The interfacial rate constant of CO₂–CO reaction with the solid iron and iron oxide has been measured by means of the isotope exchange reaction at 1273 K. Changing the total flowrate, the composition of CO₂/CO of gas mixture and the reaction area, the effects of experimental conditions on rate constant were investigated. Varying the CO₂/CO ratio in the range of 0.1 to 10, the dependence of rate constant on CO₂/CO ratio was studied. It is found that the relationship of the rate constant and CO₂/CO ratio can be divided to three regions represented iron, wüstite and magnetite phase field, respectively. In three phase regions, the slopes of the linear relationship of logk_c and log(CO₂/CO) were about –0.48, –1.14 and –1.71, respectively. The microstructure of reaction surface of sample was also observed by SEM, with the CO₂/CO ratio, the reacted surface became less smooth and showed the crystal morphology.

View full abstract

The permittivity and permeability measurements of ferric oxide (Fe₂O₃) were carried out over a broad temperature range from 24°C to above 1000°C at 915 and 2450 MHz. The real part and imaginary part of complex relative permittivity (ε_r′ and ε_r″) of ferric oxide slightly increase with temperature below 450°C, above which ε_r′ increases significantly while ε_r″ presents a broad dielectric loss peak between 450 and 1000°C. Contrary to ε_r′ and ε_r″, the real part and imaginary part of complex relative permeability (μ_r′ and μ_r″) remain relatively invariable (1 and 0, respectively) until 700°C. The μ_r′ values subsequently exhibit a decreasing tendency due to the increased electrical conductivity at higher temperatures while the μ_r″ values stay negligible as temperature increases. The results demonstrate that the dielectric loss is the primary factor contributing to microwave absorption of Fe₂O₃. The calculation of microwave penetration depth shows that Fe₂O₃ undergoes a transition from a microwave transparent material to a good microwave absorber with increasing temperature.

View full abstract

The solution model of Mn–Fe–Si–C system is essential in the development of high Mn containing steels and alloys, and its effectiveness depends on the well established information of constituting lower order systems. The solution property of Mn–Si system, which is one of such constituting binary systems, is not well established yet because the experimental data are limited in number and scope. The present study determines the temperature independent partial enthalpy and excess entropy of Mn and Si from the relevant experimental data. The resulting excess Gibbs free energy function was found to adequately describe the solution properties of Mn–Si system.

View full abstract

The development of a new viscometer based on rotating crucible method and the viscosity measurement of SiO₂–CaO–CaF₂ system were carried out. The viscometer was hermetically closed, and the atmosphere around the melts was highly controllable. In the viscosity measurement, an inner cylinder made of graphite was immersed into the melts in a graphite crucible under Ar at elevated temperatures (1467–1782 K). The torque given to the inner cylinder by rotating the crucible was measured by means of a supersensitive torque sensor, and the viscosity was determined. The minimum limit of the viscosity determined was 20 mPa·s. A special attention was paid to keep the temperature uniformity inside the furnace for precise measurement by placing tungsten plates as the thermal shield, and the excellent temperature uniformity (±0.2 K in a whole crucible) was obtained. The basicity (= C_CaO/C_SiO2 in mass) of the melts was 0.79 or 1.25, and the concentrations of CaF₂ were 5, 10, 15, 20, 25 mass%. As the results, the viscosity of the melts showed a good Arrhenian type linearity in any samples. The loss of fluoride from the melts after the measurement was very small which suggests the consistency of melt composition. The viscosity of the melt decreased with increasing the concentration of CaF₂, but the degree of decrease was smaller than literature values. The decrease in the viscosity of the melts with the addition of CaF₂ in acid region is slightly greater than that in basic region.

View full abstract

Iron ore sintering involves the movement of a flame front down a particulate bed, and a series of physico-chemical reactions over a large temperature range. In the literature simple and more sophisticated iron ore sintering models have been reported. In this paper a more comprehensive numerical model which incorporates most of the significant processes and heat transfer modes proposed in earlier models is given. Therefore, sub-models are available to describe the relationship between airflow rate through the bed and flame front speed, the evaporation and condensation of water ahead of the front, the calcination of fluxes nearer to the front, the reactions that occur in the front and cooling of the bed with the departure of the front. Improvements were made to several areas – such as coke combustion, and the melting and solidification processes – to more accurately quantify the phenomena involved. More recent progress in understanding the fundamentals of sintering from BHP Billiton studies have also been incorporated into the model. To date, twelve sinter pot tests have been used for validation studies. Reasonably good agreement was obtained between predicted and measured results – in areas such as bed temperature profiles and waste gas temperature and compositions. Work is continuing to further improve the model, and broaden the validation work to include other bed temperature profile parameters.

View full abstract

Discrete element method (DEM) was applied to charging process in iron ore sintering and vertical size segregation in the feed bed was anlyized, where a particular charactor of sintering feed, plastisity and stikiness, was expressed by adjusting the rolling friction coefficient to fit a result from a preliminary test on sintering feed flowing out of a box. DEM simulations and comparison with plant data gave the following findings:
1) When the rolling coefficient was 4.0, the calculation resembled the experiment in terms of the angle and the spread of feed after resting.
2) The simulation for a charging apparatus recreated vertical size segregation in feed bed in good agreement with plant data.
3) The simulation showed that the size segregation increased with decreasing chute angle, the change of which was explained with the fact that the horizontal velosity of particle at the bottom end of the chute also increased unless the feed made a build-up on the chute.

View full abstract

The liquid flow in the packed bed strongly affects the productivity and operational stability of the blast furnace. It is controlled by the physical properties and wettability of the melt and the structure of the packed bed. However, this is a difficult object of analysis, as melt properties and conditions in the furnace are inhomogeneous. In this research, a three-dimensional model of the liquid flow in a packed bed was constructed using the MPS method, which is one of particle method, in order to express melt behavior. The liquid flow on solid surfaces was expressed by optimizing the relationship of energy at a three-phase (solid, gas, liquid) interface. The appropriateness of the calculation model was verified by analyzing and comparing the behaviors of liquids with different physical properties flowing on the surface of a single sphere experimentally and in calculations. Next, this model was applied to analysis of the flow of liquids dripping in a packed bed, and a simulation was calculated for a three-dimensional liquid flow comprising dispersion and coalescence of the liquid. Among the knowledge obtained in this study, the viscosity of a liquid migrating through a packed bed influences the velocity and flow path of the liquid, but has little effect on the amount of liquid which remains in a static condition in the packed bed.

View full abstract

For green production of iron ore sintering, it is significant to substitute fossil fuels by biomass which is a kind of clean and renewable energy. In this paper, three kinds of biomass fuels such as charcoal, charred-straw and molded-sawdust were studied as sintering fuels. The results show that, with the proportion of biomass replacing coke breeze increasing, the vertical sintering speed raises, but the yield and the tumble index of sinter decrease, so the replacement proportion should be appropriate for satisfying the productivity and the quality of sinter. The suitable replacing proportions of charcoal, charred-straw and molded-sawdust are 40%, 20% and 15% respectively, in which the emission of COx can be decreased by 18.65%, 7.19% and 5.39%, SOx by 38.15%, 31.79% and 28.90%, NOx by 26.76%, 18.31% and 15.49% respectively.

View full abstract

The integrated process, which combined the carbothermic reduction assisted with microwave irradiation followed by milling and magnetic separation, was investigated as a potential approach for dephosphorization treatment of iron ores bearing high P. Mineralogical Research indicated that the phosphorous in oolitic hematite ores exists in two forms, inter-grew with iron oxide or gather together into a bigger size as gangues. The increase of microwave irradiation time had substantial effect on the reduction of Fe while the influence of carbon dosage was not obvious. As the reduced iron and unreduced P-contained gangues were not efficiently separated, a high iron recovery ratio and a reasonable dephosphorization rate were incompatible under the conditions investigated in this study. This process was successful in the aspect of iron recovery but the dephosphorization rate was unsatisfactory. Microwave-assisted reduction could be a promising pre-reduction process which could be followed by smelting process and dephosphorization treatment.

View full abstract

Through analyzing the factors that influence end-point manganese content during BOF steelmaking process, multiple linear regression model for prediction of end-point manganese content was obtained on the basis of actual production data. Given the advantages of artificial neural network, it was used to predict end-point manganese content during BOF steelmaking process, and BP neural network model was established. By means of combining the characteristics of genetic algorithm and BP neural network completely, a combined GA-BP neural network model was established. The verification and comparison of the above three models show that the combined GA-BP neural network model has the highest prediction accuracy. The hit rate of the combined GA-BP neural network model is 90% and 84% respectively when predictive errors of the model are within ±0.03% and ±0.025%. Compared with two models aboved, the combined GA-BP neural network model could provide the most accurate prediction of end-point manganese content, and thus represents a good reference for real production.

View full abstract

The control of molten steel temperature is one of the important parameters for producing the superior quality of steel. Ladle is one of the major vessels to be monitored for controlling this temperature. A computational fluid dynamics (CFD) based mathematical model was developed for temperature prediction of ladles and molten steel. The model was validated with data collected from the plant and maximum 4% deviation between predicted and measured data was noticed. The effect of various parameters on temperature drop of molten steel was studied. The results obtained show low impact of slag thickness, tapping temperature and ladle life. However, the importance of initial refractory temperature was noticed. The CFD model developed in present work can generate the history of temperature drop of molten steel and refractory walls of any ladle at any stage. The model can be utilized to predict the suitable tapping temperature.

View full abstract

Solubility of nitrogen in liquid Fe–Cr–Ni–Mo stainless steel was measured by sampling method in the temperature range from 1723 K to 1923 K under a N₂ gas of 1 atm. As a result, the solubility increased with increasing Cr, Mo and Mn content, whilst it decreased with increasing Ni content. In addition, the solubility increased with decreasing temperature. Thermodynamic analysis was carried out taking Fe-20 mass% Cr alloy as a solvent referring to the manner proposed by Anson. The first attempt showed that the calculated solubility, applying the reported first-order interaction coefficients of N against Cr, Ni, Mo and Mn, did not agree with the measured values. This inconsistency was found to be attributed to the interaction coefficient of N against Ni (e^Ni_{N Fe20Cr}). Therefore, reassessment to derive e^Ni_{N Fe20Cr} available up to 30 mass% Ni has been made with the data obtained by us and Anson. Thereby the corresponding coefficient was derived as 0.0063 with which the solubility could be well predicted.

View full abstract

A submerged entry nozzle (SEN) for thin slab casters operating at casting speeds as high as 7.5 m/minute was developed based on fundamental grounds of boundary-layer theory and water modeling experiments. Experimental techniques included tracer injection to observe overall fluid flow patterns, high speed video camera and image analysis to follow dynamic changes of meniscus levels and particle image velocimetry to measure water speeds in the mold. This design was compared with two other SEN designs of nozzles under current commercial use at various thin slab casters. Direct comparisons of mathematical simulations and experimental results among the three SEN's evidenced that this new SEN yields very stable flows which are independent from casting speed and nozzle immersion depth. Fluid flow developed by this SEN consists of a double roll pattern without generation of superficial vortices. The two other SEN's yield instable discharging jets due to and excessive shearing effects with the surrounding fluid inducing severe dissipation of kinetic energy which promotes severe tailing effects inducing strong meniscus oscillations. The proposed design has reported good industrial performances and a longer operating life because the slag protection belt suffers less wear thanks to smaller velocities of the bath in contact with the SEN wall.

View full abstract

An advanced detection method based on a modulation technique is described in radio-frequency-powered glow discharge plasma optical emission spectrometry (r.f. GD-OES). A frequency-sensitive separation using a fast Fourier transform (FFT) analyser, where a pulsated bias-current was introduced into an r.f. GD plasma, was available for improving the limit of determination for the atomic emission analysis. The FFT analyser has an ability to disperse signal components by frequency, and it is thus employed to select the component of a particular frequency. A dc bias current introduced into the GD plasma can enhance the emission intensities of analyte species greatly, and furthermore, it can be easily pulsated to modulate the emission intensities from the plasma. The modulated emission signal was selectively detected with the FFT analyser, with removing any noise components from the overall signal. The duty ratio of the pulsed bias current largely affected the amplitude of the FFT frequency components, because the pulse waveform comprised sine-function components having frequencies integral-times as much as the fundamental frequency, whose contribution coefficients depended on the duty ratio. This detection method was applied to the determination of vanadium and molybdenum in low-alloyed steel samples. The detection limits were obtained to be 6.2 × 10^–3 mass% V and 2.0 × 10^–3 mass% Mo in low-alloyed steel samples.

View full abstract

A two-dimensional on-line redox derivatization HPLC system was developed where “heart-cutting” chromatography, in conjunction with on-line redox derivatizaion, was used to isolate specific analytes in complex matrix samples. Cobalt in a stainless steel sample was used as a model analyte to evaluate the performance of the HPLC system developed. We adopted a small column packed with porous graphitic carbon (PGC) treated with hydrogen peroxide as an oxidation derivatization unit and two C₁₈ silica columns which had adsorbed hexadecyltrimethylammonium ion as the first and second separation columns. After complexation of metal components with ethylenediaminetetraacetate (edta), the sample was directly submitted to the analysis by the HPLC system without any preseparation steps. In the first dimension separation cobalt was eluted as Co(II)-edta and was separated from trivalent metal complexes such as Fe(III) and Cr(III), while it was selectively oxidized to Co(III)-edta in the PGC column and then completely separated from the divalent metal complexes such as Ni(II) on the second column. A successful application of this method to accurate and precise determination of trace amount of cobalt in a stainless steel is demonstrated.

View full abstract

The quantitative structural information such as the relative abundance of silicate discrete anions (Qⁿ units) and the concentration of three types of oxygens, viz. free–, bridging– and nonbridging oxygen can be obtained from micro–Raman spectra of the quenched CaO–SiO₂–MnO glass samples. Various transport properties such as viscosity, density, and electrical conductivity can be expected as a simple linear function of ‘ln (Q³/Q²),’ indicating that these thermophysical properties are strongly dependent on a degree of polymerization of silicate melts in the composition of silica content greater than about 30 mol%. The present methodology is believed to be widely extended to various slag systems in ferrous and non–ferrous metallurgical communities.

View full abstract

CVC (Continuous Variable Crown) work roll contour technology since its invention has been applied in more than 150 hot rolling mills worldwide for controlling strip shape. However, serious uneven wear on backup rolls and even roll spalling problems can occur due to contact pressure concentrations that exist between work and backup rolls when either cylindrical or CVC backup rolls are configured with CVC work rolls. To prevent these defects, an optimized backup roll contour was designed by combining VCR (Variable Crown Backup Roll) and CVC backup roll contours. This solution features benefits of the VCR design – reducing roll stack deflection and equalizing contact pressure distribution between work and backup rolls – as well as increasing the control capability of work roll bending forces. Implementing the optimized backup roll contour eliminated non-uniform roll wear and spalling, and strip profile was also improved.

View full abstract

Blistering occurs when oxide scale swells during oxidation at high temperatures. Blistered scale causes surface defects when rolled. The present study investigated the effect of nitrogen on blister growth when steel is oxidized at high temperatures, and drew the following conclusions. Atmospheric conditions before oxidation affect blister growth. Blisters nucleate but do not grow, when a steel sample is held in Ar gas or in vacuum before oxidation. Blisters inflate when a steel sample is held in N₂ gas before oxidation. The gas inside the grown blisters is mainly N₂ gas. The steel surface is nitrided in N₂ gas at high temperatures. It is deduced that the steel surface is nitrided before oxidation, and the nitrogen component causes blister growth upon its release as N₂ gas at the scale/steel interface.

View full abstract

In this experimental work, several single and double chemical color etching techniques have been developed to identify in details the as cast optical microstructural observation in a low alloy as cast hardening white iron. The etching techniques are based on the various metallographic reagents involving 2%nital, Marshall's, Glyceregia, 5%ammonium persulfate, Kalling's NO.1 and aqueous ferric chloride chemical solutions. Evidence is presented which indicates that the microstructural components including retained austenite, martensite, pearlite, secondary and M₃C eutectic carbides can be distinguishable colorful from each other in the multiphase white cast iron microstructure. The results indicate that the retained austenite has been appeared in a good contrasting resolution by different colors such as brilliant green and brilliant white after using double etching procedures consisting of Marshall's followed by Kalling's NO.1 and Marshall's with the subsequent use of 5%ammonium persulfate respectively, while the martensite is stained by lightly brown color after using these associated double metallographies. The martensite has been also colored by sharply contrasted brilliant brown after using a single etching technique of 5%ammonium persulfate. The secondary carbide has been stained in a good contrasting observation as dark and bright particles with double etching techniques based on the Marshall's followed by Glyceregia and Marshall's with aqueous ferric chloride solutions, respectively. The X-ray diffraction (XRD) analysis and scanning electron microscopy (SEM) micrographs are confirmed the as cast optical microstructural evolution developed in the low alloy white iron.

View full abstract

Ternary β-type Ti–10Cr–(V, Fe, Mo) alloys with self-tunable Young's moduli were subjected to solution treatment and cold rolling, and their microstructures and mechanical properties were investigated. During cold rolling, a band-like structure, which is considered to be {332}_β<113>_β mechanical twin, and deformation-induced ω phase are formed in alloys with certain chemical compositions. The number of bands increases with an increase in the cold-rolling reduction ratio and V content as well as with a decrease in Mo content. On the other hand, the Young's modulus increases during cold rolling, and the increase in Young's modulus is considered to be caused by the deformation-induced ω phase transformation. Furthermore, the tensile strength decreases slightly and the elongation tends to increase with an increase in the alloying element contents, while the effect of the V and Mo contents on the trend in changing the number of mechanical twin is opposite. These tensile properties are derived from the complicated factors among the plastic deformation mode, the type of mechanical twinning, and deformation-induced ω phase transformation, depending on the β stability and the kind of alloying element.

View full abstract

This paper presents an analytical model to describe the precipitation kinetics during isothermal holding following high temperature deformation in Ti micro-alloyed steels. This model was based on the classical nucleation and growth theory (CNGT) and Avrami equation. Using this model, the precipitation-time-temperature (PTT) diagrams for the kinetics of precipitation were easily obtained, and results show a good agreement between the experimental observation and the predictions of the model in terms of the characteristics of the PTT curves, including the shape and nose temperature and chemical composition dependence.

View full abstract

The effect of H on the mechanical properties of a high Mn twinning-induced plasticity steel was evaluated by comparing the properties of H-free and H-charged tensile specimens deformed at different temperatures prior to H-charging. The H-charging resulted in a reduction of the total elongation. The fracture surface of the tested samples revealed an inhomogeneous fracture appearance with an intergranular fracture in the range of 10 μm to 20 μm below the sample surface. The observations strongly suggest that it is this intergranular fracture in the surface layer which is the cause of the reduction of elongation in H-charged TWIP steel.

View full abstract

Determination of the mobilities for ordering transformation in Fe₃Al has been attempted by comparing the rates of experimentally observed antiphase boundary (APB) migration with that simulated by a phase filed method (PFM). The simulated boundary mobility of D0₃-APB was sensitive to the ordering mobility. This allowed us to determine the ordering mobility for D0₃-LRO at 673 K to be 3 × 10^–11 m³·J^–1·s^–1. However, the simulated boundary mobility of B2-APB was insensitive to the value of ordering mobility, and therefore the ordering mobility for B2-LRO could not be determined. The difference is ascribed to solute-drag which was quite significant at B2-APB.

View full abstract

The fractal dimension of the fracture surface (D) was estimated in the different length scales of the fractal analysis, which were associated with the size of characteristic microstructures on the fracture surface, using thin plate specimens of stainless steels fatigued by repeated bending. The fracture surfaces were produced by ductile fatigue fracture in the 21Cr and SUS316 steels and by quasi-cleavage fracture in the SUS631 steel. The values of D evaluated on small regions of the fracture surface were displayed by the fractal dimension map (color-coded map, FDM) for the analysis of the fatigue fracture process. The fractal dimension represented not only the microstructural patterns on the fracture surface but also the damage caused by cyclic compressive loading during fatigue, irrespective of fracture mechanisms of steels. The fatigue fracture processes in the stainless steels were simulated using a set of the FDMs, in which the imaging conditions were appropriately chosen. The result of simulation using the FDMs was reasonably correlated to the fatigue crack shape detected by the heat tinting method in the 21Cr and SUS631 steels. The present fractal analysis is essentially applicable to the fracture surfaces of various types of fatigued specimens subjected to cyclic compressive loading.

View full abstract

Hydrogen trap states by TiC and V₄C₃ precipitates in bcc-Fe are investigated by numerical calculations. The trap states at interstitial site and, carbon vacancy in metal carbide and bcc-Fe/metal-carbide interface were studied by ab-initio calculation. The calculated trap energies of these sites for TiC compared with the energy at interstitial site in bcc-Fe were respectively –58 kJ/mol, 125 kJ/mol and 48 kJ/mol and those for V₄C₃ were respectively –106 kJ/mol, 116 kJ/mol and –6 kJ/mol. The activation energy of detrapping from an isolated carbon vacancy is estimated at 183 kJ/mol for TiC and at 222 kJ/mol for V₄C₃ from the difference of the calculated energy at carbon vacancy and that at interstitial site in metal carbide. Hydrogen trap energy in coherent strain field around of TiC and V₄C₃ coherent precipitates in bcc-Fe are also calculated by Finite Element Method (FEM). The calculated energies are respectively less than 29 kJ/mol and less than 15 kJ/mol. These results indicate the main trap site of TiC is TiC/bcc-Fe interface, because TiC contains few carbon vacancies and has large activation energy of detrapping at the sites. That of V₄C₃ is carbon vacancies because V₄C₃ contains abundant carbon vacancies and the activation energy of migration between the neighbored carbon vacancy sites is expected to be lower than the calculated value . The estimated main trap sites of TiC is in good agreement with 3 Dimensional Atom Prove (3D-AP) observation results which reported that hydrogen atoms observed at TiC/bcc-Fe interface of TiC precipitate in bcc-Fe.

View full abstract

TiB₂-based cermets with various Co contents were fabricated from elemental powders “in situ” by means of the Self-propagating High-temperature Synthesis, SHS, and Pseudo- Hot Isostatic Pressing, P-HIP method. The sample pressed into a cylindrical compact was ignited in a steel can by an external heating element coiling the can. After SHS initiation, which was detected by rapid temperature increase, the samples were quickly pressed pseudo-isostatically under a pressure of 192 MPa and held for 5 min. Samples with predominant concentration of TiB₂, which varied from 70 to 85 vol.% with the addition of 5 vol% of Ti, were investigated in this study. Appreciable differences in terms of microstructure, density and hardness were observed depending on the composition. The average TiB₂ grain size increased while porosity decreased with rising concentration of TiB₂. The material synthesized with increased to 85% concentration of superhard TiB₂ grains and minimized concentration of Co exhibited greatest densification, highest hardness of about 2400 HV, and the most homogenous microstructure. The reaction mechanism was reportedly proposed, based on temperature monitoring during combustion and previously reported references.

View full abstract

The decomposition of simple chlorinated organic compounds comprising benzene derivatives (mono, di and tri-chlorobenzenes) as the imitation substance of polychlorobiphenyls (PCBs) was investigated by using basic molten salts in order to establish a safe, simple and highly efficient decomposition process for PCBs. Chlorobenzen liquid or the solution was injected into basic molten salts (KOH–K₂CO₃ or NaOH–Na₂CO₃) at 573–973 K with imitation air, and the residual concentration of chlorobenzenes in the exhaust was measured after the decomposition treatment. The decomposition efficiency reached to very high of 99.9999% in an optimum condition. The decomposition efficiencies of chlorobenzenes were affected by the flow rate of the air and the initial concentration of chlorobenzen in the gas mixture. The complete destruction for benzene derivatives required excess oxygen condition.

View full abstract