Reduction of MnO from slag and wetting of carbonaceous materials were studied in reaction of synthetic and industrial ferromanganese slags with graphite and coke substrates by the sessile drop method at 1450–1550°C. Reduced metal was found at slag-substrate interface and at gas–slag-substrate boundary of slag perimeter. The reduced metal was mainly manganese with small amount of dissolved silicon and carbon. The reduction rate of MnO increased with increasing activity of MnO in slag and temperature. The rate of MnO reduction from industrial slag was faster than from the synthetic slag; slag was more reactive with coke substrate than with graphite substrate. The reduction rate was observed to increase with increasing ash content in the substrate. The dynamic contact angle between carbon substrate and slag varied in a range of 80° to 140° and it decreased in the process of reaction. Lower contact angle was observed for a substrate with higher ash content.

View full abstract

The phase development in the course of carbothermal reduction of ilmentie concentrates and synthetic rutile was studied in temperature programmed reduction (350–1600°C) and isothermal reduction at 1100°C in hydrogen and 1300°C in argon and helium. Ilmenites and synthetic rutile were reduced in a tube reactor with continuously flowing gas. The rate and extent of reduction were monitored by online off-gas analysis. Samples reduced to different extent were subjected to XRD and SEM/BSE analyses. Pseudorutile and ilmenite were the main phases in ilmenite concentrates; rutile was the main phase in synthetic rutile. The phase changes in the course of reduction followed the same sequence in both hydrogen and inert gases. Pseudorutile was converted to ilmenite and titania; iron oxides in ilmenite were quickly reduced to metallic iron. Titania was reduced to titanium suboxides and further to titanium oxycarbide. Reduction of ilmenites and synthetic rutile in hydrogen was much faster than in inert atmosphere. The rate of conversion of titanium oxides to oxycarbide was affected by iron content in the ilmenites. The reduction of lower grade ilmenite with high iron content was faster in hydrogen; but slower in an inert gas. The latter was attributed to the higher porosity of highly weathered ilmenite and synthetic rutile.

View full abstract

In the processes with two interacting immiscible liquids, the overall rate of the inter-phase reactions is typically controlled by chemical reaction at the interface, mass transport within each phase, or combination of both. Under these conditions, the overall rate of the reaction between the two phases is closely related to the behaviour of their interface, the interfacial area in particular. In this work, aiming to study emulsification and the associated interfacial area increase in bottom blown metallurgical systems, water modeling was used. For two different water–oil systems, the droplet size distribution, water–oil interfacial area increase, and their dependence on gas flow rate and physical properties of the liquids were studied. The total interfacial energy and the fractions dissipated in forms of surface and potential energy were calculated.

View full abstract

The mixing time in a side-blown converter was studied using physical modelling. Water was used to simulate steel and a KCl tracer was added during the experiments. Thereafter, the mixing time was determined experimentally by measuring the electrical conductivity in the water bath. Experiments were done for two bath diameters of 200 mm and 300 mm, respectively. Furthermore, for gas flow rates between 30 cm³/s and 800 cm³/s as well as bath heights ranging from 106 to 314 mm. The mixing times were also calculated based on an expression involving the Strouhal and Reynolds numbers. The experimentally determined mixing times were found to be within ±20% of the theoretical values, which is considered to be good in physical modelling. Overall, the mixing time was found to be influenced by the gas flow rate and the vessel diameter, but not by the bath height.

View full abstract

A mixture of MgO and graphite with a molar ratio (1 : 1) was subjected to planetary ball milling for 1, 2, 4, and 8 h with the intention of enhancing carbothermic reduction reaction of MgO during subsequent thermal treatment. Unmilled and milled mixtures were characterized using a combination of X-ray diffraction (XRD) analysis, Raman spectroscopy, scanning electron microscopy (SEM), surface area analysis (SSA), and thermogravimetric analysis (TGA). The reduction reaction of milled samples occurred at lower temperature than that of the unmilled sample. A longer milling time engenders a lower reaction temperature. These results are attributable to the increased interfacial area of the sample mixture, nanometer-size refinement of MgO crystallites and amorphization of graphite obtained using the milling process.

View full abstract

Lead, considered as a tramp element in steel, was successfully removed from a 50% Fe–50% Pb mixture by evaporation in argon at 1448 K. The progress of the removal shows that it takes 5 h to completely remove the lead from the mixture, the nonvolatile iron (solid) being left behind in the crucible (alumina). Results point toward a diffusion-controlled evaporation, and the system, which closely simulates Stefan's tube for studying evaporation of a liquid through a stagnant gas film, allowed the measurement of the interdiffusivity of lead and argon (D_Pb–Ar). Since this study dealt with a pseudosteady state diffusion, rather than the steady state one for which the diffusion flux equation is well established, the required pseudosteady state equation was first developed, and, subsequently, utilized with the help of the rate data to find the D_Pb–Ar at 1448 K. The measured value, which is 1.766(±0.378) cm²/s, compares very well with the estimated value of 1.774 cm²/s.

View full abstract

It is known that the molten iron oxide reduction by Fe–C melt shows quite different behavior depending on whether the surface of Fe–C melts is fully covered with molten iron oxide or not. There have been only two studies of the molten iron oxide reduction by Fe–C melt with the existence of free surface. In both studies, the reduction appears to be controlled by the reaction of C+O→CO (g). The rate of this reaction has not been fully established. In the present study, based on the reexamination of overall reaction rates of previous two studies by Dancy and Lloyd et al., the forward and backward reaction rate constants of the reaction C+O→CO (g) at 1873 K have been evaluated by applying the order of magnitude evaluation method. The forward rate constant, in the unit of mol/m² s, is given by: k=1.33×10⁸ exp(−250000/RT), and the backward rate constant, in the unit of mol/m² s, is deduced to: k_CO=2.4×10¹¹ exp(−277000/RT).

View full abstract

In this work, the feasibility of increasing lump ores proportion in blast furnace (BF for short) was investigated. The results showed: (1) Not only the physical and chemical properties, but also the metallurgical properties of lump ores, such as the reducibility, thermal decrepitation properties and softening properties, were not worse than those of the pellets, so the BF production would not be influenced greatly when the pellets were replaced by lump ores. (2) The own softening and melting properties of the lump ores (pellets) were dramatically improved by interaction between sinters and lump ores (pellets) found in the experiments, while there was no obvious interaction between lump ores and pellets, pellets and pellets, lump ores and lump ores. The interaction occurred with the contact of burden as in BF, and the main reaction product was CaFeSiO₄. Except temperature, the reaction was influenced by the chemical composition, micro-structure and contact conditions, and so on. (3) The softening and melting properties of the integrated furnace charge were improved, and the high-temperature interactivity of iron bearing materials was enhanced when the proportion of lump ores increased and the proportion of pellets decreased, for the reactivity of the lump ore was stronger than that of the pellets. (4) Furthermore, the collocation pattern of lump ores and ratio between lump ores was optimized according to interaction. When the proportion of lump ores was up to 23% at the excellent collocation pattern and appropriate ratio, the burdens still conformed to the requirements of ironmaking.

View full abstract

The optimal granulation moisture, which means the least water content added into the mixture for obtaining suitable size distribution of the granules, is of vital importance for sintering. The moisture capacity, which means the maximum water content held in the iron ores of unit mass, is suggested, and defined. An apparatus for the moisture capacity measurment was developed. Five samples were selected for the measurments. In order to study the relationship between the moisture capacity and some characters of iron ores, the surface area and pore volume was measured by the method of liquid nitrogen absorption. It was found that the moisture capacity increaseed with increasing the external surface area of the iron ore, and decreased with increasing the pore volume in the iron ores. The iron ores were granluted with various water content level in the labtotary rotaing drum, and then loaded in the sintering pot for measuring the permeability of the burden. The size distribution of the granules were also measured for asisting confirming the optimal water content in granlation. The application of moisture capacity indicated that the iron ore of high moisture capacity needed more water added in the granulation process to get high permeability.

View full abstract

It is important to understand the role of solid phases such as solid CaO or 2CaO·SiO₂ in hot metal dephosphorization. In the present study, the reaction behavior of P₂O₅ at the interface between solid 2CaO·SiO₂ and liquid CaO–SiO₂–FeO_x–P₂O₅ slags saturated with solid 5CaO·SiO₂·P₂O₅ at 1573 K was investigated.
The result shows that the solid 2CaO·SiO₂ reacts with the liquid phase in the two phase mixture to form the P₂O₅ condensed phases. Compared with the results by using the homogeneous slag as reported in the previous papers, the formation of P₂O₅ condensed phases is more evident and the back dissolution of P₂O₅ condensed phases into the surrounding slag is restrained in this study. These results indicate that the dissolution of P₂O₅ condensed phases can be considerably restrained in the case that the bulk slag is saturated with solid 5CaO·SiO₂·P₂O₅ before reaction.

View full abstract

In the current paper, a kinetic model was developed to study the entrapment of inclusions in the molten steel flowing through a Submerged Entry Nozzle (SEN) during billet continuous casting process. The trajectory of inclusions was calculated by considering the drag force, lift force and gravitational force. The entrapment locations of inclusions on SEN wall were predicted. The effects of nozzle diameter, casting speed, billet dimension, and inclusion diameter on SEN clogging were quantitatively discussed. The results indicate the inclusions with diameter larger than 100 μm are not able to be entrapped by the nozzle wall; and the entrapment probability will increase quickly with decreasing size of inclusions. The distribution of the entrapped inclusions along the nozzle length is non-uniform and the volume fraction of inclusions in the clogging materials should be considered in order to more precisely predict the accumulated weight of molten steel that can be poured before the nozzle is fully blocked by clogging. Under the conditions assumed: 150 mm×150 mm billet, 2.0 m/min casting speed, approximately 25°C superheat, 1 m length of the SEN (Al₂O₃–C materials), 20 μm inclusions diameter in a single size, 30 ppm T.O and 40 mm nozzle diameters, the prediction shows that ~351 ton steel can be poured for the current billet continuous caster.

View full abstract

We began development of swirling-flow submerged entry nozzles in 1997 as a fundamental and effective measure for controlling the flow pattern in continuous casting molds. As a first step, we developed a swirling-flow submerged entry nozzle for round billet casting at the Wakayama works. We then began developing swirling-flow submerged entry nozzles for slab casting. The main purpose of the present work was to demonstrate that the formation of swirling flow in submerged entry nozzle improves productivity and the quality of products in continuous casting. We examined swirling-flow submerged entry nozzles with a swirl blade in these main bodies because such an arrangement is the easiest way to apply swirling flow to submerged entry nozzles in continuous casters without investment by facilities. We had only to change the submerged entry nozzle in the experiment. Swirling-flow submerged entry nozzles for slab casting were developed and their operation examined at the Wakayama and Kashima works. It was found that the proposed submerged entry nozzles increased the casting speed and improved the surface quality of slabs and steel sheets.

View full abstract

The adaptation of blast furnaces to the new technologies has increased the operation information so that the sensor information can be known at every moment. However this often results in the supply of excessive data volume to the plant operators. This paper describes an industrial application for self-organized maps (SOM) in order to help them make decisions regarding blast furnace control by means of pattern recognition and the matching of temperature profiles supplied by the thermocouples placed on the above burden. The classification of patterns via easy color coding indicates to the operator what the blast furnace operational situation is, thus making the necessary corrections easier.

View full abstract

Fault diagnosis plays an important role during the process of blast furnace ironmaking for producing safety. In this paper the important parameters of puddling process are selected as judgments criterions of fault diagnosis by analyzing the changes of these parameters. Support vector machine (SVM) is used to establish the fault diagnosis model for its suitable characters for fault classification. But the stability and accuracy of model based on single SVM could not meet the needs of practical ironmaking. Therefore, a SVM ensemble based on bagging is presented to establish a novel fault diagnosis system. The real-time producing data are collected in 5# blast furnace of a steel enterprise for training and testing the fault diagnosis models with single SVM and SVM ensemble. The experiments about the comparison between single SVM and SVM ensemble and about the SVM ensembles with different number of individual SVM are made. The experimental results demonstrate that the performance of novel fault diagnosis system based on SVM ensemble is better than the one based on single SVM, and the best fault diagnosis system that can meet the practical needs of ironmaking is found.

View full abstract

Numerical analysis models for galvanic corrosion have been developed in recent years, but most of the ordinary models can simulate only current density and potential distribution in an electrolyte solution. In order to make clear corrosion mechanism, more information such as pH, ion and corrosion product distributions are necessary. However, it is difficult to compute these phenomena by mathematical model because various kinds of ions and complex reactions exist in corrosion progress. Calculating ion movement with satisfying electroneutrality is especially difficult.
We have developed a new numerical analysis model for galvanic corrosion that can calculate ions movement and reactions. In this model, ion density distribution is corrected by solving Poisson's equation to satisfy electroneutrality. Reactions are calculated based on chemical equilibrium.
Galvanic corrosion of a model Fe/Zn couple in a NaCl solution was calculated with this model. The difference of ion and corrosion product distributions in 50 ppm and 50000 ppm NaCl solutions was discussed. Corrosion product distribution obtained by this numerical analysis model agreed well with OH distribution of corrosion product measured by FT-IR method qualitatively.

View full abstract

When galvanized steel strip is produced through a continuous hot-dip galvanizing process, the thickness of the adhered zinc film is controlled by normally impinging a thin plane nitrogen gas jet. In such a gas wiping process, frequently there appears stain of check-mark, hereinafter called “check mark stain”. The check mark stain is due to non-uniform zinc coating over the surface of the steel strip. Presence of such check-mark stain lowers the quality, productivity and profitability of the end products. From our proceeding research, it was found that there are alternating stream-wise vortices impinging on the steel strip that move almost periodically to the right and to the left along the stagnation line due to the jet flow instability. This instability is closely related to buckling of the center sheet of the plane jet. Since higher stagnation pressure removes more molten zinc adhered on the surface, the zinc coating thickness is thinner at the high pressure point. In addition, since the strip moves upward with a constant speed, the non-uniform coating surface is formed with a variety of patterns like “W”, “V” and “X”.
In the present study, in order to avoid the appearance of the check-mark stain, a new type of the air-knife system is proposed. This system consists of the main jet and a guide jet located beneath the main jet. The main jet removes the molten zinc and the guide jet prevents the formation of alternating impinging vortices on the steel strip so that it suppresses the cause of the check-mark stain. The design concept of the proposed air-knife system is verified by investigating the 3-D turbulent flow in the impingement jet region obtained numerically by using a commercial code FLUENT. Large eddy simulation (LES) technique is used to solve the governing conservation equations of mass, momentum and heat for 3-D compressible turbulent flow field.

View full abstract

The aim of the present study was to investigate the possibilities of reaching yield strengths beyond 600 MPa for low carbon bainitic hot strip steels by vanadium microalloying together with suitable base alloying. The processing conditions and levels of carbon and nitrogen chosen in this laboratory investigation correspond to those of a typical 8 mm hot strip steel containing 0.04 mass% carbon and 0.010 mass% nitrogen from electric arc furnace practice processed in conventional or compact strip mills. It was found that a base alloying corresponding to 1.4 mass% Mn, 1.0 mass% Cr and 0.25 mass% Mo is required to form a fully bainitic structure after coiling at 400°C. The decisive factors determining the strength of bainitic hot strip steels are firstly the bainite transformation temperature and secondly the extent to which recovery of the densely dislocated bainitic ferrite can be prevented. The results of this study demonstrate that vanadium microalloying effectively prevents the recovery of the bainitic ferrite and leads to retention of the strength of the virgin bainite after coiling. This is primarily due to retardation of recovery by fine vanadium carbonitrides precipitates on dislocations and only to a lesser extent to true precipitation strengthening. With 0.08 mass% V together with 0.010–0.020 mass% N the yield strength lies in the range of 750–790 MPa compared to 680 MPa for a similar reference steel without vanadium. By raising the chromium content to 2%, yield strengths in the range of 840–880 MPa have been reached. This is attributed to a lowering of the bainite transformation temperature resulting from the higher base alloying.

View full abstract