Thermodynamic modeling of dilute components in multicomponent solutions

Those performing applied calculations with thermodynamic database computing systems often find that an essential component of a particular solution phase is missing from the database. Provided that one is interested only in dilute solutions of this component, and only over a relatively narrow concentration range of the major components, the solution can often be treated with acceptable accuracy for practical purposes by assuming that the activity coefficient of the dilute component is constant, independent of composition over the composition range of interest (Henry’s Law). It is shown that, in this case, the only required model parameter, apart from the constant activity coefficient γ , is a constant ν , defined as the total number of independent species, not already present in the solvent solution, into which the dilute component dissociates. That is, the equation for the chemical potential of the dilute component is independent of the model used for the solution in the database. Consequently, simple software can be written permitting one to add dilute species to existing solution databases, while performing calculations, by supplying values only of the constants γ and ν . A sample application is given of the calculation of Fe3+/Fe2+ and Cu2+/Cu+ redox ratios in a 9-component glass melt.