Predicting kinetic interface condition for austenite to ferrite transformation by multi-component continuous growth model

A critical issue restricting the application of Hillert-Agren-Liu solute-drag based model to predict interfacial conditions at a migrating ferrite-austenite interface is the lack of value for trans-interface diffusion parameter, i.e., the L parameter. Even an estimation of the parameter's order of magnitude is difficult due to its ambiguous physical interpretation. In this paper we extend a different solute-drag based model, namely the binary continuous growth model originally developed for rapid solidification, toward austenite to ferrite phase transformation to avoid this long-standing issue. The extensions consist of the treatments of multi-alloying components including interstitial carbon element and Gibbs-Thomson effect. The extended multi-component continuous growth model employs a physical parameter with clear physical meanings, i.e., interface diffusive speed in describing trans-interface energy dissipation, and can predict kinetic interface conditions and spontaneous diffusion-controlled to diffusion-less transition without using the ambiguous L parameter. The model is verified by the good agreement of its calculation results with those predicted by Hillert-Agren-Liu model for Fe-C alloys and Fe-C-Mn alloys. Further the model's advantages over Hillert-Agren-Liu model are demonstrated by calculating the kinetic interface condition phase diagrams of Fe-C-Mn-Ni alloys. It is concluded that the extended multi-component continuous growth model is valuable in unifying the efforts in addressing the common question of predicting deviations from local equilibrium at a fast-migrating interface during solidification and solid-state phase transformation.

Automated summary

This paper addresses the issue of lacking a value for the trans-interface diffusion parameter in the Hillert-Agren-Liu solute-drag based model by extending a different solute-drag based model. The extended model utilizes a physical parameter with clear physical meanings to describe the trans-interface energy dissipation, and can predict kinetic interface conditions and diffusion-controlled to diffusion-less transition without using an ambiguous parameter. The model is validated by comparing its results with the Hillert-Agren-Liu model and demonstrated its advantages in calculating kinetic interface condition phase diagrams.