Abstract:
There is a lack of tundish models focused on the analysis of slag-steel-inclusion interaction, which are very important for understanding the steel reoxidation by tundish slag. In previous studies from literature in this field, it is found relevant deviations between calculated and experimental results, especially for predictions of oxygen and silicon contents. Besides, no information is provided regarding temporal behavior of inclusions as steel composition changes with time during steel/slag interaction. In this context, the present work aims at contributing to the development of a comprehensive model able to accurately predict steel composition as a function of time. For this purpose, a reaction zone model coupling thermodynamics and mass transfer, adapted from (Kim et al., 2018) [11], is employed. In addition, the fundamentals of this model are used for developing a sub-model able to simulate steel/inclusion interaction, allowing to predict the behavior of a single inclusion with time (size and phase transformations). The slag-steel-inclusion interaction is simulated in FactSage macro processing module. The model is able to accurately reproduce compositional evolution of molten steel, using laboratory scale experimental data from literature. It is found that the greater the SiO2 concentration in the slag, the greater silicon and oxygen pick-up. Si is quickly incorporated into bulk steel; its concentration can increase from 178 to 1487 ppm in 20 minutes. The inclusion growth curves follow the O concentrations in bulk steel, and inclusion size increases with increasing SiO2 content in the slag. While, in the case of high silica slag, inclusion is composed only of Al2O3 phase, in the case of low silica slag, inclusion consists of Al2O3 core surrounded by a liquid calcium aluminate layer, because Ca is transferred from slag to steel, promoting liquid oxide phase formation.
