Historically, the materials science and engineering field has relied almost exclusively on experimentation to establish the processing-structure-property relationships of materials. Empirical models were initially developed through trial and observation correlating a property to the experimental conditions. As our understanding through materials science has evolved, semi-empirical and mechanistic models have been derived that are more powerful and more predictive.
Thermo-Calc is based on the CALPHAD approach: a self-consistent, phase-based approach that relates the underlying thermodynamics and phase equilibria of chemical systems by capturing the composition and temperature dependencies. Databases are constructed through a critical evaluation of experimental data (combined with ab initio predictions where needed) of the underlying binary and ternary systems, allowing Thermo-Calc to make predictions for multicomponent systems of interest.
The approach has also been extended to consider the modelling of atomic mobilities that can be used to describe kinetic phenomena, such as diffusion and precipitation.
Molar volume data is also assessed, which can be used to model density and thermal expansion. By combining the thermodynamic and mobility data, predictions can be made in a wide range of areas such as welding, heat treating, corrosion, and alloy production.