Department of Chemical Engineering

Permanent URI for this community

https://scholar.sun.ac.za/handle/10019.1/321
Department Process Engineering now has a new name, and will be known from March 2023, as Department of Chemical Engineering.

Browse

Browsing Department of Chemical Engineering by browse.metadata.advisor "Bogaers, Alfred"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Momentum transfer from Arc to slag bath in an ilmenite smelting DC Arc furnace - a computational analysis
    (Stellenbosch : Stellenbosch University, 2021-12) Makgoale, Tumelo; Akdogan, Guven; Bogaers, Alfred; Zietsman, Johan; Stellenbosch University. Faculty of Engineering. Dept. of Process Engineering.
    ENGLISH ABSTRACT: Direct current (DC) arc furnaces are used for a variety of applications, including steelmaking, chromite smelting, ilmenite smelting, and nickel laterite smelting. Understanding the behaviour of DC plasma arcs is essential, as they are central to the operation of DC arc furnaces. The intent of this research project was to study and understand the interaction between the arc jet and slag bath in terms of momentum transfer to assist in furnace design and operation, particularly for ilmenite smelting processes. In an electric smelting furnace, there are a number of driving forces resulting in flow of the molten slag and alloy baths, and which influence the melting rate and temperature distribution. These include carbon monoxide bubbling, electromagnetic forces, arc jetting, and natural convection. The purpose of this research project was to find a computationally efficient and representative modelling method to describe an arc, and use this method to understand momentum transfer from arc to slag bath in DC arc furnaces. The first objective was to identify and investigate modelling methods that have been developed to describe plasma arcs, and to also select the most appropriate method to incorporate the description of an arc into multiphysics models. The criteria for an appropriate model includes: equivalence to arc behaviour, ability to reliably describe the interaction between plasma arc jet and molten bath, computational efficiency, and numerical stability during simulations. A steady-state turbulent gas jet was selected as an appropriate representation of an arc as it is capable of accounting for the thrust generated by an arc, yet is computationally simple enough to be included within a full multiphysics model suitable for furnace design. The second objective was to develop an understanding of the turbulent gas jet modelling method so that it can be applied effectively in simulations to provide an accurate and reliable arc description. A sensitivity study of the various parameters that can be modified when formulating a turbulent jet was performed. The analysis included: jet inlet velocity, gas density and viscosity, inlet nozzle diameter, jet length, as well as arc thrust. The idea was to isolate and highlight which of these parameters have significant influences on the impingement forces, and thereby highlighting which of these parameters should be chosen with care. An important finding from this investigation was the influence of the ρ 1 2 /μ ratio in maintaining Reynolds number equivalence. This approximation allows us to work with lower arc jet velocities without changing momentum transfer rate to the slag bath, while improving numerical stability and reducing computational cost. This investigation also showed that the inlet nozzle diameter has an influence on both momentum transfer to the slag bath and indentation size on the bath surface. Since there is no real physical argument to base the choice of inlet nozzle diameter, this parameter should be chosen with care. The last objective was to perform steady-state pilot plant scale simulations for a DC arc furnace to investigate the impact of furnace design and operating parameters using the turbulent jet approxi- mation as an arc. From these simulations, a 10 kA arc, with a length of 0.2 m, resulted in an average slag bath velocity of 0.0311 m s−1, an average transferred momentum of 3.342 × 10−3 kg m s−1, and a total slag bath kinetic energy of 9.3727 × 10−7 kW h. The results obtained in this study confirm that arc jetting can potentially be a major momentum driver within open bath furnaces, but the exact magnitudes may differ significantly for industrial scale furnaces.