Band 20, Heft 10 - Special Issue of “Metallurgical Reaction Engineering”

Na 2 O, CaF 2 , B 2 O 3 are commonly used fluxes in metallurgical slags. These fluxes reduce the viscosity of slags are related to changing the melt structures. In this study, molecular dynamics simulation was used to analyze the main effects of Na 2 O, CaF 2 , and B 2 O 3 respectively on the melt structure of the traditional CaO–SiO 2 metallurgical slags. The results showed that since Na +  ions have only one charge, and the electrostatic restraint between the multi-charged ions and the depolymerized [SiO 4 ] 4− tetrahedrons cannot occur on Na +  ions. The overall mobility of the particles in the melt structure is improved with the replacement of Ca 2+ with Na + . Therefore, the fluidity of CaO–SiO 2 –Na 2 O slags is stronger than that of CaO–SiO 2 , and the viscosity is reduced. F − ions mainly bond with Ca 2+ ions to form complexes. The resulting Ca–F structures destroy the electrostatic restraint between Ca 2+ ions and depolymerized depolymerized [SiO 4 ] 4− tetrahedrons in the Ca–O structures, and the fluidity of CaO–SiO 2 –CaF 2 slags are improved. The effect of B 2 O 3 on the melt structure of CaO–SiO 2 is related to the basicity. In this study, when the value of basicity is 1.87, the addition of B 2 O 3 increases the degree of polymerization of the CaO–SiO 2 melt structure, and the viscosity may increases.

In this paper, the flow field of the submerged entry nozzle (SEN) in the continuous casting process is analyzed in detail by numerical simulation. It is found that the front and rear positions of the SEN outlets and the bottom of the nozzle are prone to the accumulation of inclusions. Compared with the actual SEN, the clogging positions with the simulated results are consistent. The composition of nozzle nodules at two different positions is analyzed by scanning electron microscope (SEM), and the inclusions are alumina. The changes in molten steel level characteristics in the mold without and after clogging are compared and analyzed. It is found that the velocity of molten steel near the surface with nozzle clogging reaches 0.483 m/s, the free level fluctuation value of molten steel is 8.9 mm, which is easy to cause steel defects.

Oxygen enrichment is believed to improve productivity and reduce fuel consumption, while gas fuel injection is assumed to improve the product yield and quality in iron ore sintering process. It is important to understand the mutual effect of oxygen enrichment and gas fuel injection for combined usage of these two techniques. A mathematical model is developed to simulate the sintering process with coke oven gas (COG) injection and oxygen enrichment, particularly focusing on predicting the quality and yield of sinter production, as well as NO x emission. The model is validated by comparing the model predictions with sintering pot test data, and numerical simulations are carried out to investigate the mutual effect of oxygen enrichment and COG injection. The results show that, compared with the conventional sintering process, with 0.5% COG injection and 30% oxygen enrichment, the mean melt quantity index ( MQI ) is increased by 4.1% and the mean cooling rate ( CR ) is decreased by 62.5%, showing that the sinter quality is improved significantly. The sinter yield is increased by 44.5%, whereas the NO x emission is increased by 8.3%. With the increase of oxygen enrichment from 21% to 30%, the sinter yield increases prominently first and then decreases a little, attaining its maximum at 30% of oxygen concentration. In addition, increasing oxygen concentration will increase the conversion rate of coke-N to NO x . Therefore, excessive oxygen enrichment is not only bad for sinter strength and yield, but also increases NO x emission.

The motion behavior of bubbles in a riser tube is studied in order to analyze the bubble evolution characteristics. Gas distribution and bubble movement in risers and vacuum chambers have important effects on liquid steel flow, mixing and refining process. It is found that the initial diameter of argon bubbles in the riser tube decreased with decreasing vacuum degree. The diameter of argon bubbles in the riser tube increased with increasing gas flow rate. The bubbles could be divided into the single bubble rising zone and the bubble breaking coalescence zone in the rising tube. After the bubbles were blown in, they changed from regular spherical shapes to flat shapes in the single bubble rising zone, and then broke apart into small bubbles in the bubble breaking coalescence zone. Variations in the gas flow rate and vacuum degree had significant effects on the regional distribution of bubble motion and bubble residence time. The critical height of the single bubble rising zone and the bubble breaking coalescence zone were stable when the bubble travel distance was greater than 280 mm.

The traditional combined model of a mixed flow reactor and a plug flow reactor in series with a dead zone has been commonly used to characterize tundish flows in continuous casting of steel. However, there are noticeable differences in residence time distribution curve profile between model predictions and the corresponding experimental measurements. In order to pursue a better characterization of typical tundish flows, two new combined models are outlined in the current work: One involves uniform tanks and a plug flow reactor in series with a dead zone, while the other features two non-uniform tanks and a plug flow reactor in series with a dead zone. The application of the models to a tundish with a turbulence inhibitor in the pouring compartment reveals that both of the two new models can provide a much better representation of the experimental measurements in comparison with the traditional model. Regarding the two new models, the one with two non-uniform tanks is seen to perform better than that with uniform tanks.

The packed structure, such as packing density and particle size distribution, of COREX shaft furnace, directly affects the gas flow and reaction process. A two-dimensional steady-state mathematical model was developed to study the influence of five packed structures on the gas flow, pressure distribution, species composition, and solid metallization rate in a COREX shaft furnace with center gas supply. The results show that the gas velocity is relatively uniform along the radial direction under Case-P. Under Case-InV and Case-V, the gas velocity increases and decreases gradually from the center to the wall zone respectively. The gas velocity contour in the upper part of the shaft in Case-M is ‘M’ shape, while it shows ‘W’ shape in Case-W. The order of pressure drop under five packed structures is Case-P > Case-M > Case-W > Case-InV > Case-V, and for the solid metallization rate, the order is Case-V > Case-W > Case-P > Case-M > Case- InV. As Case-V has the lowest pressure drop and largest metallization rate, the charging matrix in practical production should develop towards a ‘V’ shaped burden profile in the upper of the packed bed in the COREX shaft furnace.

Particle mixing is an important unit operation in many industry processes. This work employed the discrete element method (DEM) to characterize the mixing performance of iron ore particles in a horizontal high shear mixer. The simulation results were first compared with the positron emission particle tracking (PEPT) data reported by Forrest et al. to validate the DEM model. Then the effects of key operation factors such as impeller rotation speed, number of impellers and filling ratio on the mixing efficiency were discussed. The Lacey mixing index was employed to evaluate the mixing performance. The results show that the mixing efficiency of particles increased with increasing the number of impellers and their rotation speeds. On the other hand, the filling ratio was negatively related to the particles velocity and the mixing efficiency. This was because the effective space left for free movement of particles decreased as the filling ratio increased, and the mixing of particles was therefore restricted. In addition, the mixing rate of particles in the axial direction of the high shear mixer was much slower than that in the radial direction.

During the operation of the CSCR activated carbon desulfurization and denitrification flue gas purification system, due to the unreasonable design of flue gas pipe diameter and the uneven distribution of activated carbon porosity, the uniformity of flue gas distribution is prone to problems, which affects flue gas removal efficiency and increases operating costs. In this article, the Fluent software was used to establish a three-dimensional numerical model of the flue gas purification system, and the flow characteristics of the flue gas were studied by continuously optimizing the structure of the system model. The effects of different pipe diameters, the accumulation method of activated carbon in the desulfurization and denitration module and the distribution of porosity on the flue gas are analyzed, and the effects on the average residence time and variance of flue gas were further analyzed. The results show that the average residence time, the variance, the dead volume fraction, and well-mixed volume decreased, while the dispersed plug volume fraction increased for a module after optimization. The average residence time and the dispersed plug volume fraction decreased. While the variance, the dead volume fraction, and well-mixed volume fraction increased for the system after optimization.

Scrap melting in the steelmaking process has recently attracted considerable attention. This study developed a mathematical model to investigate the effects of sizes, initial temperatures and carbon contents of scrap, and bath flow on its melting behaviors. The results showed that the volume of the solidified shell on the melting interface could be decreased or eliminated by increasing the initial temperature of the scrap. The initial carbon content also significantly influenced the melting behavior. The completion melting time decreased from 253 to 80 s when the carbon content increased from 0.10% to 0.30%. Importantly, the natural convection or driven flow formed in the bath significantly influenced the melting behaviors. Moreover, the dominating factors changed as the melting proceeded. Therefore this study provides more insights into the melting behavior of the scrap during the steelmaking process at different operating conditions.