Mixing time in a cylindrical water bath agitated by downward gas injection through a single-hole nozzle was experimentally investigated as a model study for steel refining processes. Measured values of the mixing time could not be predicted by empirical equations derived previously for bottom gas injection. The reason is that the bubble dispersion pattern caused by downward gas injection is much different from that by upward gas injection. An empirical equation for the mixing time was proposed as a function of the nozzle diameter, bath diameter, bath depth, nozzle position in the bath and gas flow rate. The measured values of the mixing time were approximated by this equation within a scatter of ±40%. This equation was also compared with the existing measured values and empirical equations to clarify the difference between upward and downward gas injection processes.
A segregation phenomenon observed when particles having different density and shape as well as particle diameter have been charged by mixing them were examined with the use of model experimental apparatus and a discrete simulation model. In consequence of it, it has come out that the segregation phenomenon of mixed particles depends on both of the particle diameter ratio and density ratio with a base particle and dependency can be estimated with a theoretical model.
Heat transfer in a preheating process of steel scraps was analyzed by taking convection combined with radiation into consideration. The analysis of simultaneous heat transfer has been carried out by experimental and mathematical methods. To examine the accuracy of calculation, the experiment was carried out in He, N2 and Ar atmosphere under reduced, ambient and high pressure. The commercial FEM software ANSYS was used for simulation. The simulated results agreed well to the experimental ones with the specimen surface emissivities of 0.4. The convective heat transfer coefficient was estimated. The coefficient increased with increase in pressure and with decrease in temperature. The ratio of heat quantity between overall and convective heat -Ψ, is varied with the crucible temperature, and can be expressed by the following correlation:Ψ=βeγT. The heat transfer coefficient can be summarized in the following dimensionless correlation: Nuξ=2.62 Reξ0.15
The experiments were performed using a small-scale furnace in order to improve the reduction rate of chromium oxide in Cr2O3 containing slag by carbon in steel. As a result of the investigation of the experiments by SOLGASMIX, the following points were clarified: 1) The reduction rate of chromium oxide in slag was improved by lowering the basicity of the slag or by adding Al2O3 or CaF2. This improvement was caused by the increase of liquid phase fraction of the slag. 2) The liquid fraction of the slag was related to the increase of the mass transfer coefficient and/or the effective reaction area of reaction. 3) The dissolution of MgO was accelerated greatly by lowering the basicity or by adding CaF2. This phenomenon was caused by the increase of MgO solubility in the slag. 4) The optimum composition of slag is (Al2O3)=15% and CaO/SiO2=2.5 from the viewpoint of the compatibility of the improvement of reduction rate of chromium oxide and the prevention of the dissolution of the refractories.
In order to widen the scope of laser ablation/ICP-MS, calibration curves were prepared by using aqueous standard solutions instead of using standard solid materials. The decontaminated sample was irradiated by a Q-switched Nd/YAG laser beam (150mJ, 1064nm) for 220s and the ablated sample was trapped in 10ml of 0.1 moll-1 nitric acid. More than 30% of the ablated sample were recovered in the solution as tiny particulates. The solution was directly introduced into the ICP-torch through a pneumatic nebulizer for the analysis by MS. Because the ionization efficiency of tiny particles was almost the same as that of aqueous ions, a series of aqueous standard solutions was employed for the preparation of calibration curves. Important experimental factors (e.g., argon flow rate, acidity of solution, and tube length) were studied in detail and optimized to obtain the highest signal intensity. The accuracy of the analytical data was significantly improved when the matrix element such as iron or zirconium was used as internal standard. The proposed method was applied to the determination of Cr, Ni, Mo, and Co in reference materials for steel and zircaloy. For the latter sample, laser irradiation was allowed to continue for 30min to ablate the sufficient amount of particles. The analytical results were in good agreement with the certified values.
The distribution of hydrogen occluded in high strength bolts tightened beyond the yield strength and exposed for 9 years has been investigated to establish an evaluation method for delayed fracture property. Hydrogen thermal desorption analysis was employed for determining hydrogen content in specimens cut out from the exposed bolts. Diffusive hydrogen causing delayed fracture is not uniform in concentration within the bolt and enriched in the threaded portion where diffusive hydrogen concentration was about three times that in the body portion. It is reasonable that diffusive hydrogen intrudes according to corrosion reaction, but no clear correlation between rusting and hydrogen concentration was recognized. From evaluation of the role of stress and plastic strain on hydrogen concentration, it is concluded that high concentration of diffusive hydrogen in the threaded portion was mainly caused by plastic straining.
Fatigue crack initiation and growth behavior of stainless steels with different volume fraction of ferritic (α)-austenitic (γ) phases were studied. Materials with five different volume fractions of α/γ phases, i.e. SUS304 (100%γ), SUS329J4L (50%γ, 28%γ, 12%γ), and SUS444 (100%α), were prepared. Fatigue tests were conducted in laboratory air using electro-hydraulic fatigue testing machine. Cracks initiated pre-dominantly from slip bands within γ grains for SUS304, and SUS329J4L, and within α grains for SUS444, i.e. transgranular crack initiation was found. Fatigue lives decreased in the order of SUS329J4L, SUS444 and SUS304. The same tendency was also seen in fatigue limits. In early crack growth region, while no remarkable differences were found in all materials, for three materials in SUS329J4L crack growth rates were faster in the order of 50%γ, 28%γ, and 12%γ, i.e. early crack growth was influenced by the phase boundary. Moreover, the early crack growth was enhanced in comparison with the long crack in all materials. After allowing for crack closure, long cracks exhibited a similar growth behavior in all materials. Based on these results, the transition from small to large crack and the effect of α/γ phase fraction were discussed.
Ultrafine-grained structure formed dynamically through severe plastic deformation at elevated temperatures has been investigated in a 0.1% C-0.4% Si-1.5% Mn steel. The effects of the strain, the strain rate and the deformation temperature on the microstructural evolution were examined using a compression technique with a pair of anvils. The maximum strain was 4, the deformation temperature was below the Ac1 temperature, and the Zener-Hollomon parameter (Z) ranged between 1012s-1 and 1016s-1. The ultrafine ferrite grains surrounded by the high angle boundaries evolved when the equivalent plastic strain exceeded the critical value of about 1. The number of newly evolved ultrafine grains increased with the strain; however, the average sizes did not depend on the strain. The grain size, d, depended on the strain rate and the deformation temperature. An equation, d(μm)=102.07Z-0.16, was obtained experimentally. The details of the phenomena are compared with the dynamic discontinuous recrystallization in ferritic steels.
It has been reported that an addition of boron increases total elongation of continuously annealed Al-killed steel sheet and that the addition of boron is sometimes utilized for improving the formability. The purpose of this study is to investigate the influence of boron content on workhardening behavior of Al-killed steel sheet. Stress-train curves of low carbon steel depended on annealing temperature, over-aging temperature and boron content. In low boron steel, the low temperature annealing and low temperature over-aging resulted in the increase of the stress required to continue the plastic deformation in tensile test since the low temperature annealing and over-aging promoted the grain refinement and the fine carbide precipitation in matrix, respectively. In high boron steel, on the other hand, there were no influences of both the annealing temperature and over-aging temperature on the stress-train curves. The independence of the annealing temperature and over-aging temperature may be attributed to strengthening of ferrite grain boundary and the fine carbides in matrix which were promoted by the large amount of boron, respectively. The n-value of low boron steel increased significantly and showed a peak at lower strain. On contrast, the n-value of high boron steel increased slightly at lower strains and did not show a peak. In high boron steel, dynamic recovery would be activated more remarkably than in low boron steel since dislocation density in the early plastic deformation region in the tensile test may become higher by ferrite grain boundary strengthening which is due to boron segregation.