THERMOCALC calculation facilities

Core THERMOCALC program

Extracting thermo data for end-members and reactions

In its dataset tabulation modes, THERMOCALC reads in the dataset file and tabulates the end-member thermodynamic data as prompted by the user. It also provides data on reactions among end-members, including for example log(fO2) data for buffers.

More details available here.

Phase diagram calculations

In its normal operational mode, THERMOCALC simply looks for equilibrium among the phases requested by the user, rather than seeking the most stable equilibrium possible. The user is therefore required to follow the appropriate geometrical rules for building up phase diagrams, and check at intervals that no more stable assemblage is possible.

THERMOCALC facilitates the following types of phase diagram calculation:

P-T projections

PT projections are grids of stable univariant and invariant reactions, without a mass balance constraint. Combined with compatibility diagrams, they provide a powerful insight into the basic phase relations of a system. Since they are independent of bulk composition, and therefore are not specific to an individual rock, there is no need to re-calculate a P-T projection if a suitable one has already been published.

More details available here.


In pseudosection calculations, THERMOCALC interrogates the HPx-eos to predict the equilibrium compositions of mineral and melt phases at specified values of pressure (P), temperature (T) and bulk composition (X). Thus, a pseudosection is a map showing how the most stable phase assemblage varies in P-T, T-X or P-X space, specific to a particular “equilibration volume” within a rock. It may be contoured for the compositions and/or proportions of phases. Pseudosections represent chemically closed systems.

More details available here.


A modebox simply represents how the modal proportions of phases present in a rock (i.e. at fixed bulk composition) would vary along a path in pressure or temperature.

More details available here.

Compatibility diagrams

Compatibility diagrams involve the projection of composition space for a rock into two dimensions, with the bulk composition being shown in relation to the compositions of phases in stable equilibria at the (P, T) of interest.

More details available here.

µ-µ diagrams

In µ-µ diagrams, chemical potentials replace components of the bulk composition as the independent variables. They are used where it is believed that gradients in chemical potential have been “stranded” during system closure. This interpretation might be posited for symplectites, for example.

More details available here.


In dogmin mode, THERMOCALC tries to calculate phase equilibria between all possible subsets of a list of phases, at a given pressure (P), temperature (T) and bulk composition (X). It then ranks the equilibria in order of stability by comparing the Gibbs energies of each assemblage (hence the name dogmin or “do G-min”).

Dogmin therefore allows THERMOCALC to behave like a G-minimiser, such as Perple_X or Theriak/Domino. However, unlike these programs, G-minimisation is not THERMOCALC’s natural mode of operation – in particular, because it’s dependent on good starting guesses for phase compositional variables to find equilibria. Dogmin is therefore not the way that phase diagrams are constructed in THERMOCALC, although it can be useful for exploratory calculations, and is the basis of an upcoming extension pack facilitating open-system calculations.

More details available here.

Multiple-reaction thermobarometry

In multiple-reaction thermobarometry mode, the user must have mineral analyses for an assemblage that is believed to represent the equilibrium assemblage, or a subset of the equilibrium assemblage. The analysed compositions are given to THERMOCALC, which converts them to activities using the HPx-eos, and incorporates them into a set of thermodynamic equilibrium constraints. Each of these constraints acts as an independent estimate of pressure or temperature. Using some assumptions about the uncertainties involved and how they are correlated, THERMOCALC infers one optimal estimate of pressure and/or temperature from the set of constraints, along with an uncertainty, a goodness-of-fit value and other diagnostic parameters.

More details available here.