Yielding Behavior of Low Carbon Martensitic Steel Sheet Containing Retained Austenite

Abstract

Quenching and Partitioning (Q&P) steel sheets, which utilize the transformation induced plasticity (TRIP) effect of retained austenite to improve the elongation of high strength steel sheets, are expected to become an important material for next-generation automotive structural parts. Although it has been reported that the yield strength (YS) of the Q&P steels consisting of tempered martensitic microstructure with retained austenite (hereafter ”Q&P steels” in this study) is affected by retained austenite, the mechanism has not yet been discussed in detail. The purpose of this study is to clarify the effect of the carbon content in retained austenite on the yielding behavior of the Q&P steels. The chemical composition of the model steel used here was 0.18%C-1.5%Si-3.0%Mn (mass%). The steels were annealed at 1143 K, then cooled to 473 K, followed by holding at the temperatures between 523 K and 673 K for 600 s. The increased carbon content in retained austenite increased the YS of the Q&P steels. It was found that the yielding of the Q&P steels was caused by the stress-induced transformation of retained austenite when the critical stress for the stress-induced transformation was lower than the elastic limit of tempered martensitic matrix. This result revealed that the increased carbon content in retained austenite was able to achieve the higher elastic limit of martensitic steels containing retained austenite.

1. Introduction

High strength steel sheets have been applied to automotive structural parts in order to satisfy both reduction of CO₂ emissions by reducing automobile weight and improvement of crashworthiness by increasing auto body strength. In order to apply high strength steel sheets to a variety of automotive parts, maximum tensile strength (TS) of high strength steel sheets is generally necessary to be increased while maintaining the same elongation as conventional steel sheets.^1,2,3,4) Moreover, recent studies have reported that the increased yield strength (YS) in steel sheets is able to increase the crashworthiness in body frame parts.⁵⁾

One heat cycle which is capable of realizing a complex phase of both tempered martensite and retained austenite in steel is Quenching and Partitioning (Q&P) treatment.^1,6,7) The Q&P treatment is a combined heat treatment consisting of interrupted quenching to a temperature between the martensite transformation start temperature (M_s) and the martensite transformation finish temperature (quenching process) and subsequent thermal treatment that enhances the carbon enrichment from martensite into the untransformed austenite (partitioning process). It has been widely reported that Q&P steel sheets have an improved balance of TS and elongation than quenching and tempering (Q&T) steel sheets^1,8,9,10,11) whose microstructure is tempered martensite without retained austenite. However, few studies have investigated the yielding behavior of the Q&P steels consisting of tempered martensite with retained austenite (hereafter referred to as just “Q&P steels” in this study). For example, Wang et al.¹¹⁾ confirmed that the YS of a Q&P steel was lower than that of a Q&T steel but did not study the mechanism in detail. Toji et al.,¹²⁾ one of the co-authors, reported that the YS of a Q&P steel was higher than that of the austempered steel mainly consisting of bainitic ferrite and retained austenite based on a comparison of specimens having the same chemical composition and similar TS. Although it was suggested that the solute carbon content in retained austenite (C_γ) of the Q&P steel influenced their YS, the mechanism was not clarified in detail. Bagliani et al.¹³⁾ reported that the YS of a Q&P steel was inversely related to the cooling stop temperature (T_CS) in the Q&P treatment, and also noted that the YS was related to the volume fraction of fresh martensite which transformed from austenite during the final water cooling. However, since it is known that the carbon content in untransformed austenite after the partitioning treatment affects the volume fraction of fresh martensite after the Q&P treatment,¹⁴⁾ it is difficult to separate the effects of C_γ and fresh martensite on the YS based on their results.

It is known that the yielding of dual-phase steels consisting of ferrite and martensite depends on the slip deformation of the soft ferrite phase.^15,16) On the other hand, it is considered that the yielding of the Q&P steels is affected not only by the slip deformation of tempered martensite and retained austenite but also by the stress-induced transformation of retained austenite.^17,18) Stress-induced transformation is one of the deformation-induced transformation and is explained as the transformation phenomenon in the elastic deformation region.¹⁷⁾ Tsuchida et al.¹⁹⁾ recently confirmed that the deformation-induced transformation and the slip deformation of retained austenite occurred during the yielding of the Q&P steels using in situ neutron diffraction. They also reported that the slip deformation of retained austenite caused the yielding of the Q&P steels because the transformation strain by the deformation-induced transformation was slight in their Q&P steels. On the other hand, there have been several reports concerning the mechanism of the yielding by the stress-induced transformation of retained austenite. Haidemenopoulos et al.¹⁸⁾ investigated the effect of tensile test temperature on the yielding behavior of the tempered martensitic steel containing 9 vol.% retained austenite using SNCM439 (Fe-0.4%C-0.28%Si-0.46%Mn-Ni, Cr, Mo), and strongly suggested that the yielding depended on the stress-induced transformation of retained austenite at tensile test temperatures below 313 K. De Cooman et al.²⁰⁾ reported that the yielding of medium Mn steels was caused by the stress-induced transformation of retained austenite when the contents of Mn and C in retained austenite were low, which meant that the stability of retained austenite decreased. These reports suggest that the stress-induced transformation of retained austenite may influence the yielding behavior of low carbon martensitic steel sheets containing retained austenite, such as Q&P steels.

The purpose of this study is to clarify the effect of C_γ on the yielding behavior of the Q&P steels, by using the Q&T steels and the Q&P steels containing almost the same volume fraction of retained austenite with different C_γ. The change in the elastic limit of the Q&P steels was discussed in terms of the slip deformation of martensitic matrix and the stress-induced transformation of retained austenite.

2. Experimental Procedure

The chemical composition of the model steel used in this study is shown in Table 1. In specimen preparation, a vacuum-melted ingot was forged to a thickness of 30 mm, and the resulting slabs were reheated to 1473 K for 3.6 × 10³ s and hot-rolled in 3 passes to a thickness of about 4.0 mm at a target finish temperature of 1173 K. The hot-rolled sheets were then cooled to 923 K, which corresponds to the coiling temperature, by immersion in an alumina fluidized-bed furnace at room temperature. These sheets were held at 923 K for 3.6 × 10³ s in an electric furnace, followed by furnace cooling. After pickling, the sheet thickness was reduced to 2.8 mm by grinding both surfaces to remove scale and surface defects, and the ground sheets were cold-rolled to 1.4 mm by 50% cold-rolling reduction. As shown in Fig. 1, (a) the Q&P treatment and (b) the Q&T treatment were conducted in alumina fluidized-bed furnaces.

Table 1. Chemical composition of model steel used in this study.

(mass%)
C	Si	Mn	P	S	Al	N
0.179	1.51	3.01	0.021	0.0027	0.041	0.0045

Fig. 1. Thermal histories of (a) Quenching and Partitioning (Q&P) treatment and (b) Quenching and Tempering (Q&T) treatment. W.C.: Water cooling.

The M_s of this steel was measured as 618 K by the thermal expansion test. In the Q&P treatment, the steel sheets were annealed at 1143 K for 180 s in the austenite single-phase region. The annealed sheets were quenched at about 15 K/s to either 473 K or 523 K, and then immediately heated at a heating rate of about 10 K/s to various partitioning temperatures. After holding at 523 K, 573 K, 623 K, 673 K and 723 K for 600 s, the sheets were water-cooled to room temperature. In the Q&T treatment, the sheets were annealed at 1143 K for 180 s, followed by water-cooling to room temperature. The subzero treatment at 75 K for 3.6 × 10³ s was conducted to minimize the retained austenite in the Q&T steel. The sheets were then heated at a heating rate of about 10 K/s, and the tempering was carried out at 523 K, 573 K, 623 K and 673 K for 600 s, followed by water-cooling to room temperature. The tempering temperatures were set to the same as the partitioning temperatures of the Q&P treatment. To keep the retained austenite in the as-heat-treated condition, a skin-pass rolling and leveling were not applied to any of the sheets.

The microstructures of these steel sheets were observed using a scanning electron microscope (SEM) (JEM 7100F; manufactured by JEOL Ltd.) and electron back-scatter diffraction (EBSD) on the transverse direction (TD) plane at the quarter-thickness position. The SEM analysis was performed at an acceleration voltage of 15 kV. EBSD was also used to observe the distribution of retained austenite, and the EBSD patterns were analyzed with an OIM Analysis v8 system (manufactured by TSL Solutions Co., Ltd.). The volume fraction of retained austenite and C_γ were measured by X-ray diffraction (XRD) (SmartLab; manufactured by Rigaku Corporation). The samples for XRD were ground from the surface to one quarter of the thickness of the heat-treated steel, and another 100 μm was subsequently removed from the ground surface using a mixture of oxalic acid and hydrogen peroxide water to exclude the influence of strain introduced by the grinding. The volume fraction of retained austenite was quantified based on the intensity of (2 0 0)_α, (2 1 1)_α, (2 2 0)_α, (2 0 0)_γ, (2 2 0)_γ and (3 1 1)_γ reflections.²¹⁾ The lattice parameter can be derived based on Bragg’s law and lattice spacing.²²⁾ C_γ was calculated from the lattice parameter a_γ (Å) obtained from the γ reflections using the following equation,²³⁾ which was a combination of the equations of Ruhl et al.²⁴⁾ for C, Mn and Si and Dyson et al.²⁵⁾ for Al:

  

$$\begin{array}{l}{\mathrm{C}}_{\gamma }=({\mathrm{a}}_{\gamma }-3.572-0.0012\left[\mathrm{M}\mathrm{n}\right]\\ \hspace{1em}\hspace{0.5em}+0.00157\left[\mathrm{S}\mathrm{i}\right]-0.0056\left[\mathrm{A}\mathrm{l}\right])/0.033\end{array}$$(1)

where [Mn], [Si] and [Al] were the concentrations of Mn, Si and Al in austenite (in mass%), respectively. These substitutional elements were regarded as identical to the chemical composition of the alloy since the maximum partitioning temperature in this study was 723 K and the long-range diffusion of these substitutional elements was difficult during the partitioning treatment. For tensile tests, JIS No. 5 tensile test pieces (gauge length: 50 mm, width of parallel part: 25 mm) were taken so that the tensile direction was the TD. The tensile tests were conducted at a constant crosshead speed of 10 mm/min at room temperature (297 K). After applying various levels of stress at the same crosshead speed, followed by unloading, the change in the amount of retained austenite during the tensile test was measured by XRD.

3. Experimental Results

3.1. Effect of Heat Treatment Conditions on Retained Austenite Formation Behavior of Q&P Steel

In this section, the effects of T_CS and the partitioning temperature in the Q&P treatment on the volume fraction of retained austenite and C_γ were investigated.

Figure 2 shows the change in the volume fraction of retained austenite in the Q&P steels with T_CS of 473 K and 523 K as a function of the partitioning temperature. The volume fraction of retained austenite in the Q&P steels with T_CS of 523 K increased with the partitioning temperature until 673 K, but then decreased when the partitioning temperature exceeded 673 K. At T_CS of 473 K, the volume fraction of retained austenite was almost the same independent of the partitioning temperature. Figure 3 shows the change in C_γ in the Q&P steel with T_CS of 473 K and 523 K as a function of the partitioning temperature. These C_γ were almost the same at both T_CS. C_γ increased as the partitioning temperature increased, but then decreased when the partitioning temperature exceeded 673 K. It was suggested that the increase in C_γ with increasing the partitioning temperatures below 623 K was related to the higher diffusion rate of carbon.⁷⁾ On the other hand, C_γ decreased with increasing partitioning temperatures above 623 K, suggesting that carbide precipitation increased.^26,27,28,29) The change in the volume fraction of retained austenite in the Q&P steel with T_CS of 523 K showed a good correlation with the change in C_γ. This indicates that the volume fraction of retained austenite increases due to the improved stability of untransformed austenite at the partitioning temperatures below 673 K. However, at the partitioning temperature of 673 K, the volume fraction of retained austenite decreases because C_γ is decreased. At T_CS of 473 K, the volume fraction of retained austenite was almost constant (about 7.5 vol%) when the partitioning temperature was increased. This result did not correspond to the change in C_γ. The relationship between the volume fraction of retained austenite and C_γ in the Q&P steel with T_CS of 473 K is explained as follows: In the Q&P treatment, the Koistinen-Marburger equation³⁰⁾ given by Eqs. (2) and (3) is known as a method for estimating the amount of untransformed austenite at T_CS.

  

$$f^{\alpha \text{'}}\left[\text{vol}\text{.\%}\right]=100-\text{100}\times exp\left\{-\alpha \left(\text{Ms}-{\text{T}}_{\text{CS}}\right)\right\}$$(2)

  

$$f^{\gamma }\left[\text{vol}\text{.\%}\right]=100-f^{\alpha \text{'}}$$(3)

where, f^α′ and f^γ are the amounts of martensite and untransformed austenite when a sample is quenched, respectively. M_s was 618 K and the value of α was set at 0.0184 based on Eq. (4), which was reported by Bohemen et al.³¹⁾

  

$$\mathrm{\alpha }=0.0224-0.0107\left[\mathrm{C}\right]-0.0007\left[\mathrm{M}\mathrm{n}\right]$$(4)

where [C] is the concentration of C in this alloy (in mass%). Figure 4 shows the relationship between T_CS and the untransformed austenite volume fraction at T_CS for this model steel composition as calculated by Eqs. (2), (3), (4), together with the experimental values for the volume fraction of retained austenite. The experimental value of the volume fraction of retained austenite in the Q&P steel with T_CS of 523 K was smaller than the calculated amount of untransformed austenite. This result indicated that the untransformed austenite at T_CS was transformed to bainite during the partitioning treatment or to martensite during the final water cooling.^6,26) On the other hand, the volume fraction of retained austenite in the Q&P steel with T_CS of 473 K was almost the same as the calculated amount of untransformed austenite. This means that the amount of untransformed austenite at T_CS is not decreased by the bainite transformation during the partitioning treatment or the martensite transformation during the final water cooling.

Fig. 2. Change in the volume fraction of retained austenite in the Q&P steels as a function of the partitioning temperature. T_CS: Cooling stop temperature.

Fig. 3. Change in the carbon content in retained austenite in the Q&P steels as a function of the partitioning temperature. T_CS: Cooling stop temperature.

Fig. 4. Relationship of the volume fraction of calculated untransformed austenite at T_CS before partitioning process and retained austenite after the Q&P treatment depending on T_CS. T_CS: Cooling stop temperature.

From these results, it was found that the Q&P steels with T_CS of 473 K provided the same volume fraction of retained austenite in the tempered martensite matrix with different C_γ. Thus, these specimens are suitable for investigating the effect of C_γ on the elastic limit of the Q&P steels. The Q&P steels with T_CS of 523 K were not used in the following section.

3.2. Comparison of Microstructures after Q&P Treatment and Q&T Treatment

Table 2 shows the heat treatment conditions, the volume fraction of retained austenite and C_γ of each specimen for the Q&P steels with T_CS of 473 K. The Q&P steels were used in the tensile test in the following section 3.3. The Q&P steels contained almost the same volume fraction of retained austenite with different C_γ. The volume fraction of retained austenite in the Q&T steels was found to be about 0%. Figure 5 shows SEM images of the Q&P steels with the partitioning temperatures of 523 K and 673 K and the Q&T steels with the same tempering temperatures. The SEM microstructures of the Q&P steels and the Q&T steels looked like the typical tempered martensitic microstructure, and the prior austenite grain size of these steels was about 15 μm. Since it was difficult to identify the retained austenite in the Q&P steels from the SEM images, an EBSD crystal orientation analysis was carried out to investigate the distribution of retained austenite in the Q&P steels. These results are shown in Fig. 6, where (a), (c) and (e) show the inverse pole figure maps of martensite (bcc) phases, and (b), (d) and (f) show those of only austenite (fcc) phase. The retained austenite in the Q&P steels was distributed in martensite blocks and at the boundaries of the blocks and prior austenite grain boundaries. The volume fraction of retained austenite observed by EBSD was less than that measured by XRD in Table 2. This difference seems to be affected by C_γ, which related to the stability of retained austenite. It was found that the unstable retained austenite was transformed to martensite as a result of grinding in the sample preparation for the EBSD measurement when its stability was low.

Table 2. Effect of the cooling stop temperature and the partitioning temperature on the volume fraction of retained austenite and the carbon content in retained austenite after the Q&P treatment.

Cooling stop temperature (K)	Partitioning temperature (K)	Volume fraction of retained austenite (vol%)	Carbon content in retained austenite (mass%)
473	523	7.4	0.48
473	573	7.1	0.80
473	623	7.2	0.94
473	673	7.8	0.95

Fig. 5. SEM images of (a), (b) Q&P steels with the cooling stop temperature of 473 K and (c), (d) Q&T steels. PT: Partitioning temperature, TT: Tempering temperature.

Fig. 6. Inverse pole figure maps of the Q&P steels with the cooling stop temperature of 473 K. (a), (c), (e) bcc phase (martensite) with PT of 523, 623 and 673 K, and (b), (d), (f) fcc phase (retained austenite) with PT of 523, 623 and 673 K, respectively. PT: Partitioning temperature.

3.3. Comparison of Elastic Limits in Q&P Steel and Q&T Steel

Figure 7 shows the nominal stress-strain curves until TS of the Q&P steels and the Q&T steels. Here, the elastic limit is the point where the tensile deformation curve separates from the elastic deformation line based on the elastic modulus of steel (205 GPa). Regardless of the heat treatment pattern, all the nominal stress-strain curves were the continuous yielding type. The Q&P steel with the partitioning temperature of 523 K had a high TS of more than 1400 MPa, but its elastic limit was below 600 MPa, less than half of its TS. The elastic limit of the Q&P steels increased with increasing the partitioning temperature. Figure 8 shows the relationship between the TS and the elastic limit of the Q&P steels and the Q&T steels as a function of the partitioning or tempering temperatures. The TS of the Q&P steels and the Q&T steels decreased with increasing the partitioning and tempering temperatures, respectively. The elastic limit of the Q&T steels slightly increased with the tempering temperature until 573 K and then decreased. This tendency of the elastic limit of the Q&T steels in the tempering temperature region from 523 K to 673 K was similar to the result reported by Suto et al.³²⁾ On the other hand, the elastic limit of the Q&P steels increased as the partitioning temperature increased and gradually approached that of the Q&T steels. As a result, the elastic limit (1010 MPa) of the Q&P steel with the partitioning temperature of 673 K was close to the elastic limit (1092 MPa) of the Q&T steel tempered at the same temperature. Since the YS of the martensitic steels is generally evaluated by the 0.2% offset yield strength (hereafter “0.2%YS”), the relationship between the elastic limit and the 0.2%YS of the Q&P steels and the Q&T steels is shown in Fig. 9. In this figure, the 0.2%YS linearly increased as the elastic limit increased, indicating that the 0.2%YS displayed a positive correlation with the elastic limit in both of the Q&P steels and the Q&T steels.

Fig. 7. Nominal stress-strain curves until maximum tensile strength of (a) Q&P steels with the cooling stop temperature of 473 K and (b) Q&T steels at each PT or TT. PT: Partitioning temperature, TT: Tempering temperature.

Fig. 8. Comparison of (a) maximum tensile strength and (b) elastic limit of the Q&P steels with the cooling stop temperature of 473 K and the Q&T steels as a function of PT or TT. PT: Partitioning temperature, TT: Tempering temperature.

Fig. 9. Change in 0.2% yield strength of the Q&P steels with the cooling stop temperature of 473 K as a function of elastic limit.

These results revealed that the change in the elastic limit with increasing the partitioning or tempering temperatures was clearly different in the Q&P steels and the Q&T steels. Therefore, it was suggested that retained austenite in the tempered martensitic matrix strongly influenced the elastic limit of the Q&P steels.

3.4. Deformation-induced Transformation of Retained Austenite in Q&P Steels during the Tensile Test

From previous reports^18,19,20) and the comparison of the elastic limits of the Q&P steels and the Q&T steels in section 3.3, the deformation-induced transformation of retained austenite is thought to occur in the Q&P steels during the tensile test. Therefore, the deformation-induced transformation behavior of retained austenite in the Q&P steels was investigated by measuring the volume fraction of retained austenite in specimens that were loaded with various stresses using a tensile testing machine and then immediately unloaded. The volume fraction of deformation-induced transformation (f^M) was calculated from the measured results by using Eq. (5).

  

$$f^M\left[\text{vol}\text{.\%}\right]=f^{\gamma 0}-f^{\gamma \sigma }$$(5)

where, f^γ0 was the initial volume fraction of retained austenite and f^γσ was the volume fraction of retained austenite after stress loading. Considering the error of the measured values of retained austenite in the XRD measurements, f^M was judged to be valid if the value was 0.5 vol% or larger in this study.

Figures 10 and 11 show the change in the volume fractions of retained austenite and f^M, respectively, as a function of true stress. The elastic limit of each test sample are also shown in these figures. The deformation-induced transformation of retained austenite was confirmed in the Q&P steels with the partitioning temperatures of 623 K or less, as shown in Fig. 10. Since the deformation-induced transformation started below the elastic limit, it was considered that this transformation was the stress-induced transformation.¹⁷⁾ On the other hand, it was also found that the deformation-induced transformation of retained austenite did not occur with the partitioning temperature of 673 K, even under true stress of 1354 MPa (true strain: 4.1%).

Fig. 10. Changes in the volume fraction of retained austenite in the Q&P steels with the cooling stop temperature of 473 K as a function of true stress. PT: Partitioning temperature, EL: Elastic limit.

Fig. 11. Change in the volume fraction of stress-induced martensite in the Q&P steels with the cooling stop temperature of 473 K as a function of true stress. PT: Partitioning temperature, EL: Elastic limit, σ_M: Critical stress for the deformation-induced transformation of retained austenite.

Tsuchida et al.³³⁾ reported a linear relationship between f^M and true stress during tensile testing using in situ neutron diffraction. In this study, Fig. 11 also shows a linear relationship between f^M and true stress during the tensile test in the Q&P steels with the partitioning temperatures of 623 K or less. Therefore, the critical stress for the deformation-induced transformation of retained austenite (σ_M) was obtained by the extrapolation of the approximate line obtained by the least-squares method of this relationship. At the partitioning temperature of 523 K, the retained austenite began to decrease from a lower stress, and the σ_M was confirmed to be 580 MPa. The σ_M of the Q&P steels increased with increasing the partitioning temperatures until 623 K.

From these results described above, the deformation-induced transformation of retained austenite was confirmed in the Q&P steels with the partitioning temperatures of 623 K or less. On the other hand, it was also found that the deformation-induced transformation of retained austenite did not occur with the partitioning temperature of 673 K.

4. Discussion

In section 3.3, it was investigated that the effect of C_γ on the yielding behavior of the Q&P steels, by using the Q&T steels and the Q&P steels containing almost the same volume fraction of retained austenite with different C_γ. Furthermore, in section 3.4, the stress-induced transformation of retained austenite was confirmed in the Q&P steels with the partitioning temperatures of 623 K or less, but the deformation-induced transformation of retained austenite did not occur in the Q&P steel with the partitioning temperature of 673 K. It was considered that the yielding of the Q&P steels was affected not only by the slip deformation of tempered martensite and retained austenite but also by the stress-induced transformation of retained austenite.^17,18,20) Therefore, the change in the elastic limit of the Q&P steels will be discussed in terms of the slip deformation of martnsitic matrix and the stress-induced transformation of retained austenite in this section.

First, the yielding behavior of the Q&P steels with the partitioning temperatures of 523 K, 573 K and 623 K is discussed. Figure 12 shows the change in the elastic limit and the critical stress for the stress-induced transformation of retained austenite (${\sigma }_{\text{M}}^{\text{S}}$) of the Q&P steels as a function of C_γ. Since both the elastic limit and the ${\sigma }_{\text{M}}^{\text{S}}$ showed a significant correlation with increasing C_γ, it is estimated that the yielding of these steels was caused by the stress-induced transformation of retained austenite. However, Tsuchida et al.¹⁹⁾ reported that the yielding of Q&P steels occurred by the slip deformation of retained austenite based on the results of in situ neutron diffraction experiments. Therefore, the dominant factors in the yielding of the Q&P steels with the partitioning temperatures of 523 K and 623 K were investigated. As shown in Fig. 13(a), Olson¹⁷⁾ reported that at the temperatures below ${\text{M}}_{\text{s}}^{\sigma }$, the YS increased with increasing tensile test temperature because the stress-induced transformation provided the yielding. On the other hand, at the temperatures above ${\text{M}}_{\text{s}}^{\sigma }$, the YS decreased with increasing the tensile test temperature because the yielding was caused by the slip deformation of the matrix microstructure and the strain-induced transformation occurred in this temperature range. Figure 13(b) shows the change in the elastic limit of the Q&P steels with the partitioning temperatures of 523 K and 623 K as a function of the tensile test temperature. In both Q&P steels, the elastic limit increased with increasing the tensile test temperatures below 303 K, including room temperature. From Fig. 13(b), it was revealed that the yielding at room temperature of the Q&P steels with the partitioning temperatures of 523 K and 623 K was caused by the stress-induced transformation of retained austenite. As shown in Fig. 12, the ${\sigma }_{\text{M}}^{\text{S}}$ and the elastic limit of these steels linearly increased as C_γ increased. Therefore, in cases where the yielding occurs by the stress-induced transformation of retained austenite, it is considered that the change in C_γ can control the elastic limit of the martensitic steels containing retained austenite. However, not only C_γ but also the size,^34,35,36,37) the shape^33,34,35,36) and the crystallographic orientation^34,35,36) of retained austenite possibly influence the ${\sigma }_{\text{M}}^{\text{S}}$, hence the further investigation is needed to clarify the influence of these factors on the yielding behavior of the low carbon martensitic steel containing retained austenite.

Fig. 12. Change in the elastic limit and σ_M of the Q&P steels with the cooling stop temperature of 473 K and the partitioning temperature of 523 K to 623 K as a function of carbon content in retained austenite. σ_M: Critical stress for deformation-induced transformation of retained austenite.

Fig. 13. (a) Schematic illustration of the initial yielding by the stress-induced transformation and the initial yielding by the slip deformation in matrix(tempered martensite or retained austenite). (b) Change in the elastic limit of the Q&P steels with the cooling stop temperature of 473 K and the partitioning temperatures of 523 K and 623 K as a function of tensile test temperature. ${\sigma }_{\text{M}}^{\text{S}}$: Critical stress for the stress-induced transformation of retained austenite. TM: Tempered martensite, γ: austenite, PT: Partitioning temperature.

Next, the yielding behavior of the Q&P steel with the partitioning temperature of 673 K is discussed. Olson¹⁷⁾ reported that the slip deformation caused the yielding of steels when the ${\sigma }_{\text{M}}^{\text{S}}$ was higher than the YS of the matrix microstructure. Since the amount of retained austenite in the Q&P steel with the partitioning temperature of 673 K did not change until TS, it was thought that the yielding occurred by the slip deformation of tempered martensite or retained austenite. As shown in Fig. 8, the elastic limit of the Q&P steel with the partitioning temperature of 673 K was close the elastic limit of the Q&T steel tempered at the same temperature. Therefore, it was suggested that the yielding of the Q&P steel with the partitioning temperature of 673 K occurred by slip deformation of tempered martensite. However, the elastic limit of the Q&P steel was about 80 MPa lower than that of the Q&T steel. It is inferred that this difference occurred because the carbon content in the tempered martensitic matrix was decreased by the partitioning of carbon from martensite to austenite in the partitioning treatment.^1,6,7) Although the effect of the carbon content on the elastic limit of tempered martensite has not been fully understood, Kawata et al.³⁸⁾ reported that the elastic limit of as-quenched martensite increased as the carbon content increased in low-carbon 18Ni steel. Further investigation will be needed to clarify the reason why the deformation-induced transformation of retained austenite is not confirmed in the Q&P steel with the partitioning temperature of 673 K using, for example, in situ neutron diffraction experiments.

Figure 14 presents a schematic summary of the effect of C_γ on the yielding behavior of the low-carbon martensitic steel sheets containing retained austenite, represented by Q&P steels. When the ${\sigma }_{\text{M}}^{\text{S}}$ is lower than the elastic limit of tempered martensitic matrix, the yielding of the steel sheets is caused by the stress-induced transformation of retained austenite. In this case, the elastic limit of the steel sheets increases with increasing C_γ. On the other hand, when the ${\sigma }_{\text{M}}^{\text{S}}$ is higher than the elastic limit of tempered martensite, it is estimated that yielding occurs by the slip deformation of tempered martensite.

Fig. 14. Schematic illustration of the yielding mechanism in martensitic steel sheet containing retained austenite. ${\sigma }_{\text{M}}^{\text{S}}$: Critical stress for the stress-induced transformation of retained austenite, EL: Elastic limit of steel sheet, EL_TM: Elastic limit of tempered martensitic matrix, γ: austenite.

5. Conclusions

The effect of the carbon content of retained austenite (C_γ) on the elastic limit of the Q&P steels was investigated. The change in the elastic limit of the Q&P steels was discussed in terms of the slip deformation of tempered martensitic matrix and the stress-induced transformation of retained austenite. As a result, the following conclusions were obtained.

(1) The microstructure of the Q&P steels with the cooling stop temperature (T_CS) of 473 K consisted of tempered martensitic matrix containing almost the same volume fraction of retained austenite (about 7.5 vol%) with different C_γ.

(2) The elastic limit of the Q&P steels with T_CS of 473 K increased with increasing the partitioning temperature.

(3) The deformation-induced transformation of retained austenite was confirmed in the Q&P steels with the partitioning temperatures of 623 K or less. Since the deformation-induced transformation started below the elastic limit, it was considered that the transformation was occurred by the stress-induced transformation.

(4) The deformation-induced transformation of retained austenite did not occur in the Q&P steel with the partitioning temperature of 673 K.

(5) The elastic limits of the Q&P steels with the partitioning temperatures of 623 K or less showed a good correlation with the critical stress for the stress-induced transformation of retained austenite (${\sigma }_{\text{M}}^{\text{S}}$).

(6) When the ${\sigma }_{\text{M}}^{\text{S}}$ is lower than the elastic limit of tempered martensitic matrix, the yielding is caused by the stress-induced transformation of retained austenite. In this case, the elastic limit of the Q&P steels increases with increasing C_γ.

References

• 1)   J. G.  Speer,  E.  De Moor,  K. O.  Findley,  D. K.  Matlock,  B. C.  De Cooman and  D. V.  Edmonds: Metallurgical and Materials Transactions A, 42 (2011), 3591. https://doi.org/10.1007/s11661-011-0869-7
• 2)   M.  Takahashi: ISIJ Int., 55 (2015), 79. https://doi.org/10.2355/isijinternational.55.79
• 3)   H.  Minami,  T.  Kobayashi and  Y.  Funakawa: JFE Tech. Rep., 24 (2019), 23.
• 4)   T.  Tsuchiyama: J. Jpn. Inst. Met., 125 (2022), 20 (in Japanese). https://doi.org/10.1299/jsmemag.125.1239_20
• 5)   K.  Sato,  T.  Futatsuka,  T.  Sakaidani,  S.  Yoshika and  Y.  Tamai: Transactions of the Society of Automotive Engineers of Japan, Inc., 53 (2022), 675 (in Japanese). https://doi.org/10.11351/jsaeronbun.53.675
• 6)   J.  Speer,  D. K.  Matlock,  B. C.  De Cooman and  J. G.  Schroth: Acta Mater., 51 (2003), 2611. https://doi.org/10.1016/S1359-6454(03)00059-4
• 7)   J. G.  Speer,  D. V.  Edmonds,  F. C.  Rizzo and  D. K.  Matlock: Current Opinion in Solid State and Materials Science, 8 (2004), 219. https://doi.org/10.1016/j.cossms.2004.09.003
• 8)   Y.  Tamura: Tetsu-to-Hagané, 56 (1970), 429 (in Japanese).
• 9)   O.  Matsumura,  Y.  Sakuma and  H.  Takechi: Transactions of the Iron and Steel Institute of Japan, 27 (1987), 570. https://doi.org/10.2355/isijinternational1966.27.570
• 10)   A. M.  Streicher,  J. G.  Speer,  D. K.  Matlock and  B. C.  De Cooman: Int. Conf. on Advanced High-Strength Sheet Steels for Automotive Applications Proceedings, (2004), 51.
• 11)   C. Y.  Wang,  J.  Shi,  W. Q.  Cao and  H.  Dong: Mater. Sci. Eng.: A, 527 (2010), 3442. https://doi.org/10.1016/j.msea.2010.02.020
• 12)   Y.  Toji,  T.  Nakagaito,  H.  Matsuda,  K.  Hasegawa and  S.  Kaneko: ISIJ Int., 63 (2023), 758. https://doi.org/10.2355/isijinternational.ISIJINT-2022-508
• 13)   E. P.  Bagliani,  M. J.  Santofimia,  L.  Zhao,  J.  Sietsma and  E.  Anelli: Mater. Sci. Eng.: A, 559 (2013), 486. https://doi.org/10.1016/j.msea.2012.08.130
• 14)   J. G.  Speer,  F. C.  Rizzo Assunção,  D. K.  Matlock and  D. V.  Edmonds: Materials Research, 8 (2005), 417. https://doi.org/10.1590/S1516-14392005000400010
• 15)   Y.  Tomota,  S.  Morooka,  K.  Ikeda,  M.  Ojima,  S.  Harjo and  Y.  Adachi: Proc. of the 2nd Int. Sympo. on Steel Science (ISSS 2009), ISIJ, (2009), 27.
• 16)   S.  Morooka,  O.  Umezawa,  S.  Harjo,  K.  Hasegawa and  Y.  Toji: Tetsu-to-Hagané, 98 (2012), 311 (in Japanese). https://doi.org/10.2355/tetsutohagane.98.311
• 17)   G. B.  Olson: in Encyclopedia of Materials Science and Engineering, ed. by M. B. Bever, Pergamon Press, Oxford, (1986), 2929.
• 18)   G. N.  Haidemenopoulos,  M.  Grujicic,  G. B.  Olson and  M.  Cohen: Acta Metall., 37 (1989), 1677. https://doi.org/10.1016/0001-6160(89)90134-X
• 19)   N.  Tsuchida,  Y.  Wada,  H.  Minami and  Y.  Toji: Mater. Sci. Eng.: A, 873 (2023), 144989. https://doi.org/10.1016/j.msea.2023.144989
• 20)   B. C.  De Cooman,  P.  Gibbs,  S.  Lee and  D. K.  Matlock: Metall. Mater. Trans. A, 44A (2013), 2563.
• 21)   H.  Maruyama: J. Jpn. Soc. Heat Treat., 17 (1977), 198 (in Japanese).
• 22)   B. D.  Cullity, trans. by G. Matsumura: Elements of X-ray Diffraction, 2nd ed., AGNE, Tokyo, (1992), 298.
• 23)   J. R.  McDermid,  H. S.  Zurob and  Y.  Bian: Metall. Mater. Trans. A, 42 (2011), 3627. https://doi.org/10.1007/s11661-011-0678-z
• 24)   R.  Ruhl and  M.  Cohen: Trans. TMS-AIME, 245 (1969), 241.
• 25)   D. J.  Dyson and  B.  Holmes: J. Iron Steel Inst., 208 (1970), 469.
• 26)   A. J.  Clarke,  J. G.  Speer,  M. K.  Miller,  R. E.  Hackenberg,  D. V.  Edmonds,  D. K.  Matlock,  F. C.  Rizzo,  K. D.  Clarke and  E.  De Moor: Acta Materialia, 56 (2008), 16. https://doi.org/10.1016/j.actamat.2007.08.051
• 27)   M. J.  Santofimia,  T.  Nguyen-Minh,  L.  Zhao,  R.  Petrov,  I.  Sabirov and  J.  Sietsma: Mater. Sci. Eng.: A, 527 (2010), 6429. https://doi.org/10.1016/j.msea.2010.06.083
• 28)   T.  Tsuchiyama,  J.  Tobata,  T.  Tao,  N.  Nakada and  S.  Takakia: Mater. Sci. Eng.: A, 532 (2012), 585. https://doi.org/10.1016/j.msea.2011.10.125
• 29)   J.  Tobata,  K. L.  Ngo-Huynh,  N.  Nakada,  T.  Tsuchiyama and  S.  Takaki: ISIJ Int., 52 (2012), 1377. https://doi.org/10.2355/isijinternational.52.1377
• 30)   D. P.  Koistinen and  R. E.  Marburger: Acta Metall., 7 (1959), 59. https://doi.org/10.1016/0001-6160(59)90170-1
• 31)   S. M. C.  Van Bohemen and  J.  Sietsma: Metall. Mater. Trans. A, 40 (2009), 1059.
• 32)   S.  Suto: Denki Seiko (Electr. Furn. Steel), 37 (1966), 172. https://doi.org/10.4262/denkiseiko.37.172
• 33)   N.  Tsuchida,  T.  Tanaka and  Y.  Toji: ISIJ Int., 60 (2020), 1349. https://doi.org/10.2355/isijinternational.ISIJINT-2019-751
• 34)   D.  Adams,  M.  Behling,  M. P.  Miles,  E. R.  Homer,  A. K.  Sachbev,  E. V.  White and  D. T.  Fullwood: Metallurgical and Materials Transactions A, 54A (2023), 1355.
• 35)   G. K.  Tirumalasetty,  M. A.  van Huis,  C.  Kwakernaak,  J.  Sietsma,  W. G.  Sloof and  H. W.  Zandbergen: Acta Metallurgica, 60 (2012), 1311. https://doi.org/10.1016/j.actamat.2011.11.026
• 36)   D.  De Knijf,  C.  Fojer,  L. A. I.  Kestens and  R.  Petrov: Mater. Sci. Eng.: A, 638 (2015), 219. https://doi.org/10.1016/j.msea.2015.04.075
• 37)   A.  Basuki and  E.  Aernoudt: J. Mater. Process. Technol., 90 (1999), 37. https://doi.org/10.1016/S0924-0136(99)00037-0
• 38)   H.  Kawata,  Y.  Honda,  K.  Nakano,  K.  Takeda and  K.  Hikida: ISIJ Int., 62 (2022), 1502. https://doi.org/10.1590/S1516-14392005000400010