Effect of partial and full austenitisation on microstructure and mechanical properties of quenching and partitioning steel

Materials Science & Engineering A 676 (2016) 56–64

Contents lists available at ScienceDirect

Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea

Effect of partial and full austenitisation on microstructure and mechanical properties of quenching and partitioning steel G. Mandal a, S.K. Ghosh a,n, S. Bera b, S. Mukherjee c a

Department of Metallurgy & Materials Engineering, Indian Institute of Engineering Science and Technology, Shibpur, Howrah 711103, India Department of Metallurgical and Materials Engineering, National Institute of Technology, Durgapur, Burdwan 713 209, India c R&D Division, Tata Steel Limited, Jamshedpur 831007, India b

art ic l e i nf o

a b s t r a c t

Article history: Received 18 June 2016 Received in revised form 9 August 2016 Accepted 23 August 2016 Available online 24 August 2016

A novel high-strength steel has been made through thermo-mechanical controlled processing with the finish rolling temperature of 750 °C followed by air cooling. Subsequently, both partial austenitisation at 800 °C and fully austenitisation at 930 °C have been attempted for equal duration of 30 min prior to one step quenching and partitioning (Q&P) at 345 °C below MS temperature (365 °C). As-rolled steel reveals ferrite-bainite-martensite microstructures with a good combination of strength and ductility. After Q&P, all the specimens have exhibited the multiphase microstructures comprising ferrite, lath microstructure (martensite and bainite), and retained austenite with the volume fractions of up to 10.50 wt%. It is evident that partitioning for 30 min leads to good carbon enrichment ( 41 wt%) of the austenite phase from the neighbouring martensite or bainite which might be due to fast partitioning kinetics and possible suppression of carbides through a combination of Si and Al additions. The attractive combination of tensile strength (921–922 MPa) and ductility (25–26% total elongation) along with low yield ratio (0.63– 0.69) are attributed to ferrite and lath microstructures along with the thin film like carbon enriched retained austenite obtained after Q&P process. & 2016 Elsevier B.V. All rights reserved.

Keywords: Alloys Q&P treatment Ferrite Martensite Retained austenite Mechanical properties

1. Introduction In recent years, there has been an increased emphasis on the development of new advanced high strength steels (AHSS) in order to reduce vehicle weight and consumption of raw materials as well as to maintain or even improve safety standards in the automobile industries. In this context, there has been increased demand for the development of complex phase or multiphase steels having ferrite-bainite-martensite microstructure with significant amount of retained austenite, thereby manifesting high combination of strength and ductility [1]. As a critical component of main commercial steels, a certain amount of retained austenite can potentially increase ductility [2,3] and toughness [4]. These are attributed to the transformation induced plasticity (TRIP) phenomenon of carbon-enriched retained austenite during plastic deformation [5]. To obtain a certain amount of carbon-enriched retained austenite in steel at room temperature, the traditional method is to increase the carbon content in steel to reduce the martensite start (MS) temperature below room temperature [6]. However, high carbon content in steel is detrimental to ductility n

Corresponding author. E-mail address: [email protected] (S.K. Ghosh).

http://dx.doi.org/10.1016/j.msea.2016.08.094 0921-5093/& 2016 Elsevier B.V. All rights reserved.

and weldability [7]. In this regard, a fundamentally new heat treatment named quenching and partitioning (Q&P) has been proposed [8,9] to create advanced high strength multiphase steels with a controlled amount of carbon-enriched retained austenite at room temperature. The quenching step was performed to form specific fraction of supersaturated martensite and retained austenite by fast quenching below the MS temperature but above the martensite finish (Mf) temperature. A subsequent partitioning treatment at the quenching temperature below the MS temperature (one-step treatment) or above the MS temperature (two-step treatment) was employed to accomplish complete diffusion of carbon from martensite to retained austenite in the absence of carbide precipitation by alloying with appropriate amount of Si and/or Al. Finally, the carbon-enriched austenite was mostly retained at room temperature. It is well known that Si significantly restricts the formation of cementite because its solubility is incredibly low or even near zero in cementite phase [10]. However, alloying with certain amount of Si is ineffective to inhibit formation of epsilon carbide [11,12]. Therefore, it is possible to form lower bainite (bainitic ferrite plus ε-carbide) rather than carbidefree bainite during the partitioning process [13]. Al which is a typical alloying addition to transformation-induced plasticity (TRIP) steel, is employed to suppress carbide formation during partitioning and as a consequence, substantial fraction of carbon

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

Fig. 1. Complete dilatometric curve of the investigated steel showing Ac1, Ac3 and MS temperatures.

Soaking: 1200°C, 120 min. Stage 1(17%, 1150°C - 1050°C) Stage 2 (20%, 1020°C - 950°C)

Temp. (°C)

enriched retained austenite is produced [14]. According to Wang and Chang [15], Al also improves the anti-corrosion behaviour by forming Al2O3 layer on the surface. However, experimental investigations with partial and fully austenitisation can improve our understanding of the Q&P process for adjusting and tailoring the required mechanical properties of Q&P steels. The studies on the relationship between microstructure and mechanical properties of high strength steels during the one-step Q&P process are limited [16]. It is reported that one step Q&P process is achieving high strength level of 700–2400 MPa with adequate ductility of 10–20% [16–18]. This combination of mechanical properties can be achieved by multiphase microstructure consisting of mostly lath martensite, bainite and retained austenite enriched with carbon. It appears that enough attention has not been paid to the fact that martensite being the matrix of steel makes a major contribution to the mechanical properties [19]. So investigation on the microstructure-properties relationship in Q&P steel is still open. The strategy of Q&P process is to make cost effective steel with high tensile strength and good ductility. However, to fulfil this requirement several processing parameters like austenitisation temperature (AT), cooling rate (CR), quenching temperature (QT) and partitioning time (PT) should be optimised. In view of the above, steel with somewhat different chemistry from conventional TRIP-assisted steel was chosen for superior mechanical properties and corrosion resistance [20,21] by alloying with copper, nickel and cobalt for potential structural applications in motor vehicles, high-rise buildings, power plants, refineries, industrial sheds, ware houses etc. [22]. In addition, alloying with Si and Al was employed to achieve carbon depletion from supersaturated martensite to austenite by restraining carbide precipitation during partitioning process. The improved mechanical properties resulting from Q&P processing were discussed based on the multiphase microstructure along with retention of carbon enriched austenite and compared with the current generation Q&P steels.

57

TNR = 933°C Stage 3 (17%, 900°C - 860°C) Stage 4 (20%, 810°C – FRT(°C))

Heating

FRT = 750°C

Time (min.)

AC

Fig. 2. Schematic illustration of Thermo-Mechanical Controlled Processing (TMCP) schedule. FRT stands for Finish Rolling Temperature.

2. Experimental procedures A laboratory scale 25 kVA air induction furnace was used to manufacture the steel with the chemical composition as shown in Table 1. After cropping the top section of the ingot containing shrinkage/pipes, the remaining portion of 200 mm
50 mm
50 mm size was hot forged down to about 19 mm
19 mm cross section. After that, the Ac1, Ac3 and MS temperatures were measured as 725 °C, 830 °C and 365 °C respectively, by a Gleeble 1500D LVDT type dilatometer using cylindrical hollow samples of 1 mm wall thickness. At first the samples were heated at a heating rate of 10 °C/s at 1000 °C and soaked for 5 min, and then cooled under constant rate of 20 °C/s to room temperature. The experiments were carried out under argon atmosphere. The complete dilatometric curve of investigated steel is shown in Fig. 1. The TNR temperature of the investigated steel had been experimentally measured to be as 933 °C [21]. The forged bars were soaked at 1200 °C and subsequently subjected to TMCP with 750 °C finish rolling temperature (FRT) according to the laboratory scale schedule as mention in Fig. 2 using two-high rolling mill (10 HP) and finally the rolled plates Table 1 Chemical composition (wt%) of the investigated alloy. C

Mn

Si

Al

Cu

Ni

Co

S

P

Fe

0.20

1.65

1.40

1.50

1.30

1.05

1.07

0.006

0.014

Bal.

Fig. 3. Schematic illustration of the thermal profile and phase transformation behaviour of Q&P steel. Ci, Cγ, Cα′, Cγ′, Cα″ and Cα′1 represent the carbon concentrations of the initial alloy, austenite, martensite, retained austenite, partitioned martensite and martensite formed during cooling after partitioning, respectively. FA and PA stand for full austenitisation and partial austenitisation.

(E8 mm) were allowed to air cooling (AC). The TMCP samples were treated by Q&P processes, as shown in Fig. 3. All samples were subjected to partial austenitisation (PA) at 800 °C and full austenitisation (FA) at 930 °C temperatures with the same holding time of 30 min, followed by rapid quenching in a salt bath maintained at 345 °C below MS temperature (365 °C) and partitioning at the same temperature for 30 min. After partitioning treatment, all specimens were air cooled to room temperature. In this context, it is noteworthy that the earlier researchers followed the partitioning time less than one minute to avoid formation of bainite as well as epsilon carbide during Q&P treatment, thereby getting maximum retained austenite. Bhadeshia et al. [23] reported that

58

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

the long partitioning time might be causing bainite transformation during Q&P treatment that reduced the final volume fraction of retained austenite. However, in real industrial practice or large scale components productions, longer partitioning time is required to produce homogeneous microstructure which leads better mechanical properties. Therefore, the present investigation was carried out on Q&P treatment for 30 min duration. The as-treated samples were metallographically prepared and etched with 2% nital solution for detailed microstructural investigation using Carl Zeiss, Axiovert 40 Mat optical microscope (OM) as well as Hitachi scanning electron microscope (SEM) (HITACHI, S-3400N) operated at 20 kV in secondary electron mode and few representative micrographs were taken. In addition, the image analysis was also made to quantify the microstructural parameters from the optical micrographs, using Axiovision

software (version 4.8). For transmission electron microscopy, typical 3 mm diameter discs were punched out from thin foils and subjected to twin jet electro-polishing, using a mixture of electrolyte of 90% glacial acetic acid and 10% perchloric acid at temperature of about 12 °C. Thin electron transparent samples were subsequently examined in a transmission electron microscope (TEM) (Tecnai, G2) at 200 kV operating voltage. Identification of retained austenite and its volume fraction in the different bulk solid samples were determined by using a ‘Bruker-Advance D8’ XRD machine integrated with a Copper tube with an operating voltage and current of 40 kV and 30 mA and the data was collected over a 2θ range of 35°–105° with a step of 0.01°/ sec. X-Ray data were analysed by means of the Joint Committee on Powder Diffraction Standards (JCPDS). The measured diffraction patterns were analysed by Panalytical X′-Pert High Score Plus

(b)

(a)

MA/Bainite

Ferrite 50µm

(d)

(c)

MA/Bainite

Ferrite 50µm

(e)

(f) MA/Bainite

Ferrite

50µm Fig. 4. Optical and SEM micrographs of the investigated steel subjected to (a) and (b) TMCP followed by AC, (c) and (d) TMCP plus PA followed by Q&P and AC and (e) and (f) TMCP plus FA followed by Q&P and AC. MA: Martensite – austenite constituent.

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

software to find the peak position precisely. The volume fraction and lattice parameter of subsequent phases of FCC (γ-phase) and BCC (α-phase) were measured from diffraction data by Rietveld analysis. Here we assume the percentage of α-phase is the sum of ferrite, bainite and martensite. During Rietveld refinement fitting the effect of grain size, strain and texture were taken care by adjusting related parameters until a good matching was reached. The parameters suggested negligible effect of texturing. Similar type of analysis with Rietveld refinement of different phases in twinning induced plasticity (TWIP) steel was reported earlier [24]. Micro-hardness values of various phases were obtained using 200 g load and 20 s dwelling time. The Vickers hardness (HV) values of the samples were evaluated in a Brinell-Vickers combined hardness tester (BV-250(SPL)), using 30 kg load for 20 s dwelling time. Average values of six measurements for each sample were reported. According to ASTM E8M standard, flat subsize tensile specimens (100 mm total length), having gauge length of 25 mm, gauge width of 6.35 mm and thickness of 3.08 mm were prepared from the rolled plate in longitudinal direction. The tests were carried out with the crosshead speed of 0.5 mm/min in an Instron 4204 testing machine with an extensometer. Three specimens were tested for each variant and the average values were reported.

59

austenite was expected to occur to a certain extent during the PA. The formation of substantial amount of ferrite, obtained after Q&P process, is obvious as it has grown during cooling from the ferrite which was present during PA. However, the amount of ferrite decreases when the investigated steel samples are processed by FA and Q&P followed by air cooling. The formation of reduced amount of ferrite in this treatment could be related to the rapid cooling under paraequilibrium conditions with the austenite prior to Q&P. Similar type of microstructure formation was reported in lowcarbon steel during Q&P treatments after partial austenitisation [27]. The steel was subjected to FA before quenching, so the ferrite formation is not usually expected. However, during cooling from 930 °C, the steel passes through the two-phase (γ þ α) region which results into the redistribution of carbon atoms of austenite. The redistribution occurs by diffusion of carbon atoms which leads the formation of regions with varying carbon concentrations in austenite and low carbon austenitic regions transform to ferrite. This is in agreement with the formation of proeutectoid ferrite (20–40%) from the austenite phase during slow cooling [28]. Similar type of observation on formation of some ferrite was reported during cooling from fully austenitised conditions prior to helium quenching [29]. 3.2. TEM analysis

3. Results and discussion 3.1. Microstructures of TMCP and Quenched & Partitioned steels Fig. 4(a)–(f) show the optical and SEM micrographs of investigated steel, subjected to TMCP with 750 °C FRT followed by AC and subsequently PA (partial austenitisation) & FA (full austenitisation) treated followed by Q&P and AC. In general, it is evident that the microstructures consist of ferrite, bainite, martensite and retained austenite. Fig. 4(a) and (b) show the micrographs of the samples subjected to TMCP with FRT750°C (Ac1 oFRT oAc3) i.e., two-phase (ferrite-austenite) region transforms to non-equiaxed (elongated) ferrite (38 75%) and balance amount of lath microstructures (bainite and martensite) after air cooling. It is important to note that while ferrite appears as a bright phase in optical micrograph, the same appears as dark phase in SEM (Fig. 4(b)). On the other hand, lath microstructure appears as dark phase mixture in optical micrograph and its preferential etching contrast is clearly revealed in SEM (Fig. 4(b)). Such kind of microstructural evolution was in agreement with the earlier report in TRIP-assisted steel [25,26]. The above hypothesis has been confirmed by the micro-hardness measurement. The hardness of ferrite varies from 290 to 320 HV, whereas mixture of bainite and martensite being the stronger phase constituents indicates higher range of 350– 425 HV. The high hardness value of ferrite is related to the solid solution hardening due to the interstitial and substitutional elements and dislocation density achieved during TMCP. The microstructures (Fig. 4(c) and (d)) of the steel specimens subjected to TMCP (750 °C FRT) and PA at 800 °C (o Ac3) for 30 min followed by Q&P at 345 °C ( oMS) for 30 min and AC consist of multiphase constituents, comprising ferrite (30 74%), and balance as lath microstructures comprising martensite, bainite and also retained austenite. The average grain size of ferrite lies within 15–25 mm. Fig. 4(e) and (f) reveal the microstructures of the specimens subjected to TMCP (750 °C FRT) and FA at 930 °C (4 Ac3) temperature followed by Q&P and AC. These microstructures essentially consist of ferrite (20 73%) having grain size in the range of 7–15 mm and the balance as the lath microstructures comprising martensite, bainite and also retained austenite. The partitioning of alloying elements in between ferrite and

Transmission electron microscopy was carried out to characterise the microstructures in greater detail which can not be done in optical and scanning electron microscopy. Fig. 5(a) shows the TEM bright field (BF) image of the specimen processed by TMCP followed by AC which reveals the triple point of ferrite grains having equiaxed and elongated morphologies with some bainite (dark contrast) along the grain boundary of ferrite. Ferrite grains adjacent to bainite display high dislocation density as a consequence of accommodation strain. The growth of bainite plates is suppressed by adjacent ferrite grains. As a result, the length of bainite plates is largely reduced. Fig. 5(b) shows the magnified image of bainite segment of the Fig. 5(a) which reveals discontinuous bainitic carbide with adjacent of bainitic ferrite that gives the morphology of granular bainite. The regions between the ferrite grains/laths are carbon-enriched, which will subsequently transform to carbide during the formation of bainite structure. Fig. 5(c) shows a junction of two ferrite grains which has been penetrated by the lower bainitic structure. It appears that some amount of austenite remains untransformed as an island that has been sheltered by the adjacent ferrite structure (denoted by arrow). The selected area electron diffraction (SAED) pattern taken from the island region/blocky type of retained austenite is shown in the inset of Fig. 5(c). The analysis of SAED pattern indicates the presence of retained austenite. Fig. 5(d) shows the TEM image of triple point of prior austenite grains which have been divided into several regions. In each region, the bainitic and martensitic laths are almost aligning in the same direction and such a region is called a packet. A triangular region of dark contrast is evident at the inter-lath locations of bainite/martensite. It may be mentioned that, during quenching martensite forms within austenite grain, leaving untransformed austenite, closer to its lath boundaries. Some more carbon enriched austenite is available during partitioning step of 30 min which may further partially transform to martensite during cooling to room temperature. At the onset of the partitioning process, the dislocation density is very high which is attributed to the accommodation strain occurred during the shape change, accompanying the formation of martensite in Q&P process. A continuing partitioning results in the formations of dislocation cells within martensite laths (denoted by arrow). Fig. 5 (e) displays a typical TEM BF image of the ferrite grain, lath type martensite with film-like inter-lath retained austenite. Ferrite

60

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

(b)

(a) Bainite

(d)

(c) Ferrite

Dislocation cell

Bainite

Dislocation cell

Ferrite

(e)

(f)

Martensite Ferrite

Fig. 5. TEM bright field (BF) image showing ferrite and bainite in the specimen subjected to (a) and (b) TMCP followed by AC, (c) and (d) TMCP followed by PA and Q&P followed by AC showing mainly ferrite and bainite with blocky retained austenite (RA) and (e) and (f) TMCP followed by FA and Q&P followed by AC showing mostly ferrite and martensite with film like retained austenite (RA). Insets of (c) and (f) are SAEDs indicating retained austenite.

grain adjacent to martensite lath displays high dislocation density as a consequence of accommodation strain. Fig. 5(f) displays ferrite plate, lower bainite and lath martensite with high dislocation density and film-like inter-lath retained austenite along the lath boundaries. The SAED pattern taken from the region of inter-lath retained austenite is shown at the inset of Fig. 5(f). The analysis of SAED pattern confirms the presence of retained austenite. The

main difference in the microstructure between full and partial austenitization is martensite phase with film like retained austenite is prominent for full austenitisation, whereas bainite phase with blocky type retained austenite is predominant for partial austenitisation. Such a thin film-like retained austenite exhibits higher mechanically stability than blocky type [30,31]. In addition, film-like retained austenite reduces the formation of micro-cracks

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

61

[32] and prevents their propagation during loading [33]; thereby absorb the dislocation from adjacent lath structure (martensite or bainite) during plastic deformation [34,35]. Therefore, film-like retained austenite with martensite and bainite are exhibited a good combination of high strength, toughness and ductility in the Q&P steels. The volume fraction of martensite (fM) formed at the quenching temperature (QT) can be predicted by the Koistinen–Marburger equation: [36]

fM = 1 − exp⎡⎣ − 0.011 × ΔT⎤⎦=1 − exp⎡⎣ − 0.011 × ( MS−Q T)⎤⎦

(1)

where ΔT is the temperature difference between MS and QT The predicted volume fraction of martensite after quenching at 345 °C is approximately 20%. Therefore, after quenching at 345 °C (o Ms of 365 °C) good amount of parent austenite is expected, prior to the partitioning process. The progress of austenite to martensite transformation can be assessed quantitatively from the above equation (1) which is of wide practical application in carbon steels. It is important to note that MS temperature is a function of both austenite grain size and carbon content of austenite. In addition, MS temperature should vary with the partial and full austenitisation approach with the presence of different amount of ferrite. In the present study, the authors have not studied the effect of austenite grain size on MS temperature but calculated the carbon content of austenite by the following empirical equation [37]:

X γ = x¯ +

V∝(x¯ − xα ) (wt%) 1−V∝

(2)

where X γ is carbon concentration of austenite, x¯ is the average carbon concentration of the alloy, V∝ is the volume fraction of ferrite and xα is carbon concentration of ferrite. Considering, x¯ ¼0.20 wt%, xα ¼ 0.025 wt%, V∝=0.20 and 0.30, the calculated values have indicated that X γ of 0.24 wt% and 0.27 wt% are obtained during full austenitisation (FA) and partial austenitisation (PA), respectively. It is evident that the amount of enrichment of carbon in austenite is more in case of PA than FA. Subsequently, the results have been fitted with the diagram of carbon content (wt%) versus MS temperature, reported by Liu et al. [38] which has indicated that the MS temperature decreases to the degree of 10 °C and 17 °C for FA and PA conditions, respectively. Hence, the estimated MS temperatures of 355 °C and 348 °C lie below the experimental MS temperature (365 °C) for FA and PA conditions, respectively. The above estimation is quite logical as quenching temperature (QT) being close to MS temperature for PA condition, lower bainitic structure is predominant and lath martensite is the major microstructural constituent for FA where MS temperature reasonably lies well above QT temperature. As the MS temperature changes between full and partial austenitization, the fraction of martensite in the microstructure changes, as only one QT was used for both the processes. However, it was very difficult to measure the martensite volume fraction for both the cases using TEM characterisation. Therefore, the total amount of lath microstructures was reported in results, discussed associated with Fig. 4. It is noted that during Q&P process, carbon enriched austenite is retained with martensite and carbides are suppressed. Carbon atoms mainly diffuse into untransformed austenite at the early stages of partitioning (0–10 min). As the partition proceeds (10– 30 min), the difference in the carbon potential between the martensite and the untransformed austenite diminishes which could results in a little increase of the carbon concentration in retained austenite.

Fig. 6. XRD patterns of investigated samples subjected to different processing conditions.

3.3. X-Ray diffraction analysis Fig. 6 shows the measured diffraction pattern of the samples subjected to (a) TMCP followed by AC, (b) TMCP and followed by PA (800 °C) and Q&P plus AC and (c) TMCP followed by FA (930 °C) prior to Q&P process followed by AC, respectively. It is evident that the (110), (200), and (211) diffraction peaks of the BCC phase (ferrite, bainite or martensite) for all specimens, whereas (111), (200), (220), (311) and (222) diffraction peaks indicate FCC austenite (γ) phase in case of both PA and FA prior to Q&P processing. On the other hand the sample subjected to TMCP followed by AC has exhibited clearly (220)γ and (222)γ diffraction peaks but the (111)γ, (200)γ and (311)γ peaks are not visible. The precise lattice parameter (ap) of the austenite phase was determined by suitable extrapolation the diffraction angle 0–90° of the variation of lattice parameter as a function of the Nelson-Riley (N-R) parameter (cos2θ/sinθ þcos2θ/θ) for the all specimens [39]. The carbon concentration of retained austenite was measured by X-ray diffraction using the aforesaid lattice parameter extrapolation method and empirical equation, [40] as given below:

a p = 3.578 + 0.044 wt% C

( Å)

(3)

The volume fraction of retained austenite (VRA %), precise lattice parameter (ap) values and carbon concentration of retained austenite (CRA %) are also indicated in Table 2. Such kind of analysis of carbon concentration of retained austenite is consistent with the results reported earlier [41]. Table 2 indicates that the Q&P processed samples have higher the volume fraction of retained austenite than TMCP processed sample. In addition, the samples subjected to TMCP prior to FA at 930 °C and subsequently Q&P followed by AC has exhibited the highest value (10.50%) of retained austenite which should be reflected on the ductility and tensile toughness of the specimen. This is apparent that the higher austenitisation temperature compared to lower austenitisation temperature leads to a more positive effect on the stability of retained austenite. Table 2 substantiates the reasonable volume fraction of retained austenite and their carbon enrichments which indicates Table 2 The VRA, ap and CRA of the different samples subjected to various processing conditions. Sample conditions

Precise lattice parameter (ap) (Å)

TMCP þAC ap ¼ 3.6219 70.0005 TMCP þACþ PA þQ&Pþ AC ap ¼ 3.63357 0.0003 TMCP þACþ FA þ Q&Pþ AC ap ¼ 3.6398 7 0.0004

VRA (%)

CRA (%)

02.40 7 1 1.00 08.30 7 1 1.26 10.50 71 1.40

62

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

Table 3 Mechanical properties of the investigate steels subjected to various processing conditions. Processing conditions

Hardness (HV)

YS (MPa)

UTS (MPa)

YR

TEL (%)

PSE (GPa %)

TT (MPa)

TMCP þAC TMCP þPA þ Q&Pþ AC TMCP þFA þ Q&Pþ AC

3707 3 302 7 3 3107 4

5747 5 584 7 5 6417 6

1099 7 6 922 7 6 9217 5

0.52 0.63 0.69

147 2 25 72 26 73

15.38 23.05 23.94

1377 3 1997 3 230 7 5

YS: yield strength; UTS: ultimate tensile strength; YR: yield ratio; TEL: total elongation; PSE: product of strength and elongation; TT: tensile toughness.

their stability at room temperature. Good amount of austenitestabilising elements e.g., Mn, Cu, Ni, Co in the current alloy plausibly leads austenite to be the stable phase at room temperature. The increased carbon content of the retained austenite could be related to the effect of Co and Al addition [42] which might accelerate to expel carbon from martensite into the coexisting austenite during partitioning. It is known that carbon enrichment and interaction between micro-constituents plays a crucial role in austenite stabilisation [43]. According to Jacques et al. [44], the stability of retained austenite depends on the strength and dislocation density of surrounding ferrite-bainite matrix of a multiphase microstructure. The presence of retained austenite in the Q&P microstructure can leads to an interesting mechanical property i.e., good formability, as a result of the TRIP effect from the retained austenite. Consequently, strength higher than that of conventional TRIP steels could be obtained due to the presence of martensite instead of bainite. Higher value of carbon enriched (1.40%) metastable retained austenite (10.50%) in case of FA at 930 °C, followed by Q&P process is considered to be beneficial as TRIP phenomenon during deformation can contribute to ductility and toughness. 3.4. Mechanical properties The mechanical properties of the investigated steels subjected to TMCP with 750 °C FRT and PA and FA prior to Q&P at 345 °C with partitioning time 30 min followed by AC are summarised in Table 3 and Fig. 7. Here total area under the stress-strain curve has been considered as the measure of tensile toughness [45]. The area under the curve has been mathematically approximated by the average of the yield and ultimate strengths, multiplied by the strain to fracture and the machine generated values of tensile toughness is shown in Table 3. It is evident that samples subjected to Q&P exhibit a better combination of high tensile strength (921– 922 MPa), good total elongation (25–26%) and tensile toughness (199–230 MPa). On the other hand, only TMCP processed sample exhibits good strength (1099 MPa) and moderate elongation (14%) and tensile toughness (137 MPa). It is clear that the microstructure

and thereby mechanical properties could be tailored as a function of austenitisation temperature (PA & FA). The FA processed samples have shown maximum total elongation and tensile toughness values compared to other samples. The lower elongation of PA compared to FA can be explained by their microstructural constituents. However, the difference of elongation is marginal (Table 3). The higher fraction of ferrite in PA is reflected in its lower yield strength. In general, lower hardness and strength along with superior ductility and toughness of Q&P steel are attributed to the complicated cumulative effect of solid solution of interstitial and substitutional atoms, decrease in carbon super-saturation in martensite, change in dislocation density, increase in the fraction of carbon enriched retained austenite and good amount of ferrite, obtained prior to quenching in one-step Q&P treatment. The absence of iron carbide in the present steel due to the chosen chemical composition might contribute to the substantial ductility. The PSE values of the samples subjected to various processes lie in the range of 15.38 GPa%–23.94 GPa% and the maximum value is obtained from the sample subject to FA followed by Q&P. In addition, samples subjected to FA prior to Q&P have been also exhibited significantly higher value of tensile toughness (230 MPa), which is nearly 16% higher than that of samples processed at PA temperature (199 MPa). Fig. 7 shows the stress–strain curves of the investigated steel specimens which exhibits absence of yield point elongation and continuous elastic to plastic transition. The continuous yielding behaviour is attributed to pearlite free microstructural evolution. Bainite also gives similar effect like pearlite but since amount of martensite is higher than bainite, the continuous yielding is seen. The tensile stress-strain curves have been further analysed to understand the strain hardening behaviour of the investigated steels. Fig. 8 presents the strain hardening rate (ds/dε) as a function of true strain (ε) curves which correspond to the engineering stress-strain curves (Fig. 7). From the strain hardening curve (Fig. 8), it is indicated that the work hardening behaviour indicates three distinct stages, denoted as stage I, stage II and stage III. The stage I represents the beginning of yielding where the strain hardening rate drops rapidly with true strain in both cases

1200 1050

Stress (MPa)

900 750 600 TMCP+AC 450

TMCP+PA+Q&P+AC TMCP+FA+Q&P+AC

300 150 0 0

5

10

15

20

25

30

Strain (%) Fig. 7. Tensile stress–strain curves obtained after various processing conditions.

Fig. 8. Strain hardening rate of experimental steels as a function of true strain.

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

Table 4 Typical mechanical property ranges for current-generation Q&P steels. Steel/Processing condition

YS (MPa)

UTS (MPa)

Total elongation (%)

Q&P 980 [28] Q&P 1180 [28] Present investigated Q&P steel

650–800 950–1150 584–641

980–1050 1180–1300 921–922

17–22 8–14 25–26

i.e., only TMCP and also TMCP followed by Q&P, because of the high density dislocation (which interaction in martensite or bainite) cancellation and rearrangement due to increasing of deformation [46,47]. In stage II, strain hardening decreases slowly with strain for only TMCP processed steel, whereas it becomes almost a region of plateau before necking for TMCP followed by Q&P steel due to TRIP effect of retained austenite which will transforms partially to a new martensite phase with increasing the deformation. During phase transformation (γ-α′), concentrated local stress has transformed from high to low stress zone, as a results initiation of micro-cracks is delay and improved the ductility of investigated steel [48,49]. In stage III, retained austenite transformation was completed and the necking will start. Finally, strain hardening rate has been reduced at higher strains due to the onset of dynamic recovery. Similar type of strain hardening curve has been reported earlier in case of tensile deformation behaviour of ferrite-martensite dual phase steels [50,51]. Ishikawa et al. [52] reported that the higher uniform elongation for ferrite-bainite

63

dual-phase steels can be related to the strain hardening behaviour of each phase (ferrite and bainite). It is important to note that the strain hardening rate in all steels does not decrease to zero at failure which indicates that the failure is occurred by necking [53]. Mechanical properties of industrially produced Q&P steels were reported by Wang and Speer [28] as shown in Table 4, which indicates that the present experimental Q&P steels, having good balance of strength, ductility and tensile toughness values. The lower yield ratio (0.63-0.69) and high strength along with superior ductility, obtained in the present investigated steels after Q&P process also ensures high energy absorption capacity of the steel (till fracture) which improves the trustworthiness of the steel during operation in service. The current steel with its low carbon content (0.20 wt%) can achieve not only higher hardening level but also maintain reasonable plasticity, formability and weldability. 3.5. Fractography analysis Fig. 9(a)–(c) show the SEM micrographs of fracture surfaces of the investigated steel subjected to TMCP with 750 °C FRT followed by AC and TMCP followed by PA (800 °C) & FA (930 °C) followed by Q&P (345 °C, 30 min) and AC respectively. All fracture surfaces exhibit a nature of mixed mode of fracture i.e., ductile and brittle comprising cleavage facets and dimples with some inter-granular cracks indicating considerable amount of deformation before fracture. The presence of fine dimple in the Fig. 9(c) indicates

Fig. 9. SEM micrographs of fracture surfaces of the investigated steel subjected to (a) TMCP and AC, (b) TMCP and PA followed by Q&P plus AC and (c) TMCP and FA followed by Q&P plus AC.

64

G. Mandal et al. / Materials Science & Engineering A 676 (2016) 56–64

higher ductility as a consequence of micro-void coalescence. The same micrograph is also apparently showing more dimples, indicating higher amount of energy absorbed by nucleation and growth of micro-voids before fracture. This can be substantiated by the maximum available tensile toughness value (230 MPa) compared to that of other samples (Table 3).

4. Conclusions The major conclusions drawn from the present study are listed below: 1. The Q&P steels with the partial and full austenitisation processing routes reveals pearlite-free multiphase microstructures comprising ferrite and lath microstructures (martensite and bainite) along with retained austenite. 2. Partitioning at 345 °C for 30 min leads to good carbon enrichment (4 1 wt%) of the retained austenite which might be due to the fast partitioning kinetics and to the possibility of the suppression of carbides by the Si and Al additions. Carbon partitioning may be considered as a dominant process at 30 min of partitioning time, leading to a good amount of retained austenite (8–10%). It is evident that the full austenitisation compared to partial austenitisation leads to a more positive effect on the stability of retained austenite. 3. The attractive combination of strength (921–922 MPa) and ductility (25–26% total elongation) along with low yield ratio (0.63–0.69) is attributed to the multiphase microstructure with good amount of carbon enrichment in retained austenite. 4. All the specimens reveal three-stage strain hardening behaviour and mixed mode of fracture involving ductile and brittle flat facets. The higher value of tensile toughness of the samples subjected to full austenitisation is substantiated by more dimples and less inter-granular cracking. 5. The current quenching and partitioning (Q&P) steel can be considered as a potential choice for third generation automotive applications.

Acknowledgement The authors of IIEST, Shibpur gratefully acknowledge the financial support provided by Tata Steel Limited (Grant number:3000084687/102), Jamshedpur, India, in the present investigation.

References [1] B.C. De Cooman, Curr. Opin. Solid State Mater. Sci. 8 (2004) 285–303. [2] A.Z. Hanzaki, P.D. Hodgson, S. Yue, Metall. Mater. Trans. A 28 (1997) 2405–2414. [3] C. Garcia-Mateo, F.G. Caballero, J. Chao, C. Capdevila, C.G. De Andres, J. Mater. Sci. 44 (2009) 4617–4624. [4] S.K. Putatunda, Mater. Sci. Eng. A 297 (2001) 31–43. [5] H.K.D.H. Bhadeshia, R. Honeycombe, Steels: Microstructure and Properties, Butterworth-Heinemann, Oxford, 2006. [6] W. Steven, A.G.J. Haynes, J. Iron Steel Inst. 183 (1956) 349–359.

[7] G. Krauss, Metall. Mater. Trans. A 32 (2001) 861–877. [8] J.G. Speer, D.K. Matlock, B.C. De Cooman, J.G. Schroth, Acta Mater. 51 (2003) 2611–2622. [9] D.K. Matlock, V.E. Bräutigam, J.G. Speer, Mater. Sci. Forum 426–432 (2003) 1089–1094. [10] J. Gordine, I. Codd, J. Iron Steel Inst. 207 (1969) 461–467. [11] D.V. Edmonds, K. He, F.C. Rizzo, B.C. De Cooman, D.K. Matlock, J.G. Speer, Mater. Sci. Eng. A 438–440 (2006) 25–34. [12] S.S. Nayaka, R. Anumolu, R.D.K. Misra, K.H. Kim, D.L. Lee, Mater. Sci. Eng. A 498 (2008) 442–456. [13] H.Y. Li, X.W. Lu, X.C. Wu, Y.A. Min, X.J. Jin, Mater. Sci. Eng. A 527 (2010) 6255–6259. [14] A.M. Streicher, J.G. Speer, D.K. Matlock, B.C. De Cooman, in: J.G. Speer (Eds.), Proceedings of the International Conference on Advanced High-strength Sheet Steels for Automotive Applications, Warrendale, PA: AIST, 2004, pp. 51–62. [15] C.J. Wang, Chang, Y.C. Chang, Mater. Chem. Phys. 76 (2) (2002) 151–161. [16] H.Y. Li, X.W. Lu, W.J. Li, X.J. Jin, Metall. Mater. Trans. A 41A (2010) 1284–1300. [17] H.R. Ghazvinloo, A. Honarbakhsh-Raouf, A.R.K. Rashid, Metallurgist 59 (2015) 90–96. [18] K. Zhang, P. Liu, W. Li, Z. Guo, Y. Rong, Mater. Sci. Eng. A 619 (2014) 205–211. [19] X. Tan, Y. Xu, X. Yang, D. Wu, Mater. Sci. Eng. A 589 (2014) 101–111. [20] F. Blekkenhorst, G.M. Ferrari, C.J.V.D. Wekken, F.P. Ijsseling, Br. Corros. J. 21 (3) (1986) 163–176. [21] G. Mandal, S.K. Ghosh, S. Mukherjee, J. Mater. Sci. 51 (2016) 6569–6582. [22] F.G. Caballero, S. Allain, J. Cornide, J.D.P. Vela´squez, C. Garcia-Mateo, M. K. Miller, Mater. Des. 49 (2013) 667–680. [23] H.K.D.H. Bhadeshia, R.W.K. Honeycombe, Steels : Microstructure and Properties, Third edition, Butterworth-Heinemann, Oxford, 2006. [24] G. Dini, A. Najafizadeh, S.M. Monir-Vaghefi, R. Ueji, J. Mater. Sci. Technol. 26 (2) (2010) 181–186. [25] E. Girault, P. Jacques, Ph. Harlet, K. Mols, J. Van Humbeeck, E. Aernoudt, F. Delannay, Mater. Charact. 40 (1998) 111–118. [26] J. Wang, G. Li, A. Xiao, Mater. Trans. JIM 52 (11) (2011) 2027–2031. [27] M.J. Santofimia, L. Zhao, J. Sietsma, Metall. Mater. Trans. A 40A (2009) 46–57. [28] Li Wang, J.G. Speer, Metallogr. Microstruct. Anal. 2 (2013) 268–281. [29] G.A. Thomas, J.G. Speer, D.K. Matlock, Metall. Mater. Trans. A 42A (2011) 3652–3659. [30] G.H. Gao, H. Zhang, X.L. Gui, P. Luo, Z.L. Tan, B.Z. Bai, Acta Mater. 76 (2014) 425–433. [31] Y.F. Shen, L.N. Qiu, X. Sun, L. Zuo, P.K. Liaw, D. Raabe, Mater. Sci. Eng. A 636 (2015) 551–564. [32] L. Yuan, D. Ponge, J. Wittig, P. Choi, J.A. Jimenez, D. Raabe, Acta Mater. 60 (2012) 2790–2804. [33] G.H. Gao, H. Zhang, Z.L. Tan, W.B. Liu, B.Z. Bai, Mater. Sci. Eng. A 559 (2013) 165–169. [34] K. Zhang, M.H. Zhang, Z.H. Guo, N.L. Chen, Y.H. Rong, Mater. Sci. Eng. A 528 (2011) 8486–8491. [35] Y. Wang, K. Zhang, Z.H. Guo, N.L. Chen, Y.H. Rong, Mater. Sci. Eng. A 552 (2012) 288–294. [36] D.P. Koistinen, R.E. Marburger, Acta Met. 7 (1959) 59–60. [37] H.K.D.H. Bhadeshia, D.V. Edmonds, Acta Mater. 28 (1980) 1265–1273. [38] C. Liu, Z. Zhao, D.O. Northwood, Y. Liu, J. Mater. Proc. Tech. 113 (2001) 556–562. [39] J.B. Nelson, D.P. Riley, Proc. Phys. Soc. Lond. 57 (1945) 160. [40] B.D. Cullity, S.R. Stock, Elements of X-Ray Diffraction, 3rd edition, Prentice Hall, Inc, Upper Saddle River, NJ, USA, 2001. [41] C. Garcia-Mateo, F.G. Caballero, M.K. Miller, J.A. Jimenez, J. Mater. Sci. 47 (2012) 1004–1010. [42] F. Hu, K.M. Wu, H. Zheng, Steel Res. Int. 84 (10) (2013) 1060–1065. [43] S.M.K. Hosseini, A. Zarei-Hanzaki, S. Yue, J. Mech. Eng. Automot. 3 (2013) 650–655. [44] P. Jacques, X. Cornet, pH Harlet, J. Ladrière, F. Delannay, Metall. Trans. A 29 (9) (1999) 2383–2393. [45] G.E. Dieter, Mechanical Metallurgy, SI Metric Edition, McGraw-Hill Book Company, Singapore 1988, p. 283. [46] H. Rastegari, A. Kermanpur, A. Najafizadeh, Mater. Sci. Eng. A 632 (2015) 103–109. [47] S. Zhu, Y.L. Kang, K.K. Ren, S.C. Li, Adv. Mater. Res. 299–300 (2011) 403–407. [48] M.G. Lee, S.J. Kim, H.N. Han, Int. J. Plast. 26 (2010) 688–710. [49] K. Sugimoto, N. Usui, M. Kobayashi, S. Hashimoto, ISIJ Int. 32 (1992) 1311–1318. [50] T.S. Byun, I.S. Kim, J. Mater. Sci. 28 (11) (1993) 2923–2932. [51] Z. Xiurong, C. Yunbo, W. Miaohui, Mater. Res. 15 (6) (2012) 915–921. [52] N. Ishikawa, K. Yasuda, H. Sueyoshi, S. Endo, H. Ikeda, T. Morikawa, K. Higashida, Acta Mater. 97 (2015) 257–268. [53] F. Ozturk, A. Polat, S. Toros, R.C. Picu, JIS Res. Int 20 (6) (2013) 68–74.