2016 Volume 56 Issue 11 Pages 2057-2061
Quenching and partitioning (Q&P) was successfully applied to a medium carbon and low alloy martensitic D6AC steel. Fully austenitized samples were quenched to temperatures below the Ms temperature (317°C), ranging from 240 to 275°C, and partitioned at the quench temperature for 300 s. While the as quenched sample exhibits a fully martensitic microstructure, all the Q&P processed samples have a retained austenite fraction up to about 6%. These samples were tensile tested at room temperature and their mechanical properties were compared to those obtained from different heat treated D6AC samples. Regardless of the partitioning temperature (equivalent to the quench temperature), the retained austenite in the Q&P processed samples transformed to martensite during the tensile testing, providing additional ductility above 10% of total elongation for the high strength material with a tensile strength around 1500 MPa.
Quenching and partitioning (Q&P) is a two-step (where the quench temperature differs from the partitioning temperature) or a one-step (where the quench temperature equals the partitioning temperature) heat treatment process proposed for carbon-bearing martensitic sheet steels to acquire desired combination of mechanical properties such as strength, ductility and toughness.1,2) Samples were initially heated to obtain a fully austenitic microstructure and quenched to a temperature between Ms and Mf, providing a mixed microstructure of austenite and martensite. The partitioning step was designed to re-distribute supersaturated carbon atoms in the martensitic matrix into the austenite regions. In comparison to samples for direct quenching, Q&P processed samples may exhibit increased fractions of retained austenite with its extra stability.
The Q&P process has been applied successfully to diverse advanced high strength steels (AHSS) for automotive applications. Clarke et al.2,3) confirmed the partitioning of carbon atoms into austenite for the Q&P processed 0.2C-1.6Mn-1.6Si steel, which shows the tensile strength values above 1 GPa and varied total elongations up to about 20%. Santofimia et al.4) applied the Q&P process to a partially austenitized 0.2C-1.6Mn-0.4Si-1.1Al steel. It shows a complex microstructure, which contains intercritical ferrite (presented from partial austenitization process), epitaxial ferrite (formed during the first quenching step), martensite, and retained austenite. They suggested that the carbon enrichment occurs mainly during the formation of epitaxial ferrite. De Moor et al.5) focused on the variations in the tensile properties of Q&P processed CMnSi steels having different carbon and manganese contents. In a 0.3C-3Mn-1.6Si steel, numerous heat treatment conditions provide combinations of tensile strength levels about 1500 MPa and total elongations exceeding 15%.
The Q&P process can be applied to other steel grades as well. Matlock et al.6) applied the Q&P process to a 0.35C-1.3Mn-0.74Si microalloyed bar steel with different processing parameters such as quenching temperature and partitioning time, and confirmed increased carbon content in the retained austenite. In addition, Paravicini Bagliani et al.7) compared the mechanical properties for a 0.3C-0.7Mn-1.4Si-1.5Cr-0.6Mo plate steel processed by Q&P and quenching and tempering (Q&T). It shows improved tensile strength and impact toughness for the Q&P processed samples compared to the results produced by Q&T.
D6AC steel is one of the steel grades specialized for aerospace and defense components such as landing gear, axles, shafts, and so on, which is typically produced by conventional Q&T. Its microstructure is mainly composed of tempered martensite, exhibiting high strength but limited ductility and toughness. Abbaszadeh et al.8) austempered D6AC samples to obtain a microstructure mixed with martensite and bainite for improved combination of mechanical properties. The formation of lower bainite microstructure improves both the elongation and impact toughness concurrently, whereas the increase of upper bainite deteriorates strength, ductility and toughness. Lee et al.9) compared the tensile properties of the samples experiencing different heat treatment processes such as as-quenched, tempered and spheroidized. The spheroidized specimens exhibit relatively low tensile strength below 1000 MPa, but the total elongations exceed 20%. Considering the limited ductility and toughness of the D6AC steel with conventional heat treatments, the Q&P process can be an alternative heat treatment process for its tailored combination of mechanical properties as well. Therefore, the objective of the present study is to apply the Q&P process to a D6AC steel and examine its microstructure and mechanical properties.
The D6AC steel is a medium carbon, low alloy steel possessing a relatively high fraction of carbide forming elements including Cr, Mo and V. Detailed chemical composition for the D6AC steel used in the present study is provided in Table 1. Bulk samples from the steel ingot were homogenized at 1200°C for 1 h and hot-rolled to produce sheet samples with a thickness of about 1 mm. No segregation was observed in the hot-rolled microstructure. The sheet samples were solution heat treated at 1000°C for 1 h using a tube furnace with an Ar gas atmosphere, and quenched by water. The aim is to obtain fully martensitic microstructure with minimized carbide precipitation. Note that the dissolution temperature of VC for the D6AC steel calculated by MatCalc (ver.5.62) was 899.4°C.
| C | Mn | Mo | P | S | Cu | Ni | Cr | Si | V | Fe |
|---|---|---|---|---|---|---|---|---|---|---|
| 0.425 | 0.787 | 1.03 | 0.007 | 0.006 | 0.017 | 0.549 | 1.024 | 0.274 | 0.089 | bal. |
The solution treated specimen was machined to a dimension of 3 × 1 × 10 mm3 and dilatometer tested to measure its martensite transformation kinetics including the Ms temperature. The specimen was heated to 880°C with a constant heating rate of 3°C/s, held for 10 min, and cooled to room temperature with a constant cooling rate of 100°C/s. Measured Ms temperature is about 317°C. The solution treated samples were austenitized at 880°C for 10 min using a tube furnace with an Ar gas atmosphere. These samples were quenched at four different temperatures below the Ms temperature, i.e., 275, 262, 253 and 240°C, and were partitioned at the quench temperature for 300 s, followed by water quenching. For comparison, one austenitized sample was quenched to room temperature without partitioning process (As-Q sample).
The heat treated samples were ground and polished to a final step using 1 μm diamond suspension. X-ray diffraction (XRD) analyses were performed using a Rigaku D/MAX 2500 diffractometer with a copper target operated at 40 kV and 30 mA. The 2θ scan range and scan speed employed were from 40 to 120° and 2° min−1, respectively. Retained austenite fractions for the samples were calculated from the peak intensities. Additional microstructure observations for selected Q&P processed samples were performed using an electron backscatter diffractometer (EBSD, EDAX, AMETEK, Inc), attached to a field emission-scanning electron microscope (FE-SEM) with an accelerating voltage of 30 kV, a working distance of 5 mm, and a step size of 20 nm. The results were analyzed using EDAX OIM software. Heat treated samples for the tensile tests were prepared as a sub-size specimen with a gauge length of 25.4 mm according to the ASTM-E8 standard.10) Tensile tests for the heat treated samples were performed with a strain rate of 1 × 10−3 s−1 at room temperature. Retained austenite fractions for the tensile fractured samples were measured at the region close to the fractured part in the gauge section in the same manner for the heat treated samples using XRD.
Figure 1(a) shows an on-cooling dilation curve for a solution treated sample of D6AC steel. The sample was fully austenitized at 880°C for 10 min and rapidly cooled to room temperature with a cooling rate of 100°C/s. There is a single expansion from a straight line related to the thermal contraction of austenite in the on-cooling strain curve, indicating the transformation of martensite from austenite. As mentioned in the experimental procedure, its Ms temperature was measured as 317°C. During the cooling to room temperature, martensite transformation was completed, which was confirmed by the XRD test for the As-Q sample, showing no measurable austenite peaks.
(a) On-cooling dilation curve with temperature for the D6AC steel sample austenitized at 880°C for 10 min and cooled to room temperature with a constant cooling rate of 100°C/s, and (b) martensite fraction as a function of quench temperature.
The volume fraction of martensite as a function of quench temperature was obtained by applying the lever rule due to a single transformation from austenite to martensite,11) which is shown in Fig. 1(b). The obtained martensite transformation kinetics were formulated using a following equation:
| (1) |
| Kinetic Parameter | Optimized Value |
|---|---|
| k1 | −1.2986 × 10−3 |
| k2 | 2.9241 × 10−3 |
| k3 | −1.3516 × 10−4 |
| k4 | 2.0221 × 10−6 |
| k5 | −1.6320 × 10−2 |
| k6 | 5.4051 × 10−5 |
| k7 | 1.5509 × 10−6 |
Figure 2 shows the predicted evolutions of phase fractions and austenite carbon content for the Q&P processed D6AC steel with quench temperature. Theoretical retained austenite fractions for the Q&P processed specimens were predicted using an approach proposed by Speer et al.2,3,5,12,13,14) This approach follows three assumptions: (1) full partitioning of carbon atoms between martensite and austenite during partitioning heat treatment; (2) no formation of diffusional phase transformation products such as cementite, transition carbides, ferrite, and bainite during the heat treatment; and (3) immobile interface between austenite and martensite. Initial samples were assumed to have a fully austenitic microstructure and the amount of martensite transformed at each quenching temperature was obtained from the dilatometric results. Carbon concentration of the partitioned austenite was calculated from the relative volume fractions of austenite (fγ(Q)) and martensite (fM(Q)) in the initial quench state.
The evolution of predicted phase fraction and austenite carbon content during the quenching and partitioning process of the D6AC steel. The fγ(Q), fM(Q), fγ(RT), and fM(RT) denote the fractions of initially quenched austenite and martensite at the quench temperature, and final quenched austenite and martensite to room temperature.
The Ms temperature for the carbon partitioned austenite was calculated using the following equation.15)
| (2) |
Part of the austenite was transformed to martensite during the secondary quenching to room temperature, and its fraction (fM(RT)) decreased at lower temperatures. The predicted maximum fγ(RT) of 0.48 was obtained at 236°C, where the fγ(Q) and its stability is balanced, resulting in a peak shaped distribution of the fγ(RT) against quench temperature. However, the quenching temperature for the maximum fγ(RT) is different from the temperature corresponding to the full stabilization of martensite against quenching to room temperature.
In the present study, 240, 253, 262, and 275°C were selected as the quenching and partitioning temperature for the D6AC steel. These partitioning temperatures are related to the fM(Q) from 7 to 45 vol.%. Figure 3(a) shows the XRD patterns of the As-Q and Q&P processed samples. While the As-Q sample shows only five bcc peaks ({110}α, {200}α, {211}α, {220}α, and {310}α), a small austenite peak of {220}γ was observed in the four Q&P processed samples. The amounts of retrained austenite for these samples were measured based on the peak intensities of five bcc peaks and one fcc peak and the results were compared to the predictions (Fig. 3(b)). Experimental fγ(RT) values up to about 7% are significantly lower compared to the predictions ranging from 0.12 to 0.43 for the same temperature range, and exhibit less dependence on the quench temperature. Reduced fγ(RT) values measured experimentally is similar to the results in.2,3,16,17,18)
(a) The X-ray diffraction (XRD) peak profiles for the D6AC steel of as-quenched (As-Q) sample and samples for quenching and partitioning (Q&P) for 300 s. The numbers in each peak profile indicate the temperature for Q&P process. (b) Retained austenite fractions for the Q&P processed samples. Predicted retained austenite fraction in Fig. 2 was also provided for comparison.
For the predictions, only martensitic phase transformation and full partitioning of carbon atoms from martensite to austenite were assumed. However, competing mechanisms such as the bainite or carbide formation can be activated during the actual Q&P process, and some of the carbon atoms can be segregated to dislocations and lath boundaries in the martensite microstructure.18) In particular, for the present study of D6AC steel, the large amount of carbide forming elements such as Cr, Mo and V may cause carbide precipitation or influence the diffusion kinetics of carbon atoms to austenite during the Q&P process. Although these phenomena generally reduce the amount of carbon atoms available to be partitioned to austenite compared to the predictions, the experimental results clarify part of the austenite in the D6AC steel was successfully stabilized by the Q&P process.
Figure 4(a) is an EBSD image quality map for the sample Q&P processed at 275°C, showing lath-type morphologies. Figure 4(b) is a corresponding phase map, where the brighter region denotes bcc martensite microstructure and darker region corresponds to the fcc austenite. Considering the heat treatment process, it is expected that the bcc region is mainly composed of the fresh martensite transformed after Q&P with a small fraction of martensite formed at the initial quench. In addition, the phase map confirms a small amount of austenite, about 6.2%, retained after the Q&P process, which is similar to the measured value of 6.4% by the XRD peak analysis. Most of the fcc austenite region shows a finely dispersed morphology, which is associated with martensite/austenite (M/A) constituents. As mentioned above, the austenite after the Q&P process is not fully stabilized, part of which may be transformed to martensite during secondary quenching to room temperature, providing the dispersion of the fine austenite grains in the bcc matrix microstructure.
Electron backscatter diffraction (EBSD) (a) image quality map and (b) phase map for the D6AC steel after quenching and partitioning (Q&P) at 275°C for 300 s. Brighter and darker regions denote the areas of bcc and fcc region, respectively.
Engineering stress-strain curves for the As-Q and Q&P processed samples of the D6AC steel and their tensile properties are shown in Figs. 5(a) and 5(b), respectively. All of the flow curves show continuous yielding with ductile fracture with a certain amount of post-uniform elongation. The As-Q sample exhibited the greatest yield strength (YS) of 1477 MPa and ultimate tensile strength (UTS) of 1841 MPa with limited uniform elongation (UE) of 6% and total elongation (TE) of 7%. The Q&P process decreases sample strengths and improves the ductility, which are greater with increasing quench temperature. Meanwhile, the minimum YS was obtained at 262°C and it increased at 275°C. Increased quench temperature generally softens the martensite formed at the initial quench, reducing the strength values. However, the amount of the martensite at initial quench is decreased at higher temperature and part of that was replaced by harder fresh martensite formed from partitioned austenite, resulting in the increase of YS at 275°C.
(a) Engineering stress-strain curves for the D6AC steel of as-quenched (As-Q) sample and quenching and partitioning (Q&P) of samples at 240, 253, 262, and 275°C for 300 s and (b) mechanical properties including yield strength (YS), tensile strength (UTS), uniform elongation (UE), and total elongation (TE).
Regardless of the Q&P processing temperature, the retained austenite of the tensile specimens was transformed to martensite during the tensile test, which was confirmed by XRD peak analysis. Considering the positive effect of martensite transformation during tensile deformation on the work hardening behavior, the (UTS – YS) value should be increased in the samples with larger fγ(RT). However, the relationship between the fγ(RT) and (UTS – YS) value is not clear in the present study. One possibility is that the work hardening behavior was governed by the deformation behavior of major microstructural component i.e., martensite transformed at initial quench and martensite transformed at final quench after partitioning.
Figure 6 compares the tensile properties (UTS and TE) for the D6AC steels of as quenched, tempered, spherodized, and Q&P processed samples.9,20) The as quenched specimens have the highest strength up to about 2000 MPa and limited ductility below 7%, and spherodized samples are most ductile with the TE above 20% and lowest strengths about 1000 MPa. Although the reference tempered samples and Q&P samples in the present study have similar level of TE values about 10%, the UTS values are relatively higher in the Q&P samples. It confirms that the mechanical properties of the D6AC steel can be improved by Q&P heat treatment process instead of general Q&T process.
The effects of quenching and partitioning (Q&P) heat treatment process on the retained austenite fraction and tensile properties were investigated in a D6AC steel. Hot-rolled samples were fully austenitized, quenched to different temperatures ranging from 240 to 275°C, partitioned at the quench temperature for 300 s, and quenched again to room temperature. The Q&P samples have retained austenite fractions up to about 6%, and the retained austenite was transformed to martensite during tensile deformation. The ultimate tensile strength and total elongation values for the Q&P samples are about 1500 MPa and 10%, respectively. These values are improved tensile properties compared to the conventional quenching and tempering heat treatment.
