Interaction of Injector Design, Bubble Size, Flow Structure, and Turbulence in Ladle Metallurgy Kwaku B. Owusu, Tim Haas, Prince Gajjar, Moritz Eickhoff, Pruet Kowitwarangkul, and Herbert Pfeifer In ladle metallurgy, the flow of purge gas through injectors promotes an effective mixing of the melt concerning composition and energy. In this work, different types of gas injectors, positioned eccentrically at 66% of the ladle radius are investigated in terms of the bubble size distribution, the resultant flow field velocity, and turbulent kinetic energy. The experiments are carried out in a 1:3 scale water model of a 185 t ladle using Particle Image Velocimetry (PIV) and image processing. It is shown that a porous plug provides more intensive bulk convection and a higher degree of turbulence than the other tested injectors. The differences are explained by the generation of smaller bubbles, which transfer more momentum into the liquid. The differences between the injectors are small, though. Thus, it is concluded that in comparison with other process parameters, the type of injector plays a minor role in the efficiency of ladle metallurgy. 1. Introduction With the ever-increasing demand for high-quality steel, primarily in high-tech applications, ladle metallurgy has gained attention in a number scientic studies. Nevertheless, the complex process is still not fully understood. Ladle rening is used for purposes of temperature homogenization, desulphurization, degassing, adjustment of alloying elements as well as inclusion removal. During this process, argon gas is injected into the molten steel from the bottom part of the ladle through one or a number of porous plugs. The argon disintegrates into gas bubble column(s), known as plumes. Due to the buoyancy force, the bubbles rise and escape the melt through the free surface at the top. [1,2] As they rise, the bubbles induce a recirculation ow in the ladle, that provides effective mixing. The mixing efciency is among other factors determined by the gas ow rate, plug position, bath height, and slag layer. Hence, a comprehensive under- standing of these factors is essential for effective process control and possible pro- cess optimization. Many studies report the major process variables relevant to gas stirred ladle metallurgy and consequently their inu- ences under real ladle operations are now known with a considerable level of accu- racy. [2] Independently, different studies have indicated that the gas ow rate is the key determinant in providing sufcient stirring energy while limiting slag eye formation. [2,3] It is also evident that plug position is not negligible when optimizing the steel renery operation. Nunes et al. [4] observed that better mixing is obtained when the porous plug is positioned eccentrically at mid-radius. It has also been found that mid-radius is the most favorable positioning for single and dual plug bubbling. [2] In contrast, Li et al. [5] investigated different plug positions and observed that mixing time decreases with increasing plugs radial distance. They also found that a maximum wall stress occurs at a radial plug position of 0.67 R, while a radial plug position of 0.73 R induced a different ow eld which reduced the wall shear stress. Multiple plugs located diametrically opposite at mid-bath radius have proven to provide good recirculation and signi- cantly shorter mixing time. [6] Domgin et al. [7] and Freire et al. [8] established a rm connection that plugs positioned close to one another or ladle walls produces deecting plumes, known as Coanda effect. Evidence also exists that higher bath depth provides better circulation and tends to reduce mixing time. [9] In addition, the height of the molten liquid determines the size of the slag-eye opening in the ladle. A lower bath depth is likely to cause a larger slag eye-opening, consequently exposing a larger area of the molten metal surface to the atmosphere. Cloete et al. [10] reported that an increased bath depth tends to provide higher kinetic energy inux per volume of the stirring gas and reduced viscous dissipation in the plume region. Different injector designs are in use, although their inuence on the process performance has not been quantied yet. The gas injector design is responsible for the determination of bubble evolution, regime, and diameter. [11] These gas bubble phase interactions can alter the ow pattern and inuence the ow characteristics of the entire liquid bath. Understanding the impact of different gas injector designs on the ow velocity, turbulent K. B. Owusu, P. Gajjar, P. Kowitwarangkul The Sirindhorn International Thai-German Graduate School of Engineering (TGGS) King Mongkuts University of Technology North Bangkok (KMUTNB) 1518 Pracharat 1 Rd., Wongsawang, Bangsue, Bangkok 10800, Thailand E-mail: kwaku.b-pe2016@tggs.kmutnb.ac.th K. B. Owusu, T. Haas, M. Eickhoff, P. Gajjar, Prof. H. Pfeifer Department for Industrial Furnaces and Heat Engineering (IOB) RWTH Aachen University Kopernikusstraße 10, 52074 Aachen, Germany E-mail: haas@iob.rwth-aachen.de The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/srin.201800346. DOI: 10.1002/srin.201800346 www.steel-research.de FULL PAPER steel research int. 2018, 1800346 © 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1800346 (1 of 10)