Alloying elements in steels often have dramatically different diffusion coefficients. For example, Carbon diffuses 5 to 10 orders of magnitude faster than Manganese depending on the temperature of interest. This raises the question, which elements determine the phase transformation kinetics: the slow diffusing substitutional elements (e.g. Mn or Cr) Or the fast-diffusing interstitial elements like C or Nor both? This has been extensively discussed in the literature and several models have been proposed to answer this question, namely ortho-equilibrium (ORTHO), para-equilibrium (PARA) and non-partitioning local equilibrium (NPLE).

Single precipitate growth can be kinetically simulated using the Thermo-Calc diffusion module DICTRA (diffusion controlled phase transformations) under the constraint of local equilibrium at the interface under NPLE, ORTHO, or PARA conditions by maintaining the flux balance to and from the moving interface. The former involves finding the operating tie-line at the interface and numerically solving the full multicomponent diffusion. For certain steel compositions, such calculations predict a transition from a fast growth rate under NPLE to a slow growth rate under ORTHO conditions. However, while reliably predicting experimentally observed growth rates, these calculations are computationally expensive and prone to convergence problems. For the simulation of nucleation and growth of precipitates faster and more robust models are required. Recently a simplified growth model for multicomponent systems has been developed and implemented in the precipitation module TC-PRISMA, which automatically switches from fast to slow growth rate conditions. Details and some simulation results using this new model are presented.