracture strain for crack initiation from 0.53 to 0.34. The greater effect of hydrogen precharging thus observed is attributed to existence of higher stress triaxiality in CTOD and plane strain tensile testing in comparison to axisymmetric one.
On SEM examination of fracture surfaces the uncharged tensile specimens showed only dimpled fracture, the precharged specimens showed transition from dimpled to quasicleavage and intergranular fracture near the surface. Regions close to the pre-fatigue tip in CTOD specimens depict intergranular fracture. The fractographic changes are attributed to the combined role of stress intensity and hydrogen concentration variation arising out of hydrogen transport inside the specimen.

View full abstract

The sulfide stress cracking (SSC) behavior of welded linepipes has been studied using some SSC tests, such as tensile SSC test (NACE-TM 0177-90 Method A), full thickness SSC tests, four points bending SSC tests and full scale SSC tests. Detailed metallographic examinations have been performed in order to understand the influences of steel chemistries and heat inputs in welding on the SSC resistance of the heat affected zone. The effects of the geometry of the SSC specimen and hydrogen concentration on the SSC have also been investigated. The results have indicated that homogenized bainitic ferrite microstructure, which can be attained by reducing carbon content and by applying accelerated cooling after controlled rolling in plate rolling process, improves the resistance of steel plates to the SSC. Meanwhile, most specimens from submerged arc welding welded joints fail at the heat affected zone, regardless of differences in the microstructures of base materials, and they tend to show nearly the same level of the ratio of threshold stress to yield strength in the tensile SSC test. Specimen geometry and the quantity of hydrogen concentration in steel affect on the threshold stress, and hence reducing the hydrogen concentration improves the SSC resistance of the welded joints especially in lower hardness heat affected zone. Relations between the results of the laboratory tests and full scale tests are also discussed.

View full abstract