Thermodynamic, Kinetic, and Microstructure Data for Modeling Solidification of Fe-Al-Mn-Si-C Alloys

Thermodynamic, Kinetic, and Microstructure Data for Modeling Solidification of Fe-Al-Mn-Si-C Alloys JYRKI MIETTINEN, SAMI KOSKENNISKA, VILLE-VALTTERI VISURI, MAHESH SOMANI, TIMO FABRITIUS, and JUKKA KÖMI In this study, a set of thermodynamic, kinetic, and microstructure data is presented to simulate the non-equilibrium solidification of Fe-Al-Mn-Si-C alloys. The data were further validated with the experimental measurements and then used in a thermodynamic–kinetic software, IDS, to establish the effect of the alloying and cooling rate on the solidification behavior of high-AlMnSi (Al ‡ 0.5 wt pct, Mn ‡ 2 wt pct, Si ‡ 1 wt pct) steels. The modeling results were additionally validated by conducting electron probe microanalysis (EPMA) measurements. The results reveal that (1) solidification in high-AlMnSi steels occurs at much lower temperatures than in carbon steels; (2) increasing the cooling rate marginally lowers the solidus; (3) the microsegregation of Mn in austenite is much stronger than that of Si and Al due to the tendency of Al and Si to deplete from the liquid phase; (4) the residual delta ferrite content may be influenced by a proper heat treatment but not to the extent that could be expected solely from thermodynamic calculations; (5) in high-AlMnSi steels containing less than 0.2 wt pct carbon, the cracking tendency related to the strengthening above the solidus and the shell growth below the solidus may be much lower than in carbon steels. https://doi.org/10.1007/s11663-020-01973-y  The Author(s) 2020 I. INTRODUCTION ADVANCED High-Strength Steels (AHSS) belonging to the family of Fe-Al-Mn-Si-C alloys have been extensively studied due to their high strength and good formability.[1] To control the continuous casting process, it is necessary to have a thermodynamic–kinetic software that can reproduce and interpolate measurement data with high accuracy. Modern solidification models apply computational thermodynamics and kinetic equations along with corresponding databases.[2] The reliability and self-consistency of the thermodynamic descriptions are especially important for the optimization routines. Furthermore, in online applications, the computational expense of the thermodynamic–kinetic description should be reasonably low, especially in 3D modeling applications. JYRKI MIETTINEN, VILLE-VALTTERI VISURI, and TIMO FABRITIUS are with the Process Metallurgy Research Unit, University of Oulu, PO Box 4300, 90014 Oulu, Finland. Contact e-mail: ville-valtteri.visuri@oulu.fi SAMI KOSKENNISKA, MAHESH SOMANI, and JUKKA KÖMI are with the Materials and Mechanical Engineering Research Unit, University of Oulu, PO Box 4200, 90014 Oulu, Finland. Manuscript submitted April 14, 2020; accepted September 6, 2020. METALLURGICAL AND MATERIALS TRANSACTIONS B The first aim of this investigation was to outline the necessary thermodynamic, kinetic, and microstructure data to conduct the thermodynamic–kinetic simulations for Fe-Al-Mn-Si-C alloys. To validate the modeling results, electron probe microanalysis (EPMA) measurements were taken. Finally, simulations were performed to investigate the solidification behavior of high-AlMnSi steels as a function of their compositions and cooling rate/s. Also simulated, below the solidus, were the ferrite/austenite transformations and the solute microsegregation, including the determination of the soluble grain boundary compositions. As these compositions, instead of the nominal ones, are expected to control the start of austenite decomposition,[3] they will play an important role in a later study, in which we plan to extend the current simulation work on high-AlMnSi (Al ‡ 0.5 wt pct, Mn ‡ 2 wt pct, Si ‡ 1 wt pct) steels to their austenite decomposition process. These simulations will apply new continuous cooling transformation (CCT) equations, which take into account the Al alloying that was not considered in the previously optimized CCT equations of Miettinen et al.[3] A. IDS Tool The developed descriptions are implemented in the IDS software,[47] which is a thermodynamic–kinetic software for the simulation of phase change, compound formation/dissolution, and solute distribution during the solidification of steels and their cooling/heating process after solidification. The package also simulates the solid-state phase transformations related to the austenite decomposition process below 900 C (1173 K) and calculates important thermophysical material properties (such as enthalpy, thermal conductivity, and density) from the liquid state to room temperature. The calculations of the IDS tool have been compared with numerous solidification-related measurements that show generally good agreement. Coupled with a suitable heat transfer model,[8] the IDS software is applicable for the online simulation of the continuous casting process. Assuming complete solute mixing in the liquid and a regular dendritic structure, the calculations can be made in one volume element set on the side of a dendrite arm (Figure 1). At the same time, of course, no solute exchange is allowed between the volume element and its surroundings. Using a hexagonal arm arrangement for the dendrites,[9] the volume element assumes the form of an equilateral triangle when looking perpendicularly to the dendrite arm growth. All calculations are made stepwise,[6] decreasing the temperature in the liquid region in steps of 1 C, decreasing the liquid fraction in the mushy zone by the 67 steps into which the volume element is divided, and decreasing or increasing the temperature in the solid state (below the solids) in steps of 1 C, depending on whether the steel is cooled or heated. Other assumptions simplifying the calculations are (1) the thermodynamic equilibrium holds good at the solution phase interfaces; (2) the diffusion of solutes is independent of the chemical effect of other solutes; (3) differences in the molar volumes of the phases are negligible; (4) during solidification, both ferrite and austenite begin to form as soon as it is thermodynamically possible. Depending on the composition, the solidifying Fe-Al-Mn-Si-C alloys go through one of the following solidification paths down to 900 C (1173 K) prior to the austenite decomposition: (A) L fi L + a fi a fi c + a(fi c), (B) L fi L + a fi L + c + a fi c + a (fi c), (C) L fi L + c fi c, where L denotes liquid, a denotes ferrite, and c denotes austenite. The mutual order of the phases in the above paths shows how the phases are located in the volume element of Figure 1, from the dendrite arm axis (left) to the interdendritic region (right). In the two-phase regions of L + a, L + c, and c + a, subsequent transformations of L fi a, Lfic, and a fi c, respectively, occur during cooling, and in the threephase region of L + c + a, a peritectic transformation of L + a fi c occurs. Note that the solid structure at 900 C (1173 K) may contain the austenite and ferrite, and not necessarily only the austenite (see paths A and B). Three thermodynamic model (...truncated)