Abstract

    A computer model has been created to calculate the high temperature mechanical properties and density change of steels during solidification. Such calculations were based on accurate thermodynamic description of the phase evolution, including changes in phase fraction and element concentration as a function of temperature and steel composition during the solidification process. The properties of each phase were firstly calculated based on its composition and temperature. Then the strength and density of the overall material were calculated via a mixture law. Results show that the calculated tensile strength and density values are in good agreement with experimental results in liquid, δ-ferrite, and γ-austenite single phase regions and mixed phase regions. The computer model is designed in such a way that all the calculations can be done automatically via a user friendly graphical interface when the alloy composition is given.

Introduction

    Many researchers have studied high temperature mechanical properties because these are necessary for the prediction and control of stresses in the solidified shell of steel castings.1,)2 Density is another important factor required to optimise the conditions for continuous casting and any simulation of heat-conduction, solidification, elast-plastic deformation and fluid flow in many processes.3,)4 Recently, Mizukami and co-workers attempted to formulate quantitative relationships for the calculation of tensile strength and density based on alloy composition and the phases existing during the solidification process.5,,)67 A brief description of their approach is given below using density as an example. First, thermo-dynamic calculation was used to calculate the phase fraction as a function of temperature during solidification. Then linear regression analysis was carried out to correlate the density of each phase with the temperature difference ΔT from a characteristic temperature (e.g. liquidus for the liquid phase) through optimisation against experimental measurements. Finally a linear mixture law was applied to obtain the overall density based on that of each phase. Although such models showed good agreement with their experimental data for both density and strength, the fact that no composition dependency was considered for the properties of each phase significantly limited its application to steel types outside the studied composition region. Their assumption that "the density of a phase has almost the same value when ΔT is the same" may be reasonable for the steels studied in their work, i.e. carbon steels with Si+Mn ≤ 1.06 wt%, it would not be applicable to alloy steels. Also such calculation cannot be done easily for new alloys as the models require inputs such as phase fraction, solidus (TS) and liquidus (TL) temperatures that are not readily known for new alloys.

    This paper reports our recent development of material models for the calculation of the density change and high temperature strength during solidification of steels. The present models have taken the dependency of density and strength on phase composition into account and therefore can be applied to steels of a wide composition range instead of just carbon steels. Many other important physical and thermophysical properties necessary for process modelling can be calculated as well, such as thermal expansion coefficient and Young’s modulus. For the case of strength calculation, the effect of strain rate on flow stress is predicted, as is also the composition dependence. The major advantage of the present approach is that it is far less costly and has produced calculations within useful accuracy.

    The first part of the paper describes the model development for the calculation of density and high temperature strength. The second part features the application of these models to the alloys studied by Mizukami et al. It should be noted that development of the present models had been completed well before the authors became aware of the work by Mizukami et al. and their data are used here solely for the purpose of testing the performance of the present models.

``````