AGH University of Science and Technology A New Method for Determining Chemical Composition of Refining Slag in the Ladle Furnace Jan FALKUS, Tomasz KARGUL, Pawel DROZDZ AGH University of Science and Technology, Krakow, Poland Faculty of Metals Engineering & Industrial Computer Science Department of Ferrous Metallurgy

Contents Essential Process Parameters Assumptions to models development Static Model Dynamic Model Findings Conclusions

Essential Process Parameters Slag Metal Additions weight and chemical composition of the slag getting into the ladle during tapping oxygen activity before the refining process startup final chemical composition weight after tapping from the furnace start and final temperature start chemical composition oxygen activity before and after the refining process intensity and time of argon injection chemical composition after argon injection weight of the ladle additions put into the ladle during tapping and their chemical composition weight of the additions put into metal at the ladle furnace post and their chemical composition

Assumptions to static models development Based on thermodynamic model The state close to the metal-slag equilibrium is being achieved at the refining final phase The process time has no role

Thermodynamic model Equilib determines the combination of ni , Pi , and Xi which minimizes the total Gibbs energy G of the system Where: ni : moles Pi : gas partial pressure Xi : mole fraction i : activity coefficient : standard molar Gibbs energy

Testing heats

Static model findings for metallic phase Group I Group II Prognosis findings of the model calculations for the metallic phase

Static model findings for slag phase Prognosis findings of the model calculations for the slag phase total for group 1 & 2.

Static model. Virtual experiments. Al final content in the bath in relation to FeO in slag content Mn final content in the bath in relation to FeO in slag content Si final content in the bath in relation to FeO in slag content

Tools for computational calculations

Structure of Tank Model Assumptions to the dynamic model development tank I, M1 tank II, M2 tank III, M3 Division of the ladle volume into tanks in the process of bottom injection of argon into steel tank I (M1) - 25%  Mtotal tank II (M2) - 52.9% Mtotal tank III (M3) - 22.1% Mtotal Structure of Tank Model

Arrangement of equations describing mass flow The mathematical notation of the system of equations describing bath stirring. (Metal-slag interface reactions have not been taken into account in this model yet). mi – mass of reactant in ith reactor, [Mg] Mi – mass of ith elementary reactor, [Mg] – flow rate of metal bath stream between i and j reactors, [Mg/min] t – time, [min]

Structure of hybrid model Mixing model Input FactSage Output FactSage Thermodynamic model meq – equilibrium mass of reactant in metal-slag region, [Mg] The thicknesses of metal and slag layers which reach the state of mutual equilibrium depends on both the intensity of bath stirring and the calculation time interval value.

Dynamic model findings Change of Al content in metal bath during ladle furnace refining

Conclusion The ladle furnace is a unit, for which it is purposeful to create a static model in order to determine precisely the weight of the alloy additions. When the thermodynamic model, that uses the market available thermodynamic data bases, is applied in calculations, it makes it possible to determine the equilibrium state in very complicated metal-slag configurations. Complementing the static model with the dynamic one enables it to monitor the dynamics in change of metal and slag chemical compositions, and consequently to prepare better the metal bath for a continuous caster.

Thank You