Abstract

     Ti-based alloys were one of the first material types to which thermodynamic phase diagram calculations were applied. However, the early limi-tations in modelling, particularly with respect to the uptake of elements such as oxygen and nitrogen restricted their use. Since then improvements in modelling and the increase in computing power has enabled very accurate predictions to be made for phase equilibria in real multicomponent alloys. For conventional Ti-alloys comparison between predicted and experimental values for β-transus, Vf of α and β as a function of temperature, elemental partitioning between α and β will be shown for a variety of commercial alloys. The α2-Ti3Al and γ-TiAl based alloys present their own complexities and the present status of modelling for these alloys will be presented. The addition of O will be discussed and calculations will be presented to show its effect on phase equilibria.

1. Introduction

    It is 23 years since a detailed presentation on thermodynamic phase diagram calculations for Titanium alloys was made by Kaufman and Nesor1 at the 2nd World Conference on Titanium. They presented a series of computer calculated phase diagrams for Ti-based alloys and even included an early calculation for the Ti-Al system. Since then substantial advances have been made in terms of theoretical models, computer software and hardware and it is now possible to deal with extremely complex materials on a routine basis. This paper will present a series of up-to-date examples of how the CALPHAD method (Computer CALculation of PHAse Diagrams) can be applied to both conventional Ti-alloys and the more complex α2-Ti3Al and γ-TiAl based intermetallic alloys.The roots of the CALPHAD approach lie in the mathematical description of the thermo-dynamic properties of the phases of interest. If they are stoichiometric compounds the composition is defined and a mathematical formula is then used to describe fundamental properties such as enthalpy and entropy. Where phases exist over a wide range of stoichiometries, which is the usual case for metallic materials, other mathematical models are used which account for the effect of composition changes on free energy. Details of modelling procedures can be found in the review of Ansara2. All types of models require input of coefficients which uniquely describe the properties of the various phases and these coefficients are held in databases which are either in the open literature or are proprietary.

    Once the thermodynamics of the various phases are defined phase equilibria can be calculated using software packages such as Thermo-Calc3 which is the programme used in this work. The main method of such programmes is usually a Gibbs free energy minimisation process and there are now a variety of such software packages which can perform complex multi-component calculations. For more information the recent review by Bale and Eriksson4 provides a fairly comprehensive coverage of these.

``````