Elsevier

Acta Materialia

Volume 56, Issue 15, September 2008, Pages 3900-3913
Acta Materialia

The fracture toughness of TRIP-assisted multiphase steels

https://doi.org/10.1016/j.actamat.2008.04.035Get rights and content

Abstract

The relationship between microstructure and fracture resistance of TRIP-assisted multiphase steels has been investigated by processing and characterizing eight specifically designed microstructures differing in terms of the volume fractions of the constitutive phases, i.e. ferrite, bainite, retained austenite and martensite, by the stability of the retained austenite, and by the connectivity of the phases. Fracture resistance has been quantified by the fracture strain in uniaxial tension, ε¯f, the fracture toughness at cracking initiation, JIc, and by the essential work of fracture, we. The fracture mechanisms were characterized by microscopic observations and by profilometric measurements of the crack tip necking process typical in thin sheet fracture. The fracture toughness at cracking initiation of TRIP-aided steels corresponds to an equivalent Dual Phase steel in which martensite replaces retained austenite and is thus not affected by the retained austenite stability. The ductile tearing resistance of TRIP-aided steels, quantified by we, increases significantly with the retained austenite stability. This improved tearing resistance is explained by the effect of the additional work-hardening induced by the late transformation of retained austenite within the crack tip necking zone.

Introduction

Many applications requiring good formability and high structural performances require increasing levels of both strength and ductility, properties that are, for most materials, mutually exclusive. The success of TRIP-assisted multiphase steels comes from their potential to reconcile these two properties. TRIP-assisted multiphase steels consist of an intercritical ferrite matrix with retained austenite, bainite and martensite as dispersed phases [1], [2]. The high performances of these steels result both from the multiphase character of their microstructure and from the TRIP effect, i.e. the martensitic transformation of the retained austenite induced during deformation [3]. As a consequence, both strength and uniform strain increase due to the appearance of a harder phase and to the additional local plastic yielding of the surrounding grains related to the transformation strain [4], [5], [6], [7], [8], [9], [10], [11], [12]. Most of the efforts undertaken over the last 10–15 years to optimize heat-treatments and chemical compositions have focussed on enhancing the combination of these two properties [1], [2], [3], [13], leading to industrial production of several grades and allowing significant weight savings in automotive applications.

Nevertheless, several problems related to limited fracture resistance are mentioned repeatedly, including severe notch and defect sensitivity leading to fracture during forming [14] and limited hole-expansion capacity [15]. Improving the fracture resistance without deteriorating the good strength/ductility balance is thus key to enlarging the range of applications of TRIP-aided steels. From a fundamental viewpoint, outstanding issues have not yet been clarified about the relationships between the fracture toughness, the stress-state-dependent, mechanically induced martensite transformation, and the damage mechanisms taking place within an evolving multiphase microstructure. The case of the ceramics toughened by stabilized zirconia particles which transform under loading has been studied in detail [16], [17]. Numerous investigations on the fracture resistance of fully austenitic TRIP steels [18], [19], bainitic steels [20] and austempered ductile cast irons [21], [22] have also been carried out. These studies show that the presence of retained austenite can be beneficial to the fracture toughness, which is the opposite to what has been reported for TRIP-aided multiphase steels [23], [24].

The objective of the present work was to address the fracture resistance of TRIP-aided steels by systematically investigating the effect of the retained austenite volume fraction and stability. The first preliminary task was to produce representative microstructures. For that purpose, six TRIP-aided multiphase steels were processed from three steel grades with different carbon contents. Two Dual Phase steels were also heat-treated for comparison. The second preliminary task was to properly define the fracture resistance. It is essential to realize from the outset that, in the present material, the product of the ultimate tensile strength and the uniform ductility (which is frequently called “toughness” by metallurgists) is by no means a good indicator of the fracture toughness, defined as the resistance to the initiation of cracking from a pre-existing defect. Different parameters characterizing the fracture resistance have been measured and assessed: (i) the fracture strain in a uniaxial tensile test (εf); (ii) the fracture strain in front of a propagating crack tip (R); (iii) the fracture toughness at cracking initiation (JIc); or (iv) equivalently, the critical crack tip opening displacement (δc); and (v) the ductile tearing resistance measured by the essential work of fracture (we).

An extra complexity arises when considering the fracture of thin1 sheets: cracking is accompanied by a significant amount of necking entering the fracture process zone (FPZ) [25], [26], [27]. Hence, the cracking resistance Γc can be seen as the sum of the energy spent per unit area of crack advance for damage and material separation within the FPZ, Γ0, and of the work per unit area of crack advance spent in plastic deformation within the neck Γn, i.e. Γc = Γ0 + Γn, even though these two terms are highly convoluted. Γn depends on Γ0: a larger resistance to damage leads to a larger thickness reduction in the neck before fracture. The work of unit surface spent in necking, Γn, can sometimes constitute the major part of the fracture toughness [25], [26], [27]. Furthermore, it depends on the thickness of the plate, leading to the thickness dependence of the fracture toughness [25], [26]. For that reason, the sheet thickness has been kept constant in this study. The fracture resistance at cracking initiation JIc and during steady-state propagation Γss can be specifically writtenJIc=Γ0+Γninit(Δa=0)andΓSS=Γ0+ΓnSS(Δat),respectively, where Δa is the crack advance and t is the plate thickness. The steady-state regime is attained when the crack has propagated several times the plate thickness, and the crack tip neck is no longer evolving. Γss is larger than JIc due to the increase in the term Γn during crack propagation. This increase in the tearing resistance is a real increase when compared to the usual increase in J observed in a JR curve which is, in the plane strain regime, extrinsic [28], [29]. This increase comes from the necking that keeps developing during crack propagation due to the evolution of the stress distribution whose intensity decreases during propagation, [30], and due to the tunnelling effect of the crack propagating more slowly near the outer surfaces. As advocated in many contributions in the literature [26], [31], [32], [33], the so-called essential work of fracture, we, is a suitable parameter to quantify Γss in thin sheets. In-depth analysis of these various parameters is essential to reach a complete understanding of the fracture resistance of ductile thin plates.

The outline of the paper is as follows. The materials and the experimental procedures are described in Section 2. Section 3 presents the experimental results, which are discussed in Section 4. The discussion is separated in three parts: (i) the active fracture mechanisms, (ii) the relationships between crack tip necking and transformation rate and (iii) the influence of the TRIP effect on the ductile tearing.

Section snippets

Chemical compositions and heat-treatments

The microstructure of the TRIP-assisted multiphase steels results from a two-step heat-treatment after cold-rolling [3]. The first step consists of an intercritical annealing that results in a ferrite–austenite mixture. The ratio between austenite and ferrite depends on the chemical composition and holding temperature. The second step consists of a bainite holding conducted at a temperature between 300 and 400 °C. During this second isothermal holding, part of the austenite transforms into

Uniaxial tension

Fig. 5a presents the true stress–true strain curves measured up to necking and extrapolated up to the fracture strain for the various steel grades and heat-treatments. The fracture stress is an average tensile stress approximated by the load divided by the area of the fracture surface. As expected, the strength of the TRIP-aided steels increases with the amount of carbon. The uniform and fracture strains are given in Fig. 5b as a function of the initial amount of retained austenite or

Cracking initiation, fracture toughness and fracture mechanisms

The TRIP effect is known to improve both the strength and the uniform strain of multiphase steels [4], [5], [6], [7], [8], [9], [10], [11], [12] so that the product of these two properties is often used for assessing the performance of different grades or specific heat-treatment conditions. This fact is confirmed by Fig. 5a which shows that TRIP1 steels present better plastic flow properties due to a more effective TRIP effect prior to the onset of necking. Indeed, the austenite in these

Conclusions

The main conclusions emerging from the comparison of different TRIP-assisted multiphase steels and DP steels investigated in this paper are the following:

  • The damage process starts by the cracking of the interface between martensite islands and the ferrite matrix, which means that, for TRIP-aided steels, damage occurs after the austenite transformation. Hence, any future attempt to model damage nucleation and growth in TRIP steels must first rely on proper modelling of the phase transformation

Acknowledgments

G.L. acknowledges the financial support of ArcelorMittal Research. P.J.J. acknowledges the FNRS and FRFC (Belgium). Fruitful discussions with ArcelorMittal Research and J.D. Embury are gratefully acknowledged. This work was partly carried out in the framework of the IAP program of the Belgian State Federal Office for Scientific, Technical and Cultural Affairs, under Contract No. P6/24.

References (46)

  • K. Spencer et al.

    Mater Sci Eng A

    (2004)
  • M.H. Saleh et al.

    J Mater Process Technol

    (2001)
  • F. Lani et al.

    Acta Mater

    (2007)
  • P.J. Jacques et al.

    Acta Mater

    (2007)
  • P.M. Kelly et al.

    Prog Mater Sci

    (2002)
  • D.L. Porter et al.

    Acta Metall

    (1979)
  • K.J. Kim et al.

    Mater Sci Eng

    (1978)
  • P. Jacques et al.

    Acta Mater

    (2001)
  • J.H. Chen et al.

    Acta Metall

    (1984)
  • T. Pardoen et al.

    Eng Fract Mech

    (2002)
  • T. Pardoen et al.

    J Mech Phys Solids

    (2004)
  • J. Rice et al.

    J Mech Phys Solids

    (1978)
  • Y.W. Mai

    Int Mech Sci

    (1993)
  • B. Cotterell et al.

    Eng Fract Mech

    (2005)
  • P.J. Jacques et al.

    Mater Sci Eng

    (1999)
  • P.J. Jacques

    Curr Opin Solid State Mater Sci

    (2004)
  • E. Girault et al.

    Mater Charact

    (1998)
  • T. Pardoen et al.

    Eng Fract Mech

    (2000)
  • R.M. McMeeking

    J Mech Phys Solids

    (1977)
  • T. Pardoen

    Comput Struct

    (2006)
  • Y. Sakuma et al.

    Metall Trans A

    (1992)
  • K. Sugimoto et al.

    ISIJ Int

    (1992)
  • P. Jacques et al.

    Metall Trans A

    (1998)
  • Cited by (0)

    View full text