An investigation of deformation effects on phase transformation in hot stamping processes.
To reduce the fuel consumption as well as to improve the crash safety of vehicles, the usage of hot stamping parts is increasing dramatically in recent years. Aisin Takaoka has produced hot stamping parts since 2001 and has developed various technologies related to Hot Stamping. In an actual hot stamping process, parts with insufficient strength could be produced sometimes at a prototyping phase, even under the proper forming conditions. In order to understand these phenomena, in this paper, phase transformation in a boron steel 22MnB5 under various cooling rates were investigated and the effects of pre-strain conditions on the phase transformations were characterised. Uniaxial tensile specimens were stretched under isothermal conditions to different strain levels of 0-0.3, at strain rates of 0.1-5.0/s and deformation temperatures of 650-800[degrees]C. A dilatometer was used to measure the dimensional changes of the specimen to rationalize the phase transformation of the boron steel during rapid cooling after deformation. The results showed that the deformation at austenite phase causes the reduction in martensitic phase transformation. For a certain strain level, deformation applied at a lower temperature and a higher strain rate would result in less amount of martensitic phase transformation. In addition, the relations between hot stamping conditions and Vickers hardness, which are normally used as a daily quality check in the industry, are also presented in the paper.
CITATION: Matsumoto, T., Li, N., Shi, X., and Lin, J., "An Investigation of Deformation Effects on Phase Transformation in Hot Stamping Processes," SAE Int. J. Mater. Manf. 9(2):2016,
Lightweight and high crash resistance are more and more important for the automotive industry. This has increased the usage of hot stamped parts rapidly over the last 5-10 years and will increase more in the future. In a hot stamping process, a fast cooling rate of the steel during forming and quenching in the cold dies is critical to obtain the ultrahigh strength and geometrical accuracy of the formed part. Continuous cooling transformation (CCT) diagram is normally used to determine the cooling rate for achieving full martensite phase in the formed parts. However the CCT diagram obtained from the full annealed material is not applicable for the hot stamping applications, since the phase transformation would be affected by the deformation of the material prior to quenching.
Studies on phase transformations during non-isothermal deformations have been conducted  . Under the condition of hot non-isothermal deformation with a cooling rate of 50[degrees]C/s, a higher deformation temperature can reduce the effect of plastic strain on martensite phase transformation. Strain dependant CCT diagrams for FE analysis were also investigated by other researchers  , which enhanced the understanding of the effect of plastic deformation on phase transformation during continuous cooling. However, in a real hot stamping process, the deformation temperature, strain level, and strain rate are not constant and the effects of strain at different deformation conditions on the phase transformation still have not been quantified. A process window needs to be created to ensure that full martensite in formed parts could be achieved.
In this paper, uniaxial tensile and heat treatment tests have been performed under different conditions which simulate the actual hot stamping mass production processes. It aims to investigate and analyse the relations of deformation conditions in terms of strain level, deformation temperatures and strain rates with cooling rates, and martensite phase transformation in terms of transformation temperatures and post hardness.
Material and Equipment
The material used in this study was an uncoated 22MnB5. The dimensions of the tensile specimens are shown in Fig 1. All specimen used for the experiment were machined from the same batch of as-delivered, 1.60 mm thick, boron steel sheet. The specimens were cut by using laser and cut-edges were polished. Gleeble 3800 which is shown in Fig 2 was used for the isothermal tensile and heat treatment tests. A pair of K-type thermocouples was attached onto the centre of the specimen, which was used to monitor and control the actual temperature of the specimen throughout the tests. C-shaped contact dilatometer was used to measure the dimensional changes of the steel by measuring the width of the middle of the specimen, so as to detect the phase transformation. Air cooling nozzles were employed to realise high cooling rates of specimens.
Figure 3 shows the temperature-time profile for isothermal tensile testing and quenching of the specimen. An example of deformation at the temperature of 800[degrees]C is presented. The test specimen was heated up to 730[degrees]C with the heating rate of 8[degrees]C/s and then up to 900[degrees]C with 3[degrees]C/s. This was designed according to the heating condition of atmosphere furnace which is used in actual production processes. According to material supplier, the Ac3 temperature was about 820[degrees]C under the designed heating condition, thus the austenization in the boron steel was completed after heating. After that, there were two steps for cooling. The first step took 6 seconds which is the total time of cooling of a specimen from 900[degrees]C to the temperature for tensile testing and deformation of the tensile specimen. This cooling and tensile testing condition was also designed from actual production process that transferring a heated work piece from furnace to forming tools and forming the work piece. The second step was 10 seconds for quenching which is the time of work piece being quenched in forming tools. In addition, different deformation temperatures (650, 700, 750, and 800[degrees]C), strain rates (0.1, 1.0, and 5.0/s), strain levels (0, 0.1, 0.2, and 0.3), and cooling rate (25 - 80[degrees]C/s) were designed as test conditions.
RESULTS AND DISCUSSIONS
Measurement of the Amount of Martensite Phase Transformation
Fig.4 shows an example of strain changes of a boron steel specimen during quenching measured by a dilatometer. The thermal strain was calculated according to dilatometer reading. Tb represents the starting temperature of transformation from austenite to bainite Tms and Tmf represent the starting temperature and finish temperature of martensite phase transformation, respectively. The difference between the minimum and maximum values of strain during martensite transformation, expressed by [DELTA][epsilon] in Fig. 4, was used as a criterion to evaluate the amount of martensite phase transformation. A higher value of [DELTA][epsilon] indicates a higher volume fraction of martensite obtained.
Effect of Strain Level on Martensite Phase Transformation
Fig. 5 shows the amount of strain changes during martensite phase transformation of boron steel specimens with different pre-strain levels deformed at the temperature of 700[degrees]C and 800[degrees]C, strain rate of 1/s, and quenched at different cooling rates. The results indicate that a higher strain level caused less amount of martensite phase transformation. This was because higher dislocation density caused by a higher strain level induced more bainait or ferrite phase transformation, so as to reduce the martensite phase transformation. With increasing cooling rate, the amount of martensite phase transformation increased at all conditions. However, a cooling rate of 80[degrees]C/s was not high enough to obtain full martensite phase when the strain level was greater than 0.2 under the tested deformation temperatures and strain rates. It is not easy to control the strain level of actual production parts by only production conditions. Therefore, in order to get full martensite phase, geometry of the parts should be carefully designed and it is better to check the strain level by using FE analysis beforehand.
Effect of Strain Level on Vickers Hardness
Fig. 6 shows the Vickers hardness of specimens quenched at different cooling rates, when they were pre-deformed at the different strain levels, at the temperature of 700[degrees]C and 800[degrees] and strain rate of 1/s. The hardness values of boron steel without pre-deformation quenched at 40[degrees]C/s cooling rate were 441-450Hv. These values indicated the hardness of full martensite. As shown in Fig.6, a higher strain level led to a lower hardness value at the same working and cooling conditions. For a higher strain level, it required a higher cooling rate to obtain the same hardness level. When the strain level was greater than 0.2 and strain applied at 700[degrees]C, because of the ferrite and bainite phase transformation induced by strain had occurred before martensite phase transformation, full martensite transformation condition could not be obtained. Therefore, the hardness level was lowered compared to the hardness of full martensite.
Fig.7 shows the effect of strain level on phase transformation temperatures (deformation condition: 800[degrees]Cstain rate: 1/s, cooling rate: 80, 55, 45[degrees]C/s). Corresponding starting temperature of ferrite, bainite, and martensite phase transformation and finish temperature of martensite phase transformation are marked as Tf, Tb, Tms and Tmf. The results show that ferrite, bainite, and martensite phase transformation occurred when the strain level is 0.3 and only bainite and martensite phase transformation occurred when the strain level is 0.1. Bainite phase transformation occurred at a higher temperature at strain level of 0.3, which was because higher strain level caused higher dislocation density and it enhanced bainite phase transformation. Ferrite phase transformation can be seen at strain level 0.3 was the same reason.
Effects of Deformation Temperature on Martensite Phase Transformation and Vickers Hardness
Fig. 8 shows the deformation temperature and cooling rate effects on the amount of martensite phase transformation and resulting Vickers hardness. It can be seen that a higher deformation temperature allowed more martensite phase transformation and higher Vickers hardness. From the hot stamping mass production point of view, efforts on minimizing the transfer time of heated blanks from furnace to forming tools and forming parts at a higher starting temperature could enhance martensite phase transformation so as to enable higher Vickers hardness of as-formed parts, even at a higher strain level. Applying strain causes increasing dislocation density; however, these dislocations are easier to recovery at a higher temperature.
Fig. 9 shows the effect of deformation temperature on phase transformation temperatures. It is can be seen that the deformation at a lower temperature lowered the temperature of bainite phase transformation. Ferrite phase transformation was not observed from the dilatometer measurement; however it can be to occur during the deformation at 650[degrees]C.
Effects of Strain Rate on Martensite Phase Transformation
Fig.10 shows the effect of strain rate of pre-deformation on the martensite phase transformation in boron steel and the resulting post Vickers hardness. With increasing strain rate from 0.1/s to 5.0/s, the amount of martensite phase transformation was significantly reduced.
This is because a higher strain rate caused a higher dislocation density which induced ferrite and bainite phase transformation, and as a result the amount of martensite phase transformation was reduced.
In the hot stamping mass production processes, to obtain the maximum hardness is one of the highest priorities for forming the parts. It depends on the martensite phase transformation in boron steels during quenching. In this work, the effects of deformation temperature, strain level, and strain rate on the phase transformation in a boron steel 22MnB5 during quenching have been studied. The conclusions have been drawn are summarized as:
1. A higher strain level can cause fewer amount of martensite phase transformation and a lower Vickers hardness level. The cooling rate of 80[degrees]C/s was not sufficient to enable full martensite phase transformation in the boron steel with a pre-strain level of 0.3 deformed at 800[degrees]C with a strain rate of 1/s. Therefore, when the parts are designed, it is necessary to check the strain level of the parts using FEM.
2. For a certain strain level, a higher deformation temperature can allow more amount of martensite phase transformation and a higher Vickers hardness level. This is because a higher deformation temperature provided a chance to reduce the dislocation density, which reduced the strain induced phase transformation of ferrite and bainite. As a result, more martensite phase transformation occurred. Therefore, in the actual mass production, a higher starting temperature for forming is important.
3. For a certain strain level, the strain rate during deformation also affected the amount of martensite phase transformation. A Higher strain rate can cause more strain induced ferrite and bainite phase transformation due to less time for the recovery of dislocations. As a result, less amount of martensite was obtained. The Vickers hardness dropped sharply with increasing strain rate from 0.1/s to 5.0/s.
[1.] Naderi M.. et al., "An investigation into the phase transformations during non-isothermal deformation of 22MnB5 boron steel", Materials Science and Engineering : A, volume 487, Issues 1-2 (2008)
[2.] Nikravesh M. et al., "Phase transformations in a simulated hot stamping process of the boron bearing steel", Materials and Design 84, 18-24 (2015)
[3.] Behrens B. et al., "Numerical and Experimental Analysis of the Phase Transformation during a Hot Stamping Process in Consideration of Strain dependent CCT-Diagrams" 4th International Conference on Hot Sheet Metal Forming of High-Performance steel, (2013) pp. 329-336
[4.] Behrens B. et al., "Numerical Simulation of Phase Transformation during the Hot Stamping Process", 5th International Conference on Thermal Process Modelling and Computer Simulation, (2014)
[5.] LiF.F. et al., "Effect of cooling path on the phase transformation of boron steel 22MnB5 in hot stamping process" The international Journal of Advanced Manufacturing Technology (2015)
[6.] CaiY. et al., "Simulation and experiment research of phase transformation on the hot stamping of 22MnB5 high strength steel", Material Research Innovations Vol.18 (2014)
The studies were done in cooperation between Imperial College London and Aisin Takaoka. Many thanks to the colleagues from both companies for their excellent support.
Takeki Matsumoto Aisin Takaoka Nan Li, Xin Shi, and Jianguo Lin Imperial College London
|Printer friendly Cite/link Email Feedback|
|Author:||Matsumoto, Takeki; Takaoka, Aisin; Li, Nan; Shi, Xin; Lin, Jianguo|
|Publication:||SAE International Journal of Materials and Manufacturing|
|Date:||May 1, 2016|
|Previous Article:||Hydrogen solubility effects in galvanized advanced high strength steels.|
|Next Article:||A stress-based non-proportionality parameter for considering the resistance of slip systems of shear failure mode materials.|