Abstract
Hot stamping process has been regarded as one of the most attractive processes to produce high-strength parts with merits of low-forming load and small springback. However, the elongation of the hot-stamped parts is small, so the ability of crash resistance is limited. Recently, a novel hot stamping process integrated with quenching and partitioning treatment has been presented to improve the elongation of the final parts. In this article, the quenching and partitioning hot stamping process is further studied using the boron steel B1500HS, and the feasibility is verified by a series of quenching and partitioning tests followed by mechanical tests and microstructure observations. Moreover, an experimental tool for quenching and partitioning hot stamping process is first proposed in this article, where both air cooling device and heating system are designed, and a U-channel part is produced. Finally, in order to illustrate the active role of high elongation that the quenching and partitioning hot stamping process archived, numerical simulation of crash test for a B-pillar sample is conducted using finite element analysis software LS-DYNA.
Introduction
In the last decade, the hot stamping process has been developed quickly to produce high-strength steel automobile parts. In hot stamping process, the boron steel blank is heated up to about 900 °C and kept in the furnace for several minutes to ensure homogeneous austenitization and then the blank is quickly transferred to the press to conduct the forming and quenching processes simultaneously. The strength of boron steel after hot stamping process reaches 1500 MPa, but the elongation of final part is very limited to about 5%, 1 which brings inconvenience to the subsequent manufacturing processes and, what is more, decreases the ability of energy absorption during the car collision.
The product of tensile strength and plasticity (elongation) σb•At (σb represents tensile strength and At represents elongation), also called as formability index, 2 is widely regarded as a representative for the comprehensive performance of the mechanical part during the energy absorption. The conventional hot stamping product has a formability index less than 10 GPa%, which makes the hot stamping steel only belong to the first-generation auto sheet steel. 3 The second-generation steel such as twinning-induced plasticity (TWIP) steel has a very high formability index of about 50 GPa%, but it is expensive and difficult to process because many alloy elements are added. 4 Recently, the third generation of advanced high-strength steel has received additional consideration to produce auto parts, the formability index of which is lower than the second generation but much higher than the first one, and there is no much expensive alloy element, so the price is acceptable.
In order to increase the elongation of the product, several research works have been carried out by producing the microstructure-containing residual austenite, allotriomorphic ferrite or other phases with high ductility. Ying and Dong 3 proposed a microstructure modification idea featured by multiphase, meta-stable and multiscale (M3) to improve ductility of sheet steels. Naderi et al. 5 hot stamped four high-strength nonboron alloyed steels using water and nitrogen cooling media and found that hot stamping using water cooled punch resulted in maximum formability index due to the presence of some ferrite phase. Yi et al. 6 explored a novel steel design by adding high Al component, so that the quenching process results in a dual-phase allotriomorphic ferrite and martensite microstructure at the ambient temperature with better mechanical properties. Naderi et al. 7 developed the semi-hot stamping process based on the work of Schießl et al., 8 in which the MSW1200 blank was first heated to a temperature of about 650 °C instead of austenitizing temperature; the elongation after semi-hot stamping increases evidently, but the strength decreases to 650 MPa. Another method to improve the formability index is using tempering treatment after hot stamping, which has been proved by Ying et al., 9 Meng 10 and Naderi et al. 11 Liu et al.12,13 integrated the quenching and partitioning (Q&P) process into the conventional hot stamping to develop a so-called Q&P hot stamping process, and the test of high-strength steel (0.22 C-1.58 Mn-0.81 Si-0.022 Ti-0.0024 B) shows that the elongation of the final part increases evidently, and the high strength is still retained.
The Q&P hot stamping is further studied in this article, and the merits of this process are illustrated by a series of Q&P tests with boron steel B1500HS produced by BaoSteel Co. Moreover, an experimental tool for the new process is presented, and U-channel sample is produced. Finally, the importance of higher plasticity is illustrated by a typical B-pillar crash simulation through finite element analysis (FEA) software LS-DYNA.
Description of the Q&P hot stamping process
Matas and Hehemann 14 recognized that carbon can partition from martensite to austenite. Rao and Thomas 15 reported that the carbon content of residual austenite between martensitic laths could increase during the quenching process. Speer et al. 16 proposed a novel heat treatment called as Q&P process; they quenched the medium-carbon silicon steel (0.35 C-1.3 Mn-0.74 Si) to a certain temperature and then held it for a certain duration at a given temperature to cause carbon partitioning from martensite to retained austenite to result in austenite carbon enrichment and stabilization. The mechanism of austenite carbon enrichment during the partitioning process was discussed by Clarke et al. 17
In this work, the Q&P heat treatment is introduced to integrate with the hot stamping process. As illustrated in Figure 1, the medium-carbon blank is heated up and kept for several minutes to ensure homogeneous austenitization and then the blank is quickly transferred to the press to begin the forming and quenching processes. Remarkably, the blank is not quenched to the room temperature (RT) directly as the conventional hot stamping process does, but it is quenched quickly to a certain temperature called as quenching temperature (TQ), the value of which is between martensitic transformation start temperature (Ms) and martensitic transformation finish temperature (Mf). Then, the steel sheet is held for a certain duration named as partitioning time (tP) at a certain temperature called as partitioning temperature (TP). The purpose of the partitioning stage is to cause supersaturated carbon partitioning from martensite to the retained austenite, which results in carbon enrichment and austenite stabilization. The final part is composed of dual-phase microstructures of martensite and residual austenite, which induces effective improvement of elongation. TP can be equivalent to TQ, as shown in Figure 1(a), or be larger than TQ, as shown in Figure 1(b), the former process is called as one-step Q&P hot stamping process, and the latter is called as two-step Q&P hot stamping process. The conventional hot stamping process is shown in Figure 1(c), and the characteristic of the Q&P process can be easily found through the comparison between these figures. In this article, the one-step Q&P hot stamping process, as shown in Figure 1(a), is fully studied, and the conventional hot stamping process, as shown in Figure 1(c), is also introduced to compare with the one-step Q&P process. While the two-step Q&P hot stamping process, as shown in Figure 1(b), is not discussed here, the reason is mainly because the sheet temperature changes complicatedly and can be hardly realized with the presented experimental tools.

Schematic diagram of (a) one-step Q&P hot stamping process, (b) two-step Q&P hot stamping process and (c) conventional hot stamping process.
Representative Q&P tests of B1500HS
Material characteristics
The uncoated and cold-rolled steel B1500HS produced by BaoSteel Co. is studied in this article. Its chemical composition detected by Ail-3460 spectrometer is shown in Table 1. As delivered, the base material B1500HS has a ferritic–pearlitic microstructure with upper yield strength of 343.47 MPa, lower yield strength of 341.25 MPa and tensile strength of 600 MPa. Ms and Mf of B1500HS are 410 °C and 230 °C, respectively, with the critical cooling rate being 27 °C/s. 18 The 1.6-mm-thick sheet is cut into rectangle specimen with the length of 140 mm and width of 15 mm.
Chemical composition of Bl500HS steel made by BaoSteel Co. (wt%).
Experimental design of Q&P tests
In order to make the steel fully austenitized, the specimen is heated up to 920 °C with a rate of 10 °C/s and held for 5 min 19 using Gleeble 3500 system under vacuum condition. After that, the two different types of heat treatments based on conventional hot stamping and one-step Q&P hot stamping are carried out. The experimental parameters are given in Table 2, where specimens 1–3 are treated by full martensitic transformation which represents the conventional hot stamping process, and specimens 4–14 are treated by one-step Q&P process. All the specimens are cooled down to RT finally with the mixture of water and compressed air as quenchant, and the cooling rate remains constant during quenching stage. For each set of parameters, three tests are performed, and the average values are adopted for analysis.
Parameters of Q&P tests.
VQ: quenching cooling rate; TQ: quenching temperature; TP: partitioning temperature; tP: partitioning time; RT: room temperature.
Results and discussion of Q&P tests
Full martensitic transformation process
Figure 2 shows the experimental results after full martensitic transformation process, which in fact is the corresponding heat treatment during the conventional hot stamping process. It can be seen that cooling rate has notable influences on the mechanical properties. The tensile strength increases and the elongation decreases with raising cooling rate. While the product of strength and elongation does not change significantly overall, it reaches 17.1 GPa% when the cooling rate is set as 30 °C/s.

Effects of different cooling rates during full martensitic transformation process: (a) strength and elongation and (b) product of strength and elongation.
One-step Q&P process
Regarding the process chain, quenching to TQ instead of RT and partitioning for tP are the main differentials between the one-step Q&P hot stamping process and the conventional hot stamping process. In this section, the effects of quenching temperature TQ (equals to partitioning temperature TP for one-step Q&P process) and partitioning time tP are studied, respectively.
Figure 3 show effects of different partitioning time on mechanical properties, in which TQ is set as 250 °C. It can be seen that the strength reaches the maximum value when tP equals to 20 s and then decreases continually with tP increasing. It is suggested that tP of 10 s may be not enough to make adequate martensite transformation because of the uneven cooling condition for the thick blank, which induces low strength. For the case of 20 s, more martensite is transformed, and the strength is increased evidently. After that, the carbon in martensite starts to diffuse into the residual austenite, which induces the decreasing of the strength and increasing of the elongation. The product of strength and plasticity reaches its maximum value of 21.2 GPa% in this experiment when the partitioning time is chosen as 80 s, which is 24% larger than the conventional hot stamping value of 17.1 GPa%. It is also noted that if the partitioning time is set as 40 s, the product of strength and plasticity equals to 20.5 GPa%, which is close to the optimal value but only with a half partitioning time, and it is attractive since the cycle time is saved evidently.

Effects of different partitioning time during one-step Q&P process: (a) strength and elongation and (b) product of strength and plasticity.
Figure 4 illustrates the effects of different quenching temperature on mechanical properties of B1500HS during one-step treatment, in which cases of TQ equaling to 250 °C, 300 °C and 350 °C are studied, respectively. It can be seen that both strength and elongation reach the best values for TQ = 250 °C case, as well as the product of strength and plasticity. In addition, Figure 4 also illustrates that the laws of partitioning time influencing on mechanical properties mentioned in the above section are similar for all TQ cases.

Effects of different quenching temperatures and partitioning time during one-step Q&P process: (a) strength, (b) elongation and (c) product of strength and plasticity.
Figure 5 shows the microstructure of B1500HS after different heat treatments. It can be seen that the full martensitic transformation treatment generates a very fine lath martensite structure, while for the one-step Q&P process, some white blocks of austenite can be observed, and the martensite remains small and uniformly distributed. Since the quenching temperature of 250 °C is higher than Mf, martensitic transformation has not finished, and part of austenite exists at the end of quenching stage, and those austenite will be stabilized when the carbon partitions from martensite to austenite after the stage of partitioning, which explains why blocks of austenite can be observed after the one-step Q&P process. Hard martensite plays an important role in high strength, while the residual austenite, which has good deformation ability and transformation-induced plasticity (TRIP) effects, leads to a significant increase in plasticity. As a result, the material has better overall mechanical properties.

Microstructure of B1500HS after (a) full martensitic transformation process with a cooling rate of 30 °C/s and (b) one-step Q&P process with quenching temperature of 250 °C and partitioning time of 80 s.
As shown in Figure 6, austenite contents for different quenching temperatures and partitioning time are detected using special image analysis software (Image-Pro Plus 6.0). It extracts features by color with segmentation tools that isolate an area of interest (AOI) from the rest of the image. It should be noted that calculation errors may exist during the color extracting process, but the relative values under different process parameters are reliable.

Final austenite contents by one-step Q&P process: (a) tP = 20 s and TQ = 250 °C, 300 °C and 350 °C and (b) TQ = 250 °C and tP = 20 s, 40 s and 80 s.
The residual contents of austenite is highly related to the value of TQ; higher TQ can induce higher austenite content at the end of quenching stage and leads to less oversaturated carbon in martensite which, however, is important to steady the austenite at RT, so the relationship between austenite content and TQ is nonlinear, as shown in Figure 6(a), and the austenite content reaches the maximum value when TQ equals to 250 °C. Figure 6(b) shows the effect of different partitioning time on austenite content, in which the amount of residual austenite rises as the partitioning time increases since the carbon diffusion takes more time to proceed. And the behaviors for other TQ cases are similar.
Experimental tool design for one-step Q&P hot stamping and the U-channel part tests
Experimental tool design for one-step Q&P hot stamping process
Temperature control is one of the most important aspects for the tool design of Q&P hot stamping process. Three brief characteristics should be satisfied: first, the cooling rate during the quenching stage should be high enough; second, the quenching temperate should be controlled at TQ that is higher than RT; and, third, the part should be held for a duration of tP at temperate TQ during the partitioning stage.
In this article, a novel experimental tool for the Q&P hot stamping process is presented as illustrated in Figure 7. Instead of the water cooling system as in conventional hot stamping tools, a heating system is inserted in the die and punch to keep the tool at TQ, and the required high cooling rate during the quenching step is provided by a blast system using cooling air.

Schematic diagram of tool design for one-step Q&P hot stamping process.
Figure 8 gives the experimental tool for the Q&P hot stamping to produce U-channel part. In order to heat the tool, five copper plates with resistance coil are buried in the middle of punch, blank holder and both sides of die, respectively. The temperature of the tool can be monitored with thermocouples and controlled by a temperature controller. Meanwhile, a special air channel is designed in the die holder and a detachable metal plate with many holes installed at the bottom of the die, so that cooling air can pass through and bring down the temperature of the hot blank during the quenching stage. The cooling effect can be changed and controlled by flow rate, air temperature and the shape of detachable metal plate. The tool is heated up before hot stamping process and kept at a given partitioning temperature during the whole process. Cooling air is injected as the quenching step begins and switched off when the partitioning step starts.

Experimental tool for the one-step Q&P hot stamping process to produce U-channel part.
Hot stamping experiments of U-channel part
Design of hot stamping experiments
B1500HS blank with features of 260 mm length, 150 mm width and 1.6 mm thickness is heated up and austenitized at 920 °C for 5 min in an electric heating furnace. Then, the blank is quickly transferred to the experimental tool which has been pre-heated up and kept at the temperature of TQ. The stamping and quenching step begins with press pressure of 25 MPa, meanwhile the cooling air is injected into the tool to lower the blank temperature quickly. When the blank temperature arrives at TQ, the cooling air is stopped and the partitioning step begins. After tP of the partitioning time, the tool is opened and then the part is taken out and put into water to be quenched rapidly. In this experiment, TQ is set as 250 °C, and tp is chosen as 40 s. Conventional hot stamping process is also carried out to compare with the one-step Q&P hot stamping process; it should be noted that the designed tool shown in Figure 8 can also be used for conventional hot stamping process only if the heating function is turned off. The dimension of the final part is shown in Figure 9(a), where three typical sections named as top, bottom and middle are illustrated. The formed part from conventional hot stamping and Q&P hot stamping process are shown in Figure 9(b) and (c). In order to compare deformations, the formed parts are superimposed, and the 2D section profiles are shown in Figure 9(d). It can be seen that both of them have high forming accuracy with small springback.

U-channel part: (a) dimensional drawing, (b) produced part of conventional hot stamping, (c) produced part of one-step Q&P hot stamping and (d) 2D section profiles of the formed parts.
Results and discussions of hot stamping experiments
Tensile tests are performed at RT with tensile speed of 1 mm/min to obtain strength and elongation. The location, orientation and dimensions of extracted tensile test specimens in the U-channel part are shown in Figure 10, and stress–strain curves of the specimens from tensile tests are shown in Figure 11.

Extracted tensile test specimens in the U-channel part.

Stress–strain curve from tensile tests of the different sections: (a) bottom, (b) middle and (c) top.
Figure 12 shows the comparisons of mechanical properties for three positions between the conventional hot stamping and the one-step Q&P hot stamping process. It can be seen that the one-step Q&P hot stamping process produces better overall mechanical properties. For the bottom section, the product of strength and plasticity of one-step Q&P hot stamping reaches 14.6 GPa% with tensile strength of 1547 MPa and elongation of 9.4%, almost 28% larger than that of the conventional hot stamping with the product of strength and plasticity of 11.4 GPa%. For the middle section, the product of strength and plasticity of one-step Q&P hot stamping equals to 12.9 GPa%, which is close to that of conventional hot stamping with 12.8 GPa%. For the top section, the product of strength and plasticity of one-step Q&P hot stamping reaches 14.7 GPa% while that of conventional hot stamping is only 11.4 GPa%, the former is 1.3 times larger than the latter.

Mechanical property comparisons between conventional hot-stamped components and one-step Q&P hot-stamped components: (a) tensile strength, (b) elongation and (c) product of strength and plasticity.
By comparing with the results of Q&P tests mentioned above, the mechanical properties of the parts produced by the experimental tool are a little lower. The main reason is that the contact conditions of different sections vary greatly during the forming procedure, which induces heat transfer inhomogeneous and generally not so ideal as in the Gleeble system. In addition, it is not easy to control the blowing air to cool down the hot blank uniformly during the quenching stage, which results in uneven microstructure evolution. More efficient tool and process designs will be developed in the further work. Meanwhile, because of the nonideal cooling and contact conditions, the mechanical properties of the part are inhomogeneous, as shown in Figure 12. The bottom sections of the part contact with the tool surface all the time, which results in rapid cooling and high tensile strength. While the contact condition between the middle sections and tool surface is relied on the die clearance, the heat produced by plastic deformation also plays an important role since these sections are the main deformation areas of the part. The top section has the worst cooling condition because it is single-edge contact with tool in most of the stamping time, which leads to tensile strength of 1130 MPa. If the cooling rate cannot reach the critical value during the quenching stage, part of austenite transforms to bainite instead of martensite, which induces decline in strength and increase in elongation.
Numerical simulation of B-pillar collision tests
The main advantage of the discussed Q&P hot stamping process is to increase the product of strength and plasticity of final parts, which will be finally reflected in the behavior of car crash. In this section, numerical simulations of B-pillar collision tests are presented by the commercial FEA software LS-DYNA, where the material properties of the B-pillar are chosen as the conventional hot stamping product and the one-step Q&P hot stamping product, respectively, as mentioned above.
Finite element model
The geometric model of B-pillar is illustrated in Figure 13(a), and the thickness of the pillar is 1.6 mm. Both ends of the pillar are constrained in all degrees of freedom as the boundary condition. Four points (A, B, C and D) are chosen to investigate the intrusion of different positions of B-pillar. According to the auto crash testing standards C-NACP, a 50-kg cylindrical punch with initial velocity 13.8 m/s is adopted to affect the B-pillar, as shown in Figure 13(b), where the finite element model (FEM) mesh with 23916 Belytschko–Tsay shell elements for B-pillar and 6458 rigid shell elements for punch is also presented. The element size about 5 mm is believed to be sufficient for this problem after a series of test simulations, so the general element size is set as 3–6 mm except in the corner of B-pillar where the element size is about 1 mm in order to catch the geometric features.

Finite element model of B-pillar: (a) geometric model and (b) mesh model.
Two different types of material properties are chosen to study in this collision test: one is from the conventional hot stamping process, and the other one is from the one-step Q&P hot stamping process in which TQ is set as 250 °C and tp is 40 s. The flow curve of these two materials is shown in Figure 14, and some key material parameters are given in Table 3. It should be noted that the main purpose of the B-pillar collision test in this article is to compare the effects of different elongations, so the material properties of the pillar is set as uniform. The simple maximum plastic strain failure criterion is applied in the simulation, that is, if the equivalent plastic strain reaches the material elongation value, the element will be regarded as failed.

Flow curves of B1500HS after one-step Q&P hot stamping process and conventional hot stamping process.
Material parameters of B-pillar.
Results and discussions for FE simulations
Figure 15 presents the comparison of intrusions at four representative positions. It can be found that the B-pillar with material properties of one-step Q&P hot stamping process has smaller intrusions than the one of conventional hot stamping process at all four positions, that is, the former one provides better impact resistance.

Intrusion contrasts: (a) position A, (b) position B, (c) position C and (d) position D.
It can also be illustrated in the displacement profile of B-pillar at a typical time 0.03 s, as shown in Figure 16, in which the B-pillar of conventional hot stamping has been fractured near the position A, and the maximum intrusion is 100.28 mm. Contrarily, no fracture has occurred on the B-pillar of one-step Q&P hot stamping process, and the maximum intrusion is only 95.17 mm.

Displacement profiles at 0.03 s of B-pillars produced by (a) conventional hot stamping and (b) one-step Q&P hot stamping.
The numerical simulation results show that due to higher product of tensile strength and plasticity, the B-pillar produced by one-step Q&P hot stamping process has better comprehensive mechanical properties with less intrusion and good energy absorption than the one by conventional hot stamping process.
Conclusion
In order to overcome the low elongation property of the conventional hot stamping process, the Q&P hot stamping process is further studied, a series of Q&P tests, hot stamping experiments of U-channel part and impact simulation of B-pillar are carried out successively. The results demonstrate the following:
As a result of dual-phase microstructures of martensite and residual austenite, B1500HS treated by the one-step Q&P hot stamping process obtains a higher product of tensile strength and plasticity than that of the conventional hot stamping process where full martensitic transformation occurs.
Both quenching temperature and partitioning time are important to affect final mechanical properties. When quenching temperature is set as 250 °C and partitioning time is chosen as 80 s, the presented one-step Q&P hot stamping process with material B1500HS obtains the maximum product of strength and plasticity of 21.2 GPa%, which is 24% larger than that of the conventional hot stamping process.
An experimental tool for Q&P hot stamping process is first presented in this article, where both air cooling device and heating system are designed to provide high cooling rate during the quenching stage and to maintain the partitioning temperature during the partitioning stage.
Through crash simulation, it can be seen that the B-pillar produced by the one-step Q&P hot stamping process has better mechanical properties with good energy absorption and smaller incursion distance than the one produced by conventional hot stamping process.
Footnotes
Appendix 1
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China through contract or grant number 51105247.
