The application of computational thermodynamics and a numerical model for the determination of surface tension and Gibbs–Thomson coefficient of aluminum based alloys

In this paper, a solution for Butler's formulation is presented permitting the surface tension and the Gibbs–Thomson coefficient of Al-based binary alloys to be determined. The importance of Gibbs–Thomson coefficient for binary alloys is related to the reliability of predictions furnished by predictive cellular and dendritic growth models and of numerical computations of solidification thermal variables, which will be strongly dependent on the thermophysical properties assumed for the calculations. A numerical model based on Powell hybrid algorithm and a finite difference Jacobian approximation was coupled to a specific interface of a computational thermodynamics software in order to assess the excess Gibbs energy of the liquid phase, permitting the surface tension and Gibbs–Thomson coefficient for Al–Fe, Al–Ni, Al–Cu and Al–Si hypoeutectic alloys to be calculated. The computed results are presented as a function of the alloy composition.

► Surface tension and the Gibbs–Thomson coefficient are computed for Al-based alloys.
► Butler's scheme and ThermoCalc are used to compute the thermophysical properties.
► Predictive cell/dendrite growth models depend on accurate thermophysical properties.
► Mechanical properties can be related to the microstructural cell/dendrite spacing.