Effect of microstructure morphology on mechanical properties of quenching and partitioning steel

Abstract

In order to illuminate the relationship between microstructure morphology and final properties of the quenching and partitioning(Q&P) steel, the samples with different microstructure morphology (equiaxed and lamellar) and same volume fraction of each phase are obtained by controlling the initial microstructure and Q&P heat treatment. Because of the feature of microstructure morphology, a large yield ratio and total elongation are obtained in the lamellar sample though yield strength and ultimate tensile strength are relatively lower than that of equiaxed sample. Moreover, the lamellar sample produces a continuous work-hardening rate and better the fracture toughness compared to the equiaxed sample. Thus the lamellar sample is more suitable to be used as automotive structural components.

Keywords

Steel microstructure morphology mechanical properties

Introduction

In recent years, the development of new advanced high strength steels (AHSS) has gained wide attention in order to not only reduce vehicle weight and oil consumption but also maintain or even improve safety standards in the automobile industries for deteriorating environment. Speer et al. proposed a novel and promising AHSS after quenching and partitioning (Q&P) heat treatment [1,2]. The steel obtains a better comprehensive property [3]. The Q&P heat treatment consists of (1) an isothermal treatment at the intercritical temperature or above A_c3 for partial or full austenitisation, (2) quenching to a temperature (QT) between the martensite start temperature (M_s) and martensite finish temperature (M_f), at the QT, the microstructure consists of martensite and untransformed austenite, (3) a subsequent partitioning at the same or a higher temperature where carbon diffuses from supersaturated martensite to austenite, which will then stabilise the (retained) austenite and impede further transformation during the final quenching [4 –6]. The volume fraction of each phase is controlled by isothermal treatment, a thermal martensite transformation and isothermal partitioning stage, and their volume fractions and distributions determine the mechanical properties [7]. Such an approach creates a wide variety of microstructures and correspondingly large ranges of mechanical properties in Q&P steel [8]. Some studies have demonstrated that the Q&P steel with the partial or full austenitisation processing routes reveals pearlite-free multiphase microstructures comprising ferrite and lath microstructures (martensite and bainite) along with retained austenite [9]. The retained austenite is a metastable phase. When stress concentrates on it, the transformation-induced plasticity (TRIP) effect can occur which can prevent cracks from propagating, delay the onset of necking and increase the total elongation. Therefore, normally the higher the volume fraction of retained austenite, the better the mechanical properties of the steel. But the retained austenite stability also plays an important role in the mechanical properties of Q&P steels. And stably retained austenite can result in continuous work harding during the deformation process. Thus controlling the stability of the retained austenite at room temperature is essential to achieve a good combination of mechanical strength and ductility.

Some studies have demonstrated that not only the volume fraction of each phase has a significant influence on the mechanical properties, morphological characteristics (equiaxed or lamellar) also have a significant impact [7]. Sugimoto et al. [10] found that the lamellar TRIP steel obtained excellent comprehensive mechanical properties, though the flow stress was decreased compared to the equiaxed TRIP steel. Mark et al. [11] illuminated that the lamellar microstructure also showed continuously retained austenite transformation with strain, demonstrating a good TRIP effect. And TRIP effect takes place in high carbon blocky retained austenite more easily than that in low carbon film-like retained austenite [12,13].

Although it has been widely proved that the lamellar microstructure possesses an excellent mechanical properties compared with the equiaxed microstructure, especially the lamellar retained austenite [14,15], the effect of phase volume fraction is not taken into consideration. Therefore, in order to strictly research the effect of microstructure morphology on mechanical properties at same each phase volume fraction, the current work studied a novel lamellar Q&P steel using martensitic microstructure prior to the Q&P process compared to the conventional equiaxed Q&P steel.

Experimental

The chemical composition of the steel investigated in this study is given in Table 1. The steel was forged, hot and cold rolled to 1.5 mm in thickness after 50 kg vacuum melting. The A_c1 and A_c3 temperatures were tested by dilatometry (DIL 805A) to be 692°C and 923°C, respectively, corresponding to the heating rate of 5°C/s.

Table 1

Chemical composition of the investigated steel (wt-%).

C	Si	Mn	P	S	Al	Nb
0.23	1.93	2.18	0.0081	0.0044	0.043	0.023

Essentially, the different microstructure morphologies are determined by the initial condition of the steel: cold-rolled pearlite-ferrite for the equiaxed microstructure and completely martensite for the lamellar microstructure. Therefore, the sample to obtain lamellar microstructure was first austenitised at 960°C for 20 min in electric furnace and then quenched in oil to acquire completely martensite before the Q&P process.

The volume fractions of each phase for the tested steels were measured by dilatometry and calculated based on lever rule. As shown in Figure 1, the volume fractions of austenite was calculated by the following Equation (1): (1) where k₁ and k₂ are the slope of the heating curve, b₁ and b₂ are the intercept, g is the length change when the temperature is T during the austenitising process.

The volume fractions of austenite as a function of annealing temperature for the tested steels were measured by dilatometry as shown in Figure 2. When annealed at the (Ac₁+ Ac₃)/2 temperature, the TRIP steel can obtain excellent mechanical properties [16], so the intercritical annealing temperature was chosen to be (A_c1+ A_c3)/2,810°C. Thus, the corresponding volume fractions of austenite at 810°C as a function of annealing time for two types of steels were measured by dilatometry as shown in Figure 3. In order to get the same volume fraction of austenite, annealing time of the equiaxed sample is 1,000 s, and that of the lamellar sample is 561 s. At this time, the volume fractions of austenite in two samples are both 67%.

Figure 4 shows the volume fraction of martensite as a function of quenching temperature for the two types of steels measured by dilatometry. The M_s temperatures of the equiaxed sample is 260°C, while that of the lamellar sample is 255°C. In order to obtain same volume fraction of each phase in two types of steels, the equiaxed Q&P steel should be cooled rapidly to 225°C and the lamellar steel is 223°C, and then held at target temperature for 10 s. finally, both samples are partitioning at 400°C for 150 s and quenched to room temperature. Thus, the heat treatments are shown in Figure 5 and the Q&P process is carried out by heat treatment system for thin steel sheet (CCT-AY-Ⅱ, Japan).

The tensile specimens with 25 mm gauge length were cut from the sheet parallel to the rolling direction by a wire-electrode machine. The tensile tests were carried out at a speed of 1 mm min⁻¹ on a SANS XYB605C universal testing machine. The microstructure and fracture morphology were observed using a field emission scanning electron microscope (Qutanta FEG 450, FEI Company, U.S.A.). For more detailed analysis of the microstructures, the samples were characterised by a ZEISS ULTRA 55-type ﬁeld emission scanning electron microscope with a back scatter electron diffraction (EBSD) unit. XRD tests were made by using Cu–K_α radiation. The volume fractions of retained austenite were measured by comparing the integrated intensity of the (200) _γ , (220) _γ , (311) _γ , (200) _α and (211) _α peaks [17]. The carbon concentration of the retained austenite (C_γ, mass%) was obtained using the following Equation (1) [18]: (2) where a_γ is the retained austenite lattice constant, nm.

Results

Microstructures

The original microstructures of the two types of Q&P steel (the equiaxed steel and the lamellar steel) before the Q&P process in this study are shown in Figure 6(a,b). The original microstructure of the equiaxed Q&P steel is as-cold-rolled banded ferrite and pearlite, while that of the lamellar Q&P steel is full martensite. The initial microstructure plays an important role in the austenite formation process and the morphology of formed austenite [19]. Figure 2 shows that Ac1 in the equiaxed sample is lower than that in the lamellar sample. This is because that pearlite and ferrite are banded distribution, and carbon is mainly concentrated in the pearlite region. During the austenitising process, austenite nuclei first occurs at the phase boundaries between cementite and ferrite in the pearlite. Because the moving speed of the γ/α interfaces is higher than that of γ/pearlite, the ferrite in the pearlite disappears earlier than the cementite, which demonstrates that the rate of pearlite austenitising is lower than that of ferrite austenitising. Therefore, the increasing rate of the volume fraction of austenite is lower at initial stage and then becomes higher with the rise of temperature as shown in Figure 2. While, carbon distribution is homogeneous in full martensite in the lamellar Q&P steel, so the austenite nuclei occur at higher temperature. Austenite nuclei appear not only at the prior austenite grain boundaries but also between martensite laths. The austenite formed at prior austenite grain boundaries is blocky (γ_b), while the austenite embedded between martensite laths is lamellar/acicular (γ_a). With an increase of annealing temperature, the size of γ_b increases rapidly, and the phase boundaries move toward the martensite interior. The size of γ_a increases slowly, which is because the size of length (corresponding to the direction of martensite lath) increases dramatically, but that of the width (perpendicular to the direction of the strip) does not increase much. Therefore, γ_b will coarsen and become the majority form in the intercritical microstructure, whereas coarsening of γ_a will be less obvious than that of γ_b [13]. Thus, there will be great difference in the final microstructure.

The final microstructures with different morphology (equiaxed and lamellar) after Q&P process are shown in Figure 7. In the SEM micrographs, ferrite (F) emerges dark and islands of untempered martensite and retained austenite (UM/RA) present light and smooth. The rest adjacent to M/RA islands is tempered martensite (TM) which is distinguishable because of going through different conditions [20]. TM appears in the quenching process, and is tempered during the partitioning process, while UM occurs in the second cooling process after partitioning, without tempering. Moreover, carbide precipitation is observed on the TM matrix. The formation of carbide precipitation can decrease the carbon concentration of the austenite phase, which in turn decreases the stability of the austenite. So the volume fraction of retained austenite in the final microstructure is less than that of the initial design. In the equiaxed sample, microstructure morphology of F, TM and UM/RA is blocky, while it is lath shape in the lamellar sample which is inherited from oil quenched martensite.

The volume fraction of each phase and mass fraction of carbon in retained austenite of tested steels are shown in Table 2. The volume fractions of ferrite and TM are calculated by dilatometry based on lever rule. The volume fraction and carbon content of retained austenite are investigated by XRD. Thus, the rest is untempered martensite. It indicates that the volume fraction of each phase in the two samples is mainly equal.

Table 2

Volume fraction of each constituent phase.

Sample	f_F/%	f_TM/%	f_UM/%	f_RA/%	C_RA/%
Equiaxed	32.93 ± 0.43	33.54 ± 0.43	24.95 ± 0.43	8.58 ± 0.43	1.120 ± 0.016
Lamellar	32.91 ± 0.43	33.55 ± 0.43	25.05 ± 0.43	8.49 ± 0.26	1.055 ± 0.001

Note: f_F, volume fraction of ferrite; f_M, volume fraction of TM; f_M, volume fraction of untempered martensite; f_RA, volume fraction of retained austenite; C_RA, carbon content of retained austenite.

Retained austenite characterisation

EBSD analysis results of the tested steels are shown in Figure 8, in which blue corresponds to fcc phase (RA) and others to bcc phase (F, TM and UM). The light grey refers to F and dark grey to TM. The dark region represents UM owing to its high defect densities and large quantities of substructures that resulted in low qualities of Kikuchi bands [21]. In the equiaxed sample, F is blocky, and almost all of the RA are equiaxed and present as discrete particles between ferrite and martensite. In the lamellar sample, F is lath, and most RA is acicular (γ_a), minority is blocky (γ_b). The white dash line presents the prior austenite grain boundary. γ_b is distributed along the dash line and has a lower stability and easily transforms into martensite during the deformation process [13]. Compared with γ_b, γ_a is more stable and better able to resist strain-induced martensitic transformation during the deformation process.

Fracture mechanisms

The fracture toughness of materials is dependent not only on the volume fraction of each phase and interface (grain boundary and phase boundary) but also on the morphology and distribution of the martensite [22]. Moreover, the volume fraction of interface is higher in the lamellar sample for its morphological feature. Fracture morphologies of the tested steels after uniaxial tensile test are shown in Figure 9(a,b), respectively. The fracture surface of the equiaxed sample is different at different regions. There are lots of flat facets in the left part of the fracture surface [23]. The middle part of the fracture surface contains mainly dimples which is a mark of admirable fracture toughness. Some cracks occur in the right part of the fracture surface. While the fracture of the lamellar sample is almost consisted of dimples. Consequently, the fracture toughness of the lamellar sample is relatively better than that of the equiaxed sample, which is proved in Table 3.

Table 3

Mechanical properties of samples after Q&P treatment.

Sample	R_0.2 (MPa)	R_m (MPa)	YR	A (%)	R_m× A (GPa·%)
Equiaxed	884.79 ± 54.14	1225.18 ± 29.59	0.7222	19.22 ± 1.56	23.57 ± 2.47
Lamellar	853.57 ± 43.31	1144.84 ± 10.39	0.7456	25.42 ± 0.03	29.11 ± 0.23

Note: R_0.2, yield strength; R_m, tensile strength; YR, yield ratio; A, total elongation; R_m× A, the product of tensile strength and total elongation.

Mechanical properties

The engineering stress–strain curves and the corresponding work-hardening behaviours of each samples are illustrated in Figure 10. The mechanical properties of each investigated sample, such as yield strength (R_0.2), ultimate tensile strength (R_m), yield ratio (YR), total elongation (A) and the product of tensile strength and elongation (R_m× A) are summarised in Table 3. Figure 10(a) shows that continuous yielding is observed in each sample, which is usually related to high mobile dislocation density. Though ultimate tensile strength of lamellar sample is relatively lower than that of equiaxed sample, a large yield ratio and total elongation are obtained in the lamellar sample, and the yield strength of the two samples is almost equal. Moreover, the lamellar sample produces a higher product of strength and elongation which is widely used as an indication of combination of strength and ductility. Therefore, more energy could be absorbed by the lamellar steel during the process of collision.

The work-hardening behaviour is analysed based on the following Hollomon equation associated with the necking criterion: (2) where σ is the true stress, ε is the true strain, k is a constant, n is the work-hardening exponent. Then, the work-hardening exponent n can be written as: (3)

Figure 10(b) shows the n values as a function of trues train for the tested steels and the elastic strain is deducted. The blue line corresponds to the instability criterion n = ε_u, where ε_u is the true strain at the beginning of necking. It can be seen that the work-hardening behaviour can be divided into four stages: (1) in the initial stage (ε ≤ 0.011), the n values of two samples both decrease rapidly (stage I); (2) when 0.011 ≤ ε ≤ 0.07, the n values of two samples both slowly rise for the TRIP effect, and the n values of lamellar sample is smaller than that of equiaxed sample (stage II); (3) when 0.07 ≤ ε ≤ ε_u, the n values of two samples both slowly rise and decline slightly which corresponds to the uniform elongation with wide plateaus, while the n values of lamellar sample are bigger than that of equiaxed sample (stage III); (4) when ε > ε_u, the n values of two samples both slowly decrease during necking stage (stage IV).

The work-hardening process is closely associated with the TRIP effect. Xiong et al. [12] speculated that martensitic transformation takes place in high carbon blocky retained austenite at the onset of deformation, while low carbon film-like retained austenite, despite having a much lower carbon content, is stable at strains up to 12%. Chiang et al. [24] showed that substantially all of the retained austenite transformed to martensite at 7% strain in the equiaxed microstructure, while the TRIP effect is completely over only when the strain is 17% in the lamellar microstructure. The reason for the above phenomenon is that the stability of γ_a is higher than that of γ_b. The volume fraction of γ_b in equiaxed sample is higher than that in lamellar sample, the n values of lamellar sample is smaller than that of equiaxed sample for the weaken TRIP effect in the stage II. However, there is little γ_a in the equiaxed sample, and a large amount of γ_a exist in lamellar sample. So the n values of lamellar sample is bigger than that of equiaxed sample in stage III for the continuous TRIP effect.

In addition to the morphology of RA, the morphology of neighbouring microconstituents also has an important effect on the stability of RA, which is achieved by influencing the stress partitioning. Compared to the blocky matrix, the lamellar matrix might shield neighbouring RA from externally applied stress via reducing the hydrostatic pressure without changing the equivalent stress, which in turn reduces the mechanical driving force for transformation [25]: (4) where Δg^σ is the mechanical driving force, σ_h is the hydrostatic pressure and is the equivalent stress. Moreover, the surface area of the lamellar grain is larger than that of the equiaxed grain, so the volume fraction of grain boundary in lamellar is higher than that in equiaxed sample. Moreover, from the view of microstructure, there is an angle between some lamellar grains and the tensile direction, so the lamellar structure is more prone to unstability and rotates during the deformation process. The rotation is presented as deformation compatibility from the macroscopic view. In the lamellar sample and can release energy and reduce the stress concentration which will result in an obvious higher elongation. While the equiaxed structure is more stable and the rotation of microstructure does not occur which results in a stress concentration and higher tensile strength. But the specimen fractures quickly.

The schematic illustration of the multi-stage work-hardening behaviour in the two types of Q&P steel is shown in Figure 11. In the stage I, plastic deformation occurs in the two samples. And the lamellar grain rotates easily during the tensile process because there is an angle between some lamellar grains and the tensile direction. But it will not occur in the equiaxed sample. In the stage II, the blocky retained austenite transforms into martensite quickly. Therefore, there is little retained austenite in the equiaxed sample and there is still much acicular retained austenite in the lamellar sample. In the stage III, continuous plastic deformation occurs in the equiaxed sample and there is not only continuous plastic deformation but also martensitic transformation taking place in acicular retained austenite in the lamellar sample.

Conclusion

In summary, the samples with same volume fraction of each phase and different microstructure morphology (equiaxed and lamellar) are obtained by controlling the initial microstructure. The microstructures and mechanical properties are investigated by means of XRD, FEM, EBSD and tensile test. The Q&P steel with lamellar microstructure has a better comprehensive mechanics performance than the Q&P steel with equiaxed microstructure. Moreover, the lamellar sample shows a continuous work-hardening rate for the acicular RA, which can improve sheet formability, allowing this steels to be more compatibly used for automotive structural components.

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Speer

, Matlock

, De Cooman

, Carbon partitioning into austenite after martensite transformation. Acta Mater. 2003;51:2611–2622. doi: 10.1016/S1359-6454(03)00059-4

Speer

, Edmonds

, Rizzo

, Partitioning of carbon from supersaturated plates of ferrite, with application to steel processing and fundamentals of the bainite transformation. Curr Opin Solid State Mater Sci. 2004;8:219–237. doi: 10.1016/j.cossms.2004.09.003

De Cooman

BC.

Structure–properties relationship in TRIP steels containing carbide-free bainite. Curr Opin Solid State Mater Sci. 2004;8:285–303. doi: 10.1016/j.cossms.2004.10.002

Speer

, De Moor

, Findley

, Analysis of microstructure evolution in quenching and partitioning automotive sheet steel. Metall Mater Trans A. 2011;42:3591–3601. doi: 10.1007/s11661-011-0869-7

, Gao

, Nakashima

, In-situ EBSD study of deformation behavior of retained austenite in a low-carbon quenching and partitioning steel via uniaxial tensile tests. Mater Charact. 2016;118:431–437. doi: 10.1016/j.matchar.2016.06.020

Seo

, Cho

, De Cooman

BC.

Kinetics of the partitioning of carbon and substitutional alloying elements during quenching and partitioning (Q&P) processing of medium Mn steel. Acta Mater. 2016;107:354–365. doi: 10.1016/j.actamat.2016.01.059

De Knijf

, Petrov

, Föjer

, Effect of fresh martensite on the stability of retained austenite in quenching and partitioning steel. Mater Sci Eng: A 2014;615:107–115. doi: 10.1016/j.msea.2014.07.054

Seo

, Cho

, Estrin

, Microstructure-mechanical properties relationships for quenching and partitioning (Q&P) processed steel. Acta Mater. 2016;113:124–139. doi: 10.1016/j.actamat.2016.04.048

Mandal

, Ghosh

, Bera

, Effect of partial and full austenitisation on microstructure and mechanical properties of quenching and partitioning steel. Mater Sci Eng: A 2016;676:56–64. doi: 10.1016/j.msea.2016.08.094

10.

Sugimoto

, Misu

, Kobayashi

, Effects of second phase morphology on retained austenite morphology and tensile properties in a TRIP-aided dual-phase steel sheet. ISIJ Int. 1993;33:775–782. doi: 10.2355/isijinternational.33.775

11.

Mark

, Westphal

, Boyd

, Microstructural design of multiphase advanced high strength steels. Can Metall Quart. 2009;48:237–245. doi: 10.1179/cmq.2009.48.3.237

12.

Xiong

, Chen

, Huang

, The effect of morphology on the stability of retained austenite in a quenched and partitioned steel. Scripta Mater. 2013;68:321–324. doi: 10.1016/j.scriptamat.2012.11.003

13.

Ding

, Tang

, Zhao

A novel design to enhance the amount of retained austenite and mechanical properties in low-alloyed steel. Scripta Mater. 2014;88:21–24. doi: 10.1016/j.scriptamat.2014.06.014

14.

Ding

, Tang

, Zhao

, Effect of intercritical temperature on quenching and partitioning steels originated from martensitic pre-microstructure. J Mater Res. 2014;29:2525–2533. doi: 10.1557/jmr.2014.285

15.

Ding

, Tang

, Zhao

, A new type of quenching and partitioning processing developed from martensitic pre-microstructure. Mater Manuf Process 2014;29:704–709. doi: 10.1080/10426914.2014.912304

16.

Matsumura

, Sakuma

, Takechi

Enhancement of elongation by retained austenite in intercritically annealed 0.4C-1.5Si-0.8Mn steel. ISIJ Int. 1987;27:570–579. doi: 10.2355/isijinternational1966.27.570

17.

Maruyama

X-ray measurement of retained austenite volume fraction. J Jpn Soc Heat Treat. 1977;17:198–204.

18.

Dyson

, Holmes

Effect of alloying additions on the lattice parameter of austenite. J Iron Steel Inst. 1970;208:469–474.

19.

Castro Cerda

, Sabirov

, Goulas

, Austenite formation in 0.2 C and 0.45 C steels under conventional and ultrafast heating. Mater Design. 2017;116:448–460. doi: 10.1016/j.matdes.2016.12.009

20.

Santofimia

, Zhao

, Sietsma

Microstructural evolution of a low-carbon steel during application of quenching and partitioning heat treatments after partial austenitization. Metall Mater Trans A. 2009;40A:46–57. doi: 10.1007/s11661-008-9701-4

21.

Santofimia

, Zhao

, Petrov

, Characterization of the microstructure obtained by the quenching and partitioning process in a low-carbon steel. Mater Charact. 2008;59:1758–1764. doi: 10.1016/j.matchar.2008.04.004

22.

Ghassemi-Armaki

, Maaß

, Bhat

, Deformation response of ferrite and martensite in a dual-phase steel. Acta Mater. 2014;62:197–211. doi: 10.1016/j.actamat.2013.10.001

23.

Han

, Da Silva

, Ponge

, The effects of prior austenite grain boundaries and microstructural morphology on the impact toughness of intercritically annealed medium Mn steel. Acta Mater. 2017;122:199–206. doi: 10.1016/j.actamat.2016.09.048

24.

Chiang

, Lawrence

, Boyd

, Effect of microstructure on retained austenite stability and work hardening of TRIP steels. Mater Sci Eng: A 2011;528:4516–4521. doi: 10.1016/j.msea.2011.02.032

25.

Zhang

, Findley

KO.

Quantitative assessment of the effects of microstructure on the stability of retained austenite in TRIP steels. Acta Mater. 2013;61:1895–1903. doi: 10.1016/j.actamat.2012.12.010