Steelmaking
Visualization of Intermittent Splash with Gas Blowing from a Top Lance – Breakup of Cavity Surface Causing Intermittent Splash –
2020 Volume 60 Issue 9 Pages 2118-2120
Details
2020 Volume 60 Issue 9 Pages 2118-2120
This study aims to visualize an instantaneous event of intermittent splashing from a bath surface caused by gas blowing from a top lance. The visualization techniques include computational fluid dynamics (CFD) and an experimental procedure using a high-speed camera. A remarkable finding is the fact that the difference between the air pressure inside the cavity and hydrostatic pressure outside the cavity forms a constriction and causes a strong intermittent splash.
In a steelmaking process, a high-speed oxygen jet is blown on the surface of a molten iron bath in a converter to remove impurities such as carbon and phosphorus. The high-speed jet, which is blown from a top lance, impinges on the bath surface, creating a cavity and causing an intermittent phenomenon called splash. Unfortunately, splash generates metal dust that both reduces the quality of the metal and erodes the refractory of the converter wall. It is therefore desirable to elucidate the mechanism by which gas blowing from a top lance causes splash.
Several papers address the cavity formation and splash generation caused by gas blowing from a top lance.1,2,3,4,5,6,7,8,9,10,11,12,13) For instance, according to Tokuda,7) gas jet blowing from a top lance causes a cavity, and the resulting geometry can be classified into four types, depending on the gas flow rate (see Fig. 1). The phenomenon of cavity formation and splash generation is a classical problem that was investigated with low-resolution cameras between the 1970s and late 1980s. Various studies examined the influence of conditions on the cavity shape and amount of splash, but few studies attempted to investigate the instantaneous events that actually cause a splash. Taking advantage of recent progress in video and computer hardware, the present study targets visualization of the instantaneous events of cavity formation and splash during gas blowing from a top lance. The visualization techniques employed in this study include computational fluid dynamics (CFD) and an experimental procedure using a high-speed camera. Similar to the previous studies, a so-called water (or cold) model experiment, in which molten iron is replaced with water, was adopted in this study because the two fluids have similar kinematic viscosities. In addition, the influence of surface tension is known to be sufficiently smaller than that of the Reynolds number.14)
Sketch of selected cavity patterns caused by gas blowing from a top lance (reproduced from Tokuda7)). (Online version in color.)
A transparent cylindrical container having an inner diameter of 300 [mm] is surrounded by a larger transparent rectangular tank. Both the rectangular tank and the cylindrical container are partially filled with tap water to reduce distortion in the visualized photographs. An air jet is blown through a top lance having an inner diameter of 3.0 [mm], set at the height of h = 20 [mm] from the static water surface level of HL = 200 [mm]. The air flow rate Qg = 31.0, 49.0 [L/min(npt)] is regularized with a mass flow controller. The series of instantaneous events in which the blowing air creates a cavity and causes a splash on the water surface is captured with a high-speed camera (500 [fps]).
2.2. Computational ProcedureSTAR-CCM+, a commercially available CFD software package, was employed to simulate the present target event. The STAR-CCM+ numerical code, version 11.02, was employed for all numerical predictions on a 3.30 [GHz] Intel Xeon processor with 64 [GB] RAM. The computational grids were made up of hexahedral elements. A total of 1966902 cells were employed for the entire flow domain. The cell density increased near the lance exit and near the gas-liquid interface.
STAR-CCM+ uses a control volume-based technique to solve the continuous equation and the momentum equations. A segregated implicit solver and second-order upwind interpolation scheme were employed for each computational iteration, as these provided accurate predictions of the present target situation within a water/air two-phase flow. A small time-step size, Δt = 1.0*10−4 [s], was adopted to achieve convergence in every time step, and free surface behavior was captured by a volume of fluid (VOF) model. Tracking of the interface between the phases was accomplished by the solution of a continuity equation for the volume fraction of a phase.
The realizable k-ε model was selected because it provides improved predictions for the spreading rate of a jet. The convergence of the computational solution was determined based on residuals for the continuity and x-, y-, z-velocities. The residual tolerance of all quantities was set to 10−3. The solution was considered to be converged when all of the residuals were less than or equal to these default settings (see STAR-CCM + User Guide for more details).
The two-step numerical analysis method12) has been proposed as a method to efficiently simulate the jet behavior of a top-blowing lance and the dynamic behavior of a liquid deformed thereby. However, since the jet velocity from the top-blowing lance was small in the present study, the time-step size Δt that satisfies the CFL (Courant-Friedrichs-Lewy) condition could be large. Therefore, in this model, the jet behavior caused by the top-blown lance and the dynamic behavior of the liquid surface were analyzed simultaneously.
Figure 2 shows selected snapshots from the experimental high-speed photographs at Qg = 31 [L/min(npt)] and h = 20 [mm]. The photographs successfully capture a scenario in which the breakup of the air cavity causes a strong splash on the water surface, i.e., the following scenario:
Selected snapshots of experimental photographs at Qg = 31 [L/min(npt)] and h = 20 [mm].
(i, ii) Air injection from the top lance causes the air cavity to grow.
(iii) After the cavity has grown to its full size, a constricted part appears at the point where the difference between the internal air pressure inside the cavity and external hydrostatic pressure is in balance.
(iv) The constricted part is pinched off from the cavity surface and bursts out of the water, causing a strong splash. In consequence, the cavity collapses and shrinks, after which the same process is repeated from stage (i).
A supplementary movie, Movie S1 (Supporting Information), is also available.
The nature of the cavity formation and splashing phenomenon can be elucidated in detail by CFD. Figure 3 shows selected snapshots of the computed free-surface behaviors and static pressure near the cavity. The present computational procedure can successfully simulate each of the cavity events observed in the experiment, and the computed results seem to be in good agreement with the experimental photographs. In the growing stage (i) in Fig. 3, the cavity depth increases with time. In the growing stage (ii), the cavity depth is larger and the hydrostatic pressure outside the cavity is greater than the air pressure inside the cavity. In the constrict stage (iii), the difference between the air pressure inside the cavity and hydrostatic pressure outside the cavity exceeds a critical value, constriction is observed in the cavity. After that, top-blowing gas enters the lower cavity, but it cannot escape from the cavity, so the pressure in the lower cavity increases. Due to the high pressure inside the lower cavity, the constricted part can be leaped out (see Fig. 4).
Selected snapshots of computed free-surface behaviors (top) and velocity vectors (bottom) at Qg = 31 [L/min(npt)] and h = 20 [mm].
Sketch of cavity breaking up and turning into a splash.
This study has visualized the generation process of an intermittent splash due to air injection from a top lance by using Computational Fluid Dynamics (CFD) and experimental high-speed photography. In particular, the difference between the air pressure inside the cavity and hydrostatic pressure outside the cavity forms a constriction, and the high pressure in the lower part of the cavity was found to cause a strong intermittent splash.
Slow motion movie of experimental high-speed photographs. The movie successfully captures a scenario in which the breakup of the air cavity causes a strong splash on the water surface.
This material is available on the Website at https://doi.org/10.2355/isijinternational.ISIJINT-2019-788.
