austente
Thermodynamics · guide

How CALPHAD computes a phase diagram

CALPHAD turned phase diagrams from hand-drawn experimental maps into something you can compute for any composition and temperature. The idea is simple to state and powerful in practice: write down the free energy of every phase, then let the lowest energy win.

Energy, not geometry

A phase diagram is the visible shadow of an energy competition. CALPHAD works directly on that energy: each phase gets a model for its molar Gibbs free energy as a function of composition and temperature, and the equilibrium state is whichever combination of phases minimizes the total. The diagram is then just a map of which assemblage wins where.

Modelling one phase

For a simple solution phase the molar Gibbs energy is built from three physically distinct parts:

G = Σ xᵢ·°Gᵢ + RT·Σ xᵢ·ln xᵢ + G^excess

The first term is the mechanical mixture of the pure components (°Gi, the SGTE lattice-stability data). The second is ideal mixing entropy. The third — the excess energy — carries all the non-ideal interaction and is usually a Redlich–Kister polynomial:

G^excess = xₐ·x_b · Σₖ Lₖ · (xₐ − x_b)^k

The Lk are the interaction parameters that get fitted. Magnetic ordering adds a further Inden–Hillert–Jarl term for phases like bcc iron.

Ordered phases and the compound-energy formalism

Intermetallics and interstitial solutions are not random mixtures, so they are described with sublattices. The compound-energy formalism assigns species to each sublattice and writes the energy in terms of the end-member compounds plus mixing on each sublattice — the standard way to model carbides, σ-phase, ordered B2/L1₂ phases and the like.

Finding equilibrium

With every phase's energy defined, equilibrium at a fixed temperature and overall composition is a constrained minimization:

minimize G_total = Σ_phases n^φ · G^φ subject to mass balance
Geometrically this is the common-tangent construction in a binary and the lower convex hull of the Gibbs surfaces in higher order systems. A robust solver grids the surfaces, takes the hull, then polishes locally — which is exactly how austenite's in-browser engine works.
Open the calculatorCALPHAD equilibrium calculatorCompute the equilibrium phase assemblage and fractions from composition and temperature, with SGTE lattice stabilities and Redlich–Kister excess terms — in the browser.

What you get beyond the diagram

Because the method is energy-based, the same calculation yields phase fractions, phase compositions, activities, driving forces for precipitation, and the inputs that downstream kinetics models (precipitation, solidification, diffusion) need. The phase diagram is the headline; the thermodynamic detail underneath is what makes CALPHAD useful for real alloy design.

Frequently asked

What does CALPHAD stand for?
CALculation of PHAse Diagrams. It is a method that models the Gibbs free energy of every phase in a system as a function of composition and temperature, then finds the phase assemblage that minimizes the total energy — reproducing the phase diagram and much more.
How is a phase diagram actually computed?
At a chosen temperature and overall composition, the solver minimizes the total Gibbs energy over all candidate phases subject to mass balance. Geometrically that is the common-tangent (binary) or convex-hull (multicomponent) construction on the Gibbs-energy surfaces. Sweeping temperature and composition traces the phase boundaries.
Why do two CALPHAD databases disagree?
The model parameters are assessed by fitting experimental and first-principles data, and different assessments weight that data differently — especially for metastable phases and sparsely measured corners. The phase models and even which phases are included can differ, so always state the database behind a result.

References

  1. N. Saunders, A.P. Miodownik, "CALPHAD: A Comprehensive Guide," Pergamon, 1998.
  2. H.L. Lukas, S.G. Fries, B. Sundman, "Computational Thermodynamics: The Calphad Method," Cambridge University Press, 2007.
  3. A.T. Dinsdale, "SGTE data for pure elements," CALPHAD 15(4), 1991.
  4. M. Hillert, "The compound energy formalism," Journal of Alloys and Compounds 320, 2001.

Related guides

Browse all engineering guides →