Physical modeling of gas/liquid mass transfer in a gas stirred ladle
- PDF / 759,268 Bytes
- 8 Pages / 584 x 778 pts Page_size
- 21 Downloads / 215 Views
I.
INTRODUCTION
GAs stirring has been used in steelmaking ladles to enhance inclusion removal, homogenize the melt, and in some cases, increase the rate of mass transfer between the metal and the top slag such as in the desulfurization reaction. In a previous paper, ~ modeling results for liquid/liquid mass transfer simulating desulfurization were presented. However, during these processes, nitrogen is often absorbed from the atmosphere into the steel and the steel is reoxidized at the plume eye where the injected gas opens up the slag cover. Fruehan et a/. 2'3'4 investigated the rate of the nitrogen reaction with liquid iron experimentally and mathematically. For steels low in oxygen and sulfur, as is the case in ladle refining, the rate is controlled primarily by liquid phase mass transfer, while, at intermediate levels of oxygen and sulfur, the rate is controlled by chemical kinetics and liquid phase mass transfer in series. The reoxidation of liquid steel by the atmosphere through the plume eye is also influenced by liquid phase mass transfer. 5 In some ladle practices, nitrogen is used as the bubbling gas in place of argon gas in order to reduce costs. In this case, nitrogen is not only absorbed from the atmosphere, but also from the injected gas. Therefore, a knowledge of liquid phase mass transfer through the top liquid surface and from gas bubbles injected into a liquid is important. The characteristics of gas bubbles as a function of the injected gas flow rate has been investigated by many investigators. 5-9 In the early work on the air injection into water, Leibson et al. 9 correlated the bubble size distribution with the orifice Reynolds number (NR~). In the laminar flow range (NR~ < 2000), the average bubble size increased with flow rate. However, with the onset of turbulence the bubble SEON-HYO KIM is Assistant Professor, Department of Metallurgy and Materials Science, Pohang Institute of Science and Technology, Pohang P.O. Box 125, Kyungbuk, Korea 680. R.J. FRUEHAN is Professor, Department of Metallurgical Engineering and Materials Science, Carnegie Mellon University, 3325 Wean Hall, Pittsburgh, PA 15213. Manuscript submitted March 18, 1987. METALLURGICAL TRANSACTIONS B
size decreased until the turbulent flow was fully developed (NRe > 10,000) and no significant change of bubble size was observed for Nee > 10,000. In the laminar flow range, Leibson et a l . ' s correlation of the bubble diameter as a function of orifice diameter and Reynolds number shows good agreement with the results of Davidson et al. 6 and Sano et al. 7 However, Sano et al. 7 further reported that the average bubble diameter increased with increasing orifice Reynolds number even though Reynolds numbers are in the turbulent flow regime. They explained that the observed difference of bubble sizes from Leibson et al. 's results in the turbulent flow regime was due to the fact that the bubble size obtained by Leibson et al. was for fine bubbles produced from the breakup of larger bubbles formed at the orifice, while the bubble diam
Data Loading...
