A physically based model for TRIP-aided carbon steels behaviour

https://doi.org/10.1016/S0921-5093(03)00121-7Get rights and content

Abstract

A physically based model for TRIP carbon steels is developed suitable to predict the macroscopic behaviour of multi-constituent aggregates. It includes the effects of phase composition and morphology on flow stress and strain hardening. In a first part, a detailed description of the stress-assisted and strain-induced martensitic transformation kinetics is given based on a generalised form of the Olson–Cohen model. The appearance of the much harder martensitic phase during plastic straining gives rise to a strong hardening of the retained austenite islands. The matrix behaviour is described using a model previously developed for ferritic–martensitic steels. A quite simple but accurate homogenisation approach is used to determine the TRIP steel behaviour. The predicted evolution of strain-induced martensite volume fraction, flow stress and incremental work hardening is in good agreement with experimental data and illustrates the critical importance of the retained austenite stability on the formability of TRIP steels.

Introduction

Requirements for increased formability and weight reduction in the automotive industry has largely contributed to the development of new TRIP-aided ferrous alloys. These steels exhibiting an excellent combination of high strength and ductility have a multiphase microstructure consisting of ferrite, bainite, sometimes “thermal” martensite and metastable austenite islands. This retained austenite transforms into martensite during plastic straining generating a composite deformation behaviour of the major phases and eventually a minor transformation induced plasticity (TRIP) effect which contributes to the high work-hardening rate of the microstructure and delays the onset of necking [1], [2].

As reported by Haidemenopoulos [3], the different parameters that influence the mechanical stability of the dispersed retained austenite and therefore the ductility of TRIP steels are numerous and often coupled, mainly: chemical composition, grain size and stress state of the surrounding matrix. The optimisation of such complex multiphase steels requires a detailed understanding and modelling of the mechanisms of phase transformation during mechanical testing.

The aim of the present study is to propose a physically based model of the macroscopic mechanical behaviour of TRIP-assisted multiphase steels which takes into account the composition and the morphology of each constituent. In a first part, a detailed description of the stress-assisted and strain-induced martensitic transformation kinetics is given. The subsequent hardening mechanism due to the appearance of the new hard phase in the retained austenite is taken into account. The matrix behaviour is described using a model previously developed for ferrite-martensite steels [4]. Finally, a simple but accurate homogenisation approach is used to predict the TRIP steel behaviour.

Section snippets

Strain-induced martensitic nucleation and transformation kinetics

Spontaneous martensitic transformation on pre-existing nucleation sites occurs during cooling below the MS temperature. At temperatures just above MS, the nucleation of the transformation can be stress-assisted at the same sites at increasingly higher stress for increasing temperatures (Fig. 1a). In this regime, transformation can be modelled by incorporating the thermodynamic effect of the applied stress in the theory developed for the cooling transformation (Fig. 1b).

Above the MSσ

Stress–strain behaviour of the ferrite–martensite–bainite matrix

Part of the matrix surrounding austenite islands is composed of a ferrite “thermal” martensite mixture, whose behaviour has previously been modelled for dual-phase steels [4]. This model gives the dual-phase stress–strain relationship σDP–εDP:σDP=σ0+αMμb1−exp&0xEB04;(−fM(εDP+ε0))fd+6fmdm1−exp&0xEB04;(−rM(εDP+ε0))r,where fm is the martensite volume fraction in the matrix, dm the average martensite island diameter, d the average ferritic grain size, μ=80 GPa, b=2.5×10−10 m, M=3, α=0.4, r=4.5 and ε

Stress–strain behaviour of the TRIP steel

The macroscopic behaviour of TRIP steels (Σ, E) is assumed to obey an intermediate mixture law [28] for the matrix and the metastable austenite inclusion:Σ(E)=(1−Fγ,0)σm(εm)+Fγ,0σγ(ε)E=(1−Fγ,0)εm+Fγ,0ε,where Fγ,0 is the initial retained austenite volume fraction. The Iso-W hypothesis was again applied to determine the partition of deformation and stress between the austenite and the matrix:σmdεm=σγdε.

The hardening parameter n of the TRIP steel is determined as follows:n=EΣdΣdE.

Its evolution

Conclusions

A modelling for TRIP carbon steels has been developed suitable to predict the macroscopic behaviour of multi-constituent aggregates. This physically based model includes the effects of phase composition and morphology on flow stress and strain-hardening.

1. The strain-induced and stress-assisted martensitic transformation is found to be very sensitive to changes in austenite grain size. It is also a function of the chemical composition of retained austenite and especially the carbon content.

2. The

References (31)

  • R.G. Stringfellow et al.

    Acta Metall. Mater.

    (1992)
  • J.R. Patel et al.

    Act. Met.

    (1953)
  • G. Ghosh et al.

    Mater. Sci. Eng.

    (1986)
  • R. Datta et al.

    Mater. Sci. Eng.

    (1982)
  • D.L. Bourell et al.

    Acta Metall.

    (1983)
  • Y. Estrin et al.

    Acta Met.

    (1984)
  • P. Jacques et al.

    Phil. Mag. A

    (2001)
  • K.D.H. Bhadeshia

    ISIJ Int.

    (2002)
  • G.N. Haidemenopoulos, Dispersed-phase transformation toughening in ultra high-strength steels, PhD Thesis,...
  • O. Bouaziz et al.

    J. Phys. IV

    (2001)
  • G.B. Olson et al.

    Metall. Trans.

    (1975)
  • G.B. Olson et al.

    Metall. Trans.

    (1982)
  • V. Raghavan et al.

    J. Iron Steel Inst.

    (1965)
  • G.B. Olson et al.

    Metall. Trans.

    (1976)
  • L. Kaufmann et al.

    Progress Met. Phys.

    (1958)

    • Cited by (0)

    View full text
    Copyright © 2003 Elsevier Science B.V. All rights reserved.