Abdellah Kharicha

Modeling of Multiscale and Multiphase Phenomena in Materials Processing

In order to demonstrate how CFD can help scientists and engineers to better understand the fundamentals of engineering processes, a number of examples are shown and discussed. The paper covers (i) special aspects of continuous casting of steel including turbulence, motion and entrapment of non-metallic inclusions, and impact of soft reduction; (ii) multiple flow phenomena and multiscale aspects during casting of large ingots including flow-induced columnar-to-equiaxed transition and 3D formation of channel segregation; (iii) multiphase magneto-hydrodynamics during electro-slag remelting; and (iv) melt flow and solidification of thin but large centrifugal castings.

A Dynamic Mesh-Based Approach to Model Melting and Shape of an ESR Electrode

This paper presents a numerical method to investigate the shape of tip and melt rate of an electrode during electroslag remelting process. The interactions between flow, temperature, and electromagnetic fields are taken into account. A dynamic mesh-based approach is employed to model the dynamic formation of the shape of electrode tip. The effect of slag properties such as thermal and electrical conductivities on the melt rate and electrode immersion depth is discussed. The thermal conductivity of slag has a dominant influence on the heat transfer in the system, hence on melt rate of electrode. The melt rate decreases with increasing thermal conductivity of slag. The electrical conductivity of slag governs the electric current path that in turn influences flow and temperature fields. The melting of electrode is a quite unstable process due to the complex interaction between the melt rate, immersion depth, and shape of electrode tip. Therefore, a numerical adaptation of electrode position in the slag has been implemented in order to achieve steady state melting. In fact, the melt rate, immersion depth, and shape of electrode tip are interdependent parameters of process. The generated power in the system is found to be dependent on both immersion depth and shape of electrode tip. In other words, the same amount of power was generated for the systems where the shapes of tip and immersion depth were different. Furthermore, it was observed that the shape of electrode tip is very similar for the systems running with the same ratio of power generation to melt rate. Comparison between simulations and experimental results was made to verify the numerical model.

Modeling the Effects of Strand Surface Bulging and Mechanical Softreduction on the Macrosegregation Formation in Steel Continuous Casting

Positive centerline macrosegregation is an undesired casting defect that frequently occurs in the continuous casting process of steel strands. Mechanical softreduction (MSR) is a generally applied technology to avoid this casting defect in steel production. In the current paper, the mechanism of MSR is numerically examined. Therefore, two 25-m long horizontal continuous casting strand geometries of industrial scale are modeled. Both of these strand geometries have periodically bulged surfaces, but only one of them considers the cross-section reduction due to a certain MSR configuration. The macrosegregation formation inside of these strands with and without MSR is studied for a binary Fe-C-alloy based on an Eulerian multiphase model. Comparing the macrosegregation patterns obtained for different casting speed definitions allows investigating the fundamental influence of feeding, bulging and MSR mechanisms on the formation of centerline macrosegregation.

Influence of the Slag/Pool Interface on the Solidification in an Electro-Slag Remelting Process

The Electro-Slag-Remelting (ESR) is an advanced technology for the production of components of e.g. high quality steels. In the present study a comprehensive computational model using the VOF technique for the prediction of the slag/pool interface is presented for axisymmetric and steady state conditions. In this model the distribution of the electric current is not constant in time, but is dynamically computed according to the evolution of the slag and steel phase distribution. The turbulent flow, created by the Lorentz and buoyancy forces, is computed by solving the time-averaged mass and momentum conservation equations. The turbulence effect is modelled by using a k-model. Two numerical simulations were performed, one assuming a flat interface, and a second leaving the interface free to find an equilibrium shape. The results are then analysed and compared for both cases.

Modelling macrosegregation in a 2.45 ton steel ingot

A three phase model for the mixed columnar-equiaxed solidification was proposed by the current authors [Wu and Ludwig 2006 Metall. Mater. Trans. 37A 1613-31]. The main features of the mixed columnar-equiaxed solidification are considered: the growth of the columnar dendrite trunks from the ingot surface, the nucleation and growth of the equiaxed crystals, the sedimentation of the equiaxed crystals, the thermal and solutal buoyancy flow and its interactions with the growing crystals, the solute partitioning at the solid-liquid interface during solidification, the solute transport due to melt convection and equiaxed sedimentation, the mechanical interaction/impingement between columnar and equiaxed crystals and the columnar-to-equiaxed transition (CET). However, due to the model complexity and the limited computational capability the model has not yet applied to the large steel ingots of engineering scale. This paper is going to simulate a 2.45 ton big-end-up industry steel ingot, for which some experimental results were reported [Marburg 1926 Iron Steel Inst. 113 39-176]. Here a simplified binary phase diagram for the steel (Fe-0.45 wt. %C) is considered. Comparison of the modelling results such as as-cast columnar and equiaxed zones, macrosegregation with the experimental results is made. Details about the formation sequence of the distinguished crystal zones and segregation patterns are analyzed.

Importance of Melt Flow in Solidifying Mushy Zone

A mixture solidification model is employed to study the interaction between the melt flow and the growing mushy zone. The goal is to address the importance of considering the melt flow and flow pattern (laminar or turbulent) in the growing mushy zone. A simple 2D benchmark with parallel flow passing by/through a vertically growing mushy zone is considered. Parameter studies with different velocities and flow patterns are performed. It is found that the flow velocity and flow pattern in and near the mushy zone plays an extremely important role in the formation of the mushy zone. The mushy zone thickness is dramatically reduced with the increasing melt velocity. Simulations with/without considering turbulence show significantly different results. The turbulence in the mushy zone is currently modeled with a simple assumption that the turbulence kinetic energy is linearly reduced with the mush permeability.

Using four-phase Eulerian volume averaging approach to model macrosegregation and shrinkage cavity

This work is to extend a previous 3-phase mixed columnar-equiaxed solidification model to treat the formation of shrinkage cavity by including an additional phase. In the previous model the mixed columnar and equiaxed solidification with consideration of multiphase transport phenomena (mass, momentum, species and enthalpy) is proposed to calculate the as- cast structure including columnar-to-equiaxed transition (CET) and formation of macrosegregation. In order to incorporate the formation of shrinkage cavity, an additional phase, i.e. gas phase or covering liquid slag phase, must be considered in addition to the previously introduced 3 phases (parent melt, solidifying columnar dendrite trunks and equiaxed grains). No mass and species transfer between the new and other 3 phases is necessary, but the treatment of the momentum and energy exchanges between them is crucially important for the formation of free surface and shrinkage cavity, which in turn influences the flow field and formation of segregation. A steel ingot is preliminarily calculated to exam the functionalities of the model.

Simultaneous Observation of Melt Flow and Motion of Equiaxed Crystals During Solidification Using a Dual Phase Particle Image Velocimetry Technique. Part II: Relative Velocities

A two-camera Particle Image Velocimetry (PIV) technique is applied to study the flow pattern and the equiaxed crystal motion during an equiaxed/columnar solidification process of Ammonium Chloride in a die cast cell. This technique is able to measure simultaneously the liquid and the equiaxed grain velocity pattern as already shown in Part I of this paper. The interaction between the equiaxed grains and the melt flow was explored by means of relative velocities. In single isolated configurations, the settling velocity of equiaxed crystal was found to be 41 times smaller than spheres of equivalent size. The coupling between the fluid flow and the equiaxed crystals was found to be important in areas of high crystal density. Chaotic and turbulent behaviors are found to be damped in regions of high equiaxed crystal density.

Simulation of the as-cast structure of Al-4.0wt.%Cu ingots with a 5-phase mixed columnar-equiaxed solidification model

Empirical knowledge about the formation of the as-cast structure, mostly obtained before 1980s, has revealed two critical issues: one is the origin of the equiaxed crystals; one is the competing growth of the columnar and equiaxed structures, and the columnar-to-equiaxed transition (CET). Unfortunately, the application of empirical knowledge to predict and control the as-cast structure was very limited, as the flow and crystal transport were not considered. Therefore, a 5-phase mixed columnar-equiaxed solidification model was recently proposed by the current authors based on modeling the multiphase transport phenomena. The motivation of the recent work is to determine and evaluate the necessary modeling parameters, and to validate the mixed columnar-equiaxed solidification model by comparison with laboratory castings. In this regard an experimental method was recommended for in-situ determination of the nucleation parameters. Additionally, some classical experiments of the Al-Cu ingots were conducted and the as-cast structural information including distinct columnar and equiaxed zones, macrosegregation, and grain size distribution were analysed. The final simulation results exhibited good agreement with experiments in the case of high pouring temperature, whereas disagreement in the case of low pouring temperature. The reasons for the disagreement are discussed.

Shallow water model for horizontal centrifugal casting

A numerical model was proposed to simulate the solidification process of an outer shell of work roll made by the horizontal centrifugal casting technique. Shallow water model was adopted to solve the 2D average flow dynamics of melt spreading and the average temperature distribution inside the centrifugal casting mould by considering the centrifugal force, Coriolis force, viscous force due to zero velocity on the mould wall, gravity, and energy transport by the flow. Additionally, a 1D sub-model was implemented to consider the heat transfer in the radial direction from the solidifying shell to the mould. The solidification front was tracked by fulfilling the Stefan condition. Radiative and convective heat losses were included from both, the free liquid surface and the outer wall of the mould. Several cases were simulated with the following assumed initial conditions: constant height of the liquid metal (10, 20, and 30 mm), uniform temperature of the free liquid surface (1755 K). The simulation results have shown that while the solidification front remained rather flat, the free surface was disturbed by waves. The amplitude of waves increased with the liquid height. Free surface waves diminished as the solidification proceeded.