ICAMS / Interdisciplinary Centre for Advanced Materials Simulation

Microstructure based modelling of aging kinetics and mechanical properties of dual phase steels


  • Quantification of microscopic structural measures in DP steels
  • Identification of BH properties as function of processing parameters
  • Development of a multi zone model to predict the BH effect
  • Modelling of multi-step aging kinetics by means of FE-RVE approach

  • Competing hardening and softening mechanisms in DP steels related to BH processing.

    Dual phase (DP) steels are characterized by a microstructure consisting of a ferritic matrix with dispersed martensite (5-20%). Such microstructure leads to appealing properties such as high strength, good ductility, continuous yielding and high work-hardening rate. The last step of the processing of these steels is the painting which occurs at a temperature range between 150-200 °C. At that point the steel undergoes an ageing process called Bake Hardening (BH) which is characterized by an increase in yield strength, the return of the yield point and a reduction in ductility. The bake hardening process is considered as a design criteria for these steels since they exhibit the highest BH response among other steels. The limited research that has been done in BH of DP steels suggests that the excess of strength in DP steels due to BH treatment is attributed to the high dislocation density that appears at the interface between ferrite and martensite. This density of Geometrically Necessary Dislocations (GND) at the interface is a result of the transformation of austenite to martensite. During BH carbon in ferrite migrates to newly created dislocations and locks them. Additionally the carbon segregates to grain boundaries and/or forms different kinds of carbide depending on the temperature of the process. While all these phenomena take place in ferrite, martensite goes through the process of tempering where carbon segregates, carbides precipitate and the structure of martensite changes. In this context it is possible that carbon diffuse from martensite into the GND area in the ferrite locking dislocations an assumption which is based on the fact that the area around martensite has an increased dislocation density (almost an order higher than that of ferrite) and the carbon available in ferrite for those dislocations is limited. Each of these processes may vary in intensity depending on the alloying elements, ageing conditions, quenching rate etc. The goal of this study is to investigate these competitive phenomena with respect to the process parameters (prestrain, aging time and aging temperature,) and specify the effects in the microstructure and by extension to the macroscopic properties of the steel.

    This investigation involves different types of modelling in order to accurately calculate the properties and simulate the phenomena taking place. In order to simulate the prestrain effect a Representative Volume Element (RVE) model will be applied in order to calculate the inhomogeneous strain field in the microstructure which will be used to calculate the local dislocation density in the microstructure. Since the driving force for diffusion of carbon differs for every process a modified model proposed by Balarin et al. [1] will be used to predict the kinetics of Cottrell atmosphere and a separate model for the precipitation, and carbon segregation to grain boundaries in order to take into account the grain size. A separate kinetic model will be applied to simulate the tempering of martensite along with the phase field model to calculate any carbon migration from martensite to ferrite. The inherently inhomogeneous properties of the microstructure of DP steels are further enhanced by the BH contribution. This strength profile will be fed to a commercial finite element program in order to simulate the macroscopic mechanical properties of the given alloy.

    [1] V. Ballarin, M. Soler, A. Perlade, X. Lemoine and S. Forest, Min., Met. & Matr. Soc., Vol. 40, pp. 1367-74, 2009

    « back