Finite elements of macrosegregation during binary alloys solidification

VICTOR FACHINOTTI; IGNACIO RINTOUL

Salta

Congreso; X Congreso Argentino de Mecánica Computacional; 2012

Universidad Nacional de Salta y la Asociacion Argentina de Mecanica Computacional

<!-- @page { margin: 2cm } P { margin-bottom: 0.21cm; direction: ltr; color: #000000; line-height: 115%; text-align: left; widows: 2; orphans: 2 } P.western { font-family: "Calibri", sans-serif; font-size: 11pt; so-language: en-US } P.cjk { font-family: "Times New Roman", serif; font-size: 11pt; so-language: zxx } P.ctl { font-family: "Calibri", sans-serif; font-size: 11pt; so-language: ar-SA } A:link { color: #0000ff } --> Macrosegregation is the redistribution of solute from its nominal concentration. This phenomenon may takes place during the solidification of an alloy. It is governed by the momentum, heat and solute conservation equations in the solidifying alloy, all of them mutually coupled. This contribution is devoted to model and experimental validation of macrosegregation during binary alloys solidification using improved finite element tools. Here, the alloy is assumed to have mushy-type solidification, where a solid-liquid (or ?mushy?) region develops between the fully liquid and the fully solid material. Momentum equations resemble the Navier-Stokes equations for the completely liquid material. The mushy region is modeled as a porous medium, and the flow therein is governed by the Darcy?s law. Following this approach, the solid is modeled as a zero-permeable porous body. Flow is mainly induced by thermal and solutal buoyancy forces, which are accounted for using the Boussinesq?s approximation for the gravity term in the Navier-Stokes equations. As a first approach, the domain of analysis will be assumed plane. All the equations are spatially discretized using linear triangular finite elements, where convection effects are stabilized using SUPG (Streamline Upwind Petrov-Galerkin). Incompressibility in Navier-Stokes equations is dealt with using PSPG (Pressure-Stabilized Petrov-Galerkin). In all cases, the fully-implicit backward-Euler method is used for time stepping. Finally, the model is validated by comparison against numerical and experimental results available in the literature.