Optimization of Metallurgical Reactors Using Mathematical and Physical Modeling

Abstract

Because of the tremendous cost of pilot-plantscale metal industrial trials, previous attempts to develop new metallurgical processes have generally relied on scaling up small bench-scale experiments. Because of the lack of true understanding of the rate constants, mixing phenomena, and geometric effects that control the success of the real process, many previous pilot-plant operations have been unsuccessful. Water modeling and mathematical modeling instead have become the main approaches to optimize the metallurgical reactors. To achieve success, the model must accurately predict the phenomena of true interest, including the following: -Multiphase and turbulent fluid flow within each vessel in the process, incorporating the effects of gas stirring and including transient variations, which are the main source of variability in the product -Mixing phenomena in each vessel, including solidification, heating, and remelting of a metal layer around alloy particle additions, which directly controls the success of alloying, and is important to other reactions as well -Heat-transfer phenomena, including temperature drops between each vessel -Evolution of composition during each vessel, including the thermodynamics and kinetics of the reactions between the molten metal and interfaces with the top slag and bubble surfaces, coupled with fluid flow and mixing phenomena -Effects of the true three-dimensional features of the vessel shape (including location of inlet and outlet launder), gas injection location and flow rates, and other operating conditions on the above phenomenaExtensive past work has employed physical water models to successfully investigate fluid flow phenomena in metallurgical processes such as the steel refining and casting process 1-5 and the aluminum remelting process.6-10 The first water modeling for the fluid flow in a continuous casting mold study was carried out by Afanas'Eva et al.11 for a straight bore nozzle system. Physical water models can model single-phase fluid flow because of the simulation of kinematic viscosity of steel and water. However, it is hard to use water modeling to simulate the solidification of molten metals.With the tremendous increases in the power of computer hardware and modeling software, computational modeling offers a better design tool for creating new processes. Numerical simulation has the potential to quantify the phenomena taking place in a commercial-scale process before it is constructed. Mathematical models can yield added insight into flow phenomena. Computational models based on finite-volume or finite-element solution of the Navier-Stokes equations can include phenomena such as heat transfer, multiphase flow, and solidification in steel casting without the inaccuracies inherent in a water model. Extensive research focused on the design of the metallurgical reactors using mathematical modeling such as for molten steel system 2,12-22 and molten aluminum system. 6,23THE CURRENT JOM TOPICThe current JOM topic addresses the optimization of metallurgical reactors, suc...

Water Modeling on Circulating Flow and Mixing Time in a Ruhrstahl–Heraeus Vacuum Degasser

“…

Due to the difficulties and less flexibility of industrial trials, physical modeling and mathematical modeling have been widely used to study the fluid flow phenomena during metallurgical processes, [1][2][3][4] including the multiphase fluid flow, [5][6][7] mass transfer, [8,9] chemical reaction, [10,11] bubbles and inclusions behavior, [12][13][14] and other complex phenomena in the ladle. The Ruhrstahl-Heraeus (RH) is a vital refining process to produce high-quality steel.

…”