Experimental and Numerical Modeling of Segregation in Metallic Alloys

Abstract

International audienceElectromagnetic levitation (EML) has been used as an experimental technique for investigating the effect of the nucleation and cooling rate on segregation and structure formation in metallic alloys. The technique has been applied to aluminum-copper alloys. For all samples, the primary phase nucleation has been triggered by the contact of the levitated droplet with an alumina plate at a given undercooling. Based on the recorded temperature curves, the heat extraction rate and the nucleation undercooling for the primary dendritic and the secondary eutectic structures have been determined. Metallurgical characterizations have consisted of composition measurements using a scanning electron microscope (SEM) equipped with energy dispersive X-ray spectrometry and the analysis of SEM images. The distribution maps drawn for the composition, the volume fraction of the eutectic structure, and the dendrite arm spacing (DAS) reveal strong correlations. Analysis of the measurements with the help of a cellular-automaton (CA)-finite-element (FE) model is also proposed. The model involves a new coupling scheme between the CA and FE methods and a segregation model accounting for diffusion in the solid and liquid phases. Extensive validation of the model has been carried out on a typical equiaxed grain configuration, i.e., considering the free growth of a mushy zone in an undercooled melt. It demonstrates its capability of dealing with mass exchange inside and outside the envelope of a growing primary dendritic structure. The model has been applied to predict the temperature curve, the segregation, and the eutectic volume fraction obtained upon single-grain nucleation and growth from the south pole of a spherical domain with and without triggering of the nucleation of the primary solid phase, thus simulating the solidification of a levitated droplet. Predictions permit a direct interpretation of the measurements

Three-dimensional cellular automaton-finite element modeling of solidification grain structures for arc-welding processes

“…It basically consists of identifying cells in elements, thus allowing field interpolations from FE nodes to CA cells. Considering the number of elements and cells, establishing such connections is computationally intensive and has to be reinitialized when the FE mesh is dynamically adapted during the simulation [31]. To avoid this problem a second FE mesh is defined, hereafter referred to as CA mesh that remains fixed during the simulation.…”

Prediction of Solidification Paths for Fe–C–Cr Alloys by a Multiphase Segregation Model Coupled to Thermodynamic Equilibrium Calculations

“…20) The model could also benefits from its extension to include the description of several solid phases (i.e. peritectic and eutectic microstructures as already available for binary alloys 8) ).…”

Numerical simulation of the solidification processes of copper during vacuum continuous casting

“…Therefore, the CA method possessed not only advantages of the Monte Carlo method but also physical meanings for crystallographic effects and grain selection. Some modified CA models that directly solve the transport equations at the solid-liquid interface or consider the effect of macrosegregation have been developed to simulate dendritic growth patterns [11][12][13][14][15]. Further developments of the CA model are coupling the phase-field method (PF-CA model) for the simulation of formation of macrostructures in multi-component alloy systems [16].…”