Available online at www.sciencedirect.com

SCIENCE/W) DIRECT®

Current Opinion in

Solid State &

Materials Science

ELSEVIER

Current Opinion in Solid State and Materials Science 8 (2004) 259-265

Transformation-induced plasticity for high strength formable steels

P.J. Jacques *

De´partement des Sciences des Mate´riaux et des Proce´de´s, Universite catholique de Louvain, IMAP, Place Sainte Barbe 2,

A45 B-1348 Louvain-la-Neuve, Belgium

Received 31 July 2004; received in revised form 8 September 2004; accepted 14 September 2004

Abstract

Recent advances in the development of high performance steels presenting improved properties of strength and ductility rely on the TRIP effect, i.e. on the mechanically-induced martensitic transformation of the retained austenite dispersed in a soft ferrite-based matrix. As a consequence, the stabilisation and retention of austenite at room temperature have become of primary importance, leading to specifically designed steel grades and thermal or thermomechanical treatments. Particularly, carbon enrichment of the austenite during intercritical annealing and bainite transformation was found to be very effective in retaining austenite. This meta-stable austenite then progressively transforms during straining, bringing about a large increase of the work hardening rate. This increase results from the stress and strain partitioning continuously evolving with the appearance of the hard martensite. © 2004 Elsevier Ltd. All rights reserved.

1. Introduction

Over the years, several solutions have been imagined to improve the mechanical properties of iron and steel in order to meet the requirements of more and more strin­gent applications. Indeed, depending on their chemical compositions, Fe-C based alloys present the exceptional feature that the processing route can be adapted to lead to various phases that exhibit antagonist mechanical properties ranging from soft and ductile ferrite to ultra high strength martensite. Furthermore, these phases can be combined within finely grained microstructures considered as in situ composites.

Among the different deformation mechanisms, the martensitic transformation of austenite during mechan­ical loading has been known for quite a long time [1-3]. The acronym 'TRIP' (for TRansformation-Induced Plasticity) was proposed to express the efficiency of the martensitic transformation as a deformation mech-

Tel.: [phone]; fax: [phone]. E-mail address: [e-mail]

anism [4-6]. Indeed, this phenomenon, by bringing about large enhancements of the work-hardening rate, postpones the onset of necking and thus improves the formability.

The TRIP effect has recently regained attention in the case of low alloy steels [7]. The design of new steel grades and microstructures is indeed motivated by the necessity for the steel industry to process always better suited high strength structural steels with low produc­tion costs. Responding to an unceasing demand from the automotive industry for steels with strength level higher than 500 MPa and up to 1000MPa without sac­rificing the formability properties, the 1990s have seen the development and characterisation of new formable high strength steel grades, the so-called TRIP-assisted multiphase steels [8]. As illustrated in Fig. 1, these steels present complex multiphase microstructures consisting of a ferritic matrix and a dispersion of multiphase grains of bainite, martensite and metastable retained austenite.

Once the potential improvements brought by these TRIP-aided steels demonstrated, research focussed on two main axes:

1359-0286/$ - see front matter © 2004 Elsevier Ltd. All rights reserved, doi: 10.1016/j.cossms.2004.09.006

260

P.J. Jacques / Current Opinion in Solid State and Materials Science 8 (2004) 259-265

Fig. 1. SEM (a) and EBSD (b) micrographs illustrating the typical microstructure of the TRIP-assisted multiphase steels ((a) Nital etching; (b) phase map (BCC in grey, FCC in white)).

(i) the design of processing routes bringing about finely grained micro structures containing the right amount of austenite with the adapted stability, (ii) the understanding of the work hardening and mechanical properties of composite microstructures exhibiting a TRIP effect.

The TRIP-aided steels actually correspond to a muta­tion in the way of looking at retained austenite. Indeed, while large efforts were made to avoid the presence of re­tained austenite, particularly in welds, for embrittlement and fracture reasons, recent studies dealt with the max­imisation of the amount of retained austenite presenting the right stability [9]. On the other hand, it is also of primary importance to understand the unique combina­tions of strength and ductility exhibited by the TRIP-aided steels. Contrarily to the full TRIP steels developed in the 1960s and 1970s [10-12], they indeed present the unique feature that the TRIP effect occurs for small

austenite grains dispersed in a soft ferritic matrix. It is thus required to characterise not only the TRIP effect but also the composite strengthening effect emerging from these complex microstructures.

It is proposed hereunder to focus on these two axes, i.e. to present the current general principles governing the processing of multiphase microstructures containing the appropriate austenite and to highlight the main factors governing their mechanical properties.

2. Processing of TRIP-assisted multiphase steels

In the present case of the TRIP-assisted multiphase steels, the partial stabilisation of austenite at room tem­perature is ensured by its carbon content, which is one of the strongest austenite stabiliser elements. As shown in Fig. 2, the austenite C enrichment occurs all along specifically designed thermal or thermomechanical cy­cles. After cold or hot rolling, the general morphology of the multiphase microstructure, i.e. the dispersion of ferrite and austenite grains, results from an intercritical annealing. The first carbon enrichment of the austenite accompanies the nucleation and growth of the phases. Several studies on the formation of the ferrite/austenite mixture during intercritical annealing have been carried out in the 1970s and 1980s [13-15] in the case of Dual Phase steels. However, the maximum austenite C enrich­ment during intercritical annealing does not prevent the martensitic transformation on quenching to room tem­perature. A second stage for further carbon enrichment is therefore needed.

Fig. 2. Schematic representation of the thermomechanical treatments applied to hot or cold rolled TRIP-assisted multiphase steels (c: austenite, a: ferrite, a⁰: martensite, a_b: bainite).

As shown in Fig. 2, a second hold at an intermediate temperature combined with particular alloying elements allow further C enrichment and the stabilisation of aus­tenite at room temperature [16,17]. It is well known that

P.J. Jacques / Current Opinion in Solid State and Materials Science 8 (2004) 259-265

261

Temperature (°C)

in the course of the bainite transformation, carbon redistributes from the bainitic ferrite to the surrounding austenite [18-22], leading to the characteristic bainitic morphology of ferrite/carbide mixture. Furthermore, the addition of some alloying elements is known to promote the formation of carbon-saturated austenite instead of cementite precipitation. Bainite transforma­tion has thus become of primary importance in the processing of the TRIP-aided steels.

Recent work [9] has shown that the bainite transfor­mation occurring in the case of the TRIP-aided steels presents some peculiarities. Current issues therefore in­volve the influence on the austenite retention at room temperature (i) of the alloying elements [16,17], (ii) of the austenite grain size [23] and (iii) of the austenite prior deformation [24].

Beside carbon (from 0.1 to 0.4wt.%), and manganese (0.5 to 1.5wt.%, providing some hardenability), a real challenge is to find the right alloying element that delays or suppresses the cementite precipitation during the bai­nite transformation. It is well known that silicon inhibits cementite precipitation [21,25]. However, the high sili­con levels that are needed do not fit well with the indus­trial practice of galvanised flat products [26]. Aluminium has also been shown to inhibit effectively the cementite precipitation and thus to promote the austenite carbon enrichment [17,27,28]. Some studies dealt with other ele­ments like Cu [29], P [29,30] or Ni [31]. However, hardly anything can be found in the literature on the real mech­anism by which some alloying elements inhibit this cementite precipitation.

Fig. 3(a) illustrates the evolution of the nature of the phases constituting the room temperature microstruc-ture of a typical Si-alloyed TRIP-aided steel as a func­tion of the progress of the bainite transformation. It is worth noting that Al- or Al-Si alloyed grades exhibit the same transformation behaviour [17]. Depending on the bainitic hold time, various mixtures of bainite, mar-tensite and retained austenite can be found in the micro-structure after quenching. The intercritical austenite progressively transforms to bainite, bringing about a large stabilisation of austenite at room temperature at the expense of martensite. As also shown in Fig. 3(a), this stabilisation is due to the austenite C enrichment operating up to a maximum depending on the tempera­ture at which the bainite transformation takes place.

Some maximum austenite C enrichments measured when the bainite transformation stops at different tem­peratures for a 1.5Si-1.5Mn grade are given in Fig. 3(b), together with the T₀ and Ae₃ curves calculated by the Calphad method. The T₀ curve represents the upper bound of the austenite carbon content allowing a composition-invariant transformation of austenite to ferrite (i.e. where austenite and ferrite of the same com­position present the same chemical free energy). Fig. 3(b) shows a perfect agreement between the T₀ curve

Fig. 3. (a) Transformation map and austenite C enrichment during the bainitic hold of a classical TRIP-assisted multiphase steel; (b) Comparison with the calculated T₀ and Ae3 curves of the measured maximum carbon content of retained austenite of a TRIP-aided steel held at several temperatures. The shaded box corresponds to the range of carbon content bringing about an optimised mechanical stability of the retained austenite.

and the maximum carbon content of retained austenite. The behaviour of these Si- and Al-alloyed steels can thus be explained by considering (i) the displacive mechanism of bainite formation, (ii) the carbon partitioning be­tween bainitic ferrite and residual austenite and (iii) the inhibition of cementite precipitation from austenite [20,22,32].

A shaded box has also been represented on Fig. 3(b). It corresponds to the range of carbon content bringing about an optimised mechanical stability of austenite [33]. Indeed, it is worth remembering that the bainite transformation aims at making the remaining austenite more stable so that it will progressively transform dur­ing the subsequent forming operations and thus improve

262

P.J. Jacques / Current Opinion in Solid State and Materials Science 8 (2004) 259-265

the mechanical properties through the TRIP effect. Sev­eral experimental [7,34] and now theoretical [35] studies showed that the austenite carbon content must lie within the given range for the right effectiveness of the mechanically-activated martensitic transformation. As a consequence, carbon enrichment also delineates the temperature range in which the bainite transformation can be conducted as a function of the T₀ curve and thus of the chemical composition of the grade.

Beside specific chemical compositions, another fea­ture highlight the bainite transformation occurring in the TRIP-aided steels, the austenite grain size. Indeed, as shown in Fig. 1, the austenite grains classically pres­ent a very fine size of the order of 1 or 2 lm. Fig. 4 com­pares the bainite morphology as a function of the size of the austenite grain in which it occurs. The classical bai­nite sheaf structure can be clearly seen on Fig. 4(a). The first bainitic ferrite sub-units nucleate at the austenite grain boundary and grow towards the interior of the austenite grain. New sub-units then nucleate and grow from the tip of the previous ones, bringing about the sheaf structure. This process is valid as long as tip nucle-

ation is possible, i.e. as long as the austenite grains are larger than the platelets length. By contrast, the bainite morphology is completely different when the austenite grain size is only a few micrometer as in the case of the intercritical austenite shown in Fig. 4(b). The bainite that forms in very small austenite grains presents adja­cent platelets that completely cross the austenite grain. These differences in the bainite morphology as a func­tion of the austenite grain size influence in a large way the bainite transformation kinetics. Fig. 5 presents the evolution of the normalised bainite content as a function of the transformation time for several austenite grain sizes. For smaller grains, the transformation starts ear­lier but proceeds at a slower rate. Indeed, the reduction of the grain size brings about an increase of the grain boundary area that accelerates the rate of transforma­tion thanks to an enhanced nucleation rate. This influ­ence was already shown and modelled by Rees and Bhadeshia [36]. However, when the austenite grain size is reduced to the length of one platelet, nucleation and growth of the next platelet at the tip of the previous one are no more possible. As a consequence, even if

Fig. 4. TEM micrographs and schematic representation of the growth process and resulting microstructures of bainite in the case of large (a) and small (b) austenite grains.

P.J. Jacques / Current Opinion in Solid State and Materials Science 8 (2004) 259-265

263

10 00

Bainitic Holding Time (s)

Fig. 6. True stress-true strain curves and evolution of the retained austenite content during plastic straining of several TRIP-assisted multiphase steels.

Fig. 5. Evolution of the normalised bainite volume fraction (bainite content with respect to the initial amount of austenite) as a function of the isothermal holding time for several austenite grain sizes.

the transformation starts earlier, the rate of transforma­tion is slower. The kinetics of bainite transformation in very small grains are therefore controlled only by the grain boundary nucleation rate.

3. Mechanical properties of TRIP-assisted multiphase steels

As already mentioned, the objective at the origin of the development of the TRIP-assisted multiphase steels is the improvement of the strength level without sacrific­ing the uniform elongation. However, these two proper­ties are quite antagonist ones. It is indeed very difficult to increase the stress level that can be sustained by a material without reducing its resistance to the localisa­tion of deformation and resulting fracture. These two properties of strength and deformation are actually re­lated to the work hardening rate, i.e. the increase of stress for an increase of strain, dr/de. At the microscopic scale, the work hardening rate results from the disloca­tion dynamics, i.e. the balance between the creation and annihilation of dislocations.

With respect to other steel grades and microstructures, the TRIP-aided steels allow to improve the work harden­ing rate by skillfully combining several mechanisms of strengthening and softening. Two mechanisms can effec­tively be argued in the case of the TRIP-aided steels:

(i) the TRIP effect, (ii) the composite-like nature of their microstructures.

First of all, the TRIP effect has a large influence on the resulting mechanical properties [34,37]. Fig. 6 presents

typical true stress-true strain tensile curves of TRIP-aided steels together with the evolution of the retained austenite content with true strain. These specimens pres­ent different initial amounts of retained austenite with distinct C enrichments. This Figure clearly shows that the best strength-ductility balance occurs for the speci­men presenting the largest TRIP effect distributed uni­formly all along plastic straining.

Furthermore, the occurrence of the TRIP effect for small austenite grains dispersed in a soft ferritic matrix has a large influence on the work hardening rate. Due to the shape and volume changes accompanying the transformation of austenite to martensite, local plastic­ity is generated in the surrounding ferrite grains [34]. The TEM micrograph of Fig. 7 illustrates the numerous accommodation dislocations generated within ferrite at the tip of the deformation-induced martensitic variants. With respect to the dislocations within the ferrite, the TRIP effect thus plays the role of an additional source that increases the plasticity properties. Quite simple cal­culations [34] showed a clear correspondence between the austenite transformation rate and the strength-duc­tility balance of the TRIP-aided steels.

However, the increase of the dislocation density can­not entirely explain the high strength level exhibited by the TRIP-aided steels. Another important aspect of these steels is the composite nature of their microstruc­tures. They indeed combine phases with antagonistic properties. Neutron diffraction has allowed the measure­ment of the yield strength of the different phases. Values of 500 MPa, 650 MPa, 900 MPa and 2000 MPa were found for ferrite, bainite, austenite and martensite, respectively [38]. As a result of this large variability of properties among the phases, stress and strain partition­ing occurs during loading and dictates the macroscopic stress-strain response [39,40]. Moreover, the present TRIP-aided steels constitute 'evolving composites' since

264

P.J. Jacques / Current Opinion in Solid State and Materials Science 8 (2004) 259-265

Fig. 7. BF and DF TEM micrographs illustrating the dislocations generated in the ferrite at the tip of the strain-induced martensitic variants.

Fig. 8. Comparison of the macroscopic stress and the stress calculated without taking into account the martensitic phase.

the proportions of austenite and martensite continu­ously change during straining and definitely modify the stress partitioning. As a conservative principle, the law of mixture states that the applied stress and strain are distributed among the different phases with respect to their volume fractions. In the case of stress, it writes

s) <T_a- (s

(1)

o[e) =

Thanks to neutron diffraction, it was possible to mea­sure the stress level for the BCC (ferrite (a) and bainite (a_b)) and FCC (austenite (c)) phases. Fig. 8 compares the measured macroscopic stress level with the calcu­lated stress for the 2 first terms of Eq. (1) (i.e. without taking into account of the martensite). The shaded area

thus corresponds to the strengthening resulting from the progressive appearance of this martensite.

4. Conclusion

In order to respond to stringent structural applica­tions, the TRIP-assisted multiphase steels have been developed through the control of complex phase trans­formations schemes. Thanks to the diversity of proper­ties profiles exhibited by the steel phases and the use of strain-induced phase transformation, unique combi­nations of strength and deformability can be obtained.

Acknowledgment

The author acknowledges the FNRS and the FRFC (Belgium). This work was partly supported by the Bel­gian Science Policy, within the framework of the PAI P5/08 project `<sup>`</sup>From microstructure towards plastic behaviour of single- and multiphase materials^''.

References

[1] Patel JR, Cohen M. Acta Metall 1953;1:531-8.

[2] Angel T. J Iron Steel Inst 1954:165-74.

[3] Olson GB, Cohen M. In: Antolovich SC, Ritchie RO, Gerberich WW, editors. Mechanical properties and phase transformations in engineering materials. New Orleans: Met. Soc. AIME; 1986. p. 367-90.

[4] Ludwigson DC, Berger JA. J Iron Steel Inst 1969:63-9.

[5] Magee CL, Paxton HW. Trans Metall Soc AIME 1968;242: 1741-9.

[6] Greenwood GW, Johnson RH. Proc Roy Soc London A 1965;283:403-22.

[7] Jacques PJ. In: Finel A, Mazie`re D, Veron M, editors. Thermody­namics, microstructures and plasticity. The Netherlands: Kluwer Acad. Publ.; 2003. p. 241-50.

P.J. Jacques / Current Opinion in Solid State and Materials Science 8 (2004) 259-265

265

[8] Zaefferer S, Ohlert J, Bleck W. Acta Mater 2004;52:2765-78. [9] Godet S, Georges C, Jacques PJ. In: Damm EB, Merwin MJ,

editors. Austenite formation and decomposition. Warrendale,

Pennsylvania: ISS & TMS; 2003. p. 523-36. [10] Zackay VF, Parker ER, Fahr D, Busch R. Trans Am Soc Met

1967;60:252-9.

[11] Olson GB, Cohen M. Metall Trans A 1975;6A:791-5. [12] Bhandarkar D, Zackay VF, Parker ER. Metall Trans 1972;3:

2619-31. [13] Cai XL, Garatt-Reed AJ, Owen WS. Metall Trans A 1985;16A:

543-51. [14] Speich GR, Demarest VA, Miller RL. Metall Trans A 1981;12A:

1419-28.

[15] Rigsbee JM, VanderArend PJ. In: Davenport AT, editor. Form-able HSLA and dual-phase steels. Warrendale: TMS-AIME

AIME; 1977. p. 56-86. [16] Jacques PJ, Girault E, Harlet P, Delannay F. ISIJ Int 2001 ;41:

1061-7. [17] Jacques PJ, Girault E, Mertens A, Verlinden B, Van Humbeeck J,

Delannay F. ISIJ Int 2001 ;41:1068-74. [18] Bhadeshia HKDH. Bainite in Steels. 2nd ed. London: IOM;

2001. pp. 189-200. [19] Bhadeshia HKDH, Christian JW. Metall Trans A 1990;21A:

767-97.

[20] Ohmori Y, Maki T. Mater Trans JIM 1991;32:631-41. [21] Takahashi M, Bhadeshia HKDH. Mater Trans JIM 1991;32:

689-96. [22] Jacques PJ, Girault E, Catlin T, Geerlofs N, Kop T, van der

Zwaag S, Delannay F. Mater Sci Eng A 1999;273-275:475-9. [23] Jacques PJ. J Phys 2003; 112:297-300.

[24] Godet S, Harlet P, Delannay F. Mater Sci Forum 2003;426-

4:1433-8. [25] Bhadeshia HKDH, Edmonds DV. Metall Trans A 1979;10A:

895-905. [26] Vanden Eynde X, Servais JP, Lamberigts M. Surf Int Anal

2003;35:1004-14. [27] Girault E, Mertens A, Jacques P, Houbaert Y, Verlinden B, Van

Humbeeck J. Scripta Mater 2001;44:885-92. [28] De Meyer M, Vanderschueren D, De Cooman BC. ISIJ Int

1999;39:813-22. [29] Traint S, Pichler A, Hauzenberger K, Stiaszny P, Werner E. Steel

Res Int 2002;73:259-66.

[30] Chen HC, Era H, Shimizu M. Metall Trans A 1989;20A:437-45. [31] Sakuma Y, Matlock DK, Krauss G. Metall Trans A 1992;23A:

1221-32. [32] Girault E, Jacques PJ, Ratchev P, Van Humbeeck J, Verlinden B,

Aernoudt E. Mater Sci Eng A 1999;273-275:471-4. [33] Jacques PJ. PhD thesis, UCL, 1998. [34] Jacques PJ, Furne´mont Q, Mertens A, Delannay F. Phil Mag A

2001;81:1789-812. [35] Van Rompaey T, Furnemont Q, Jacques PJ, Pardoen T, Blanpain

B, Wollants P. Steel Res Int 2003;74:365-9.

[36] Rees GI, Bhadeshia HKDH. Mater Sci Technol 1992;8:985-93. [37] Jacques PJ, Bell T. In: Cohen JN, editor. Proc Int Symp Honor

of Profs. Materials Park OH: ASM; 2000. p. 539-46. [38] Furne´mont Q. PhD thesis, UCL, 2003. [39] Furne´mont Q, Lacroix G, Godet S, Conlon KT, Jacques PJ. Can

Metal Quaterly 2004;43:35-42. [40] Jacques PJ, Ladrie`re J, Delannay F. Metall Mater Trans A

2001;32A:2759-68.

---