Accepted Manuscript

Title: Effect of contact pressure on IHTC and the formability of hot-formed

Author: Ying Chang, Shujuan Li, Xiaodong Li, Cunyu Wang, Ping Hu,
Kunmin Zhao

PII: S1359-4311(16)30003-5
DOI: http://dx.doi.org/doi: 10.1016/j.applthermaleng.2016.01.053
Reference: ATE 7613

To appear in: Applied Thermal Engineering

Received date: 2-7-2015

Please cite this article as: Ying Chang, Shujuan Li, Xiaodong Li, Cunyu Wang, Ping Hu,
Kunmin Zhao, Effect of contact pressure on IHTC and the formability of hot-formed 22MnB5
automotive parts, Applied Thermal Engineering (2016), http://dx.doi.org/doi:
10.1016/j.applthermaleng.2016.01.053.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will
undergo copyediting, typesetting, and review of the resulting proof before it is published in its
final form. Please note that during the production process errors may be discovered which could
affect the content, and all legal disclaimers that apply to the journal pertain.
Effect of contact pressure on IHTC and the formability of
hot-formed 22MnB5 automotive parts
Ying Chang a, Shujuan Li a, Xiaodong Li a,*, Cunyu Wang b,*, Ping Hu a, Kunmin Zhao a

a
Dalian University of Technology, School of Automotive Engineering, State Key Lab of Structural

Analysis for Industrial Equipment, Linggong Road, Dalian 116024, China

b
East China Branch of Central Iron & Steel Research Institute(CISRI), Beijing 100081, China

*
Corresponding author. E-mail: lixiaodong@dlut.edu.cn; wangcunyu@ec-cisri.com.cn

Postal Address: No.2 Linggong Road, Dalian, 116024, China

Highlights

 Crack characteristics of hot-formed automotive parts are revealed.

 Relation of contact pressure-temperature-phase transformation-crack is analyzed.

 Interfacial heat transfer rules following contact pressure are investigated.

 IHTC is solved by experiments and numerical method during hot forming.

 Effective local oil-coating method is applied to improve the formability.

Abstract: The interfacial heat transfer coefficient (IHTC) between the steel blank and die directly

affects temperature distribution during hot forming of advanced high-strength steel automobile

parts and further affects martensite microstructure distribution and formability. In this paper, an

experimental platform is established and Beck’s nonlinear inverse estimation method is employed

to solve the IHTC and investigate the rules of IHTC following contact pressure. This paper

demonstrates that the IHTC increases with the increase in contact pressure. An actual hot-formed

Page 1 of 21
automotive B-pillar was obtained in the experiment. Results show that a high contact pressure

corresponds to a high IHTC, which promotes fast blank cooling on the fillet corner area. By

contrast, a low contact pressure results in low IHTC and cooling rate on the side wall area. Thus, a

significant temperature difference induces asynchronous martensite transformation between the

fillet corners and the side walls, where crack is more likely to occur. As a solution, local oil

coating is used on fillet corners with high contact pressure before hot forming to improve the

formability and avoid cracks. This finding indicates that local oil coating can effectively weaken

the heat transfer performance, balance the temperature field in the different areas of the part, and

consequently achieve better formability for the hot forming process.

Keywords: 22MnB5 boron steel; Hot forming; Interfacial heat transfer coefficient (IHTC); Contact

pressure; Formability

1 Introduction

Hot stamping is a non-isothermal high temperature forming process, in which complex

advanced strength parts are produced, with the goal of no spring back, thinner and high

geometrical accuracy [1]. During the process, the advanced high-strength steel blanks are heated

to over 900℃ for completely uniform austenizing and subsequently formed and quenched in the

die[2-4]. At the end of the cycle, the part has a full martensitic microstructure with tensile strength

and yield strength exceeding 1500MPa and 1000MPa, respectively [5-8]. Hot-stamped automobile

structural components can effectively improve a vehicle’s safety, crashworthiness, and

dimensional quality, as well as reduce its weight [9, 10]. However, an effective method for

improving the formability of parts has long been needed to produce complex structural

Page 2 of 21
components, such as A-pillar, B-pillar, and side impact beams [11].

Thermal field is more significant for the formability of parts during the hot forming process

than during the conventional cold-forming process. Interfacial heat transfer coefficient (IHTC), as

an important thermophysical parameter, reflects the heat transfer ability between blank and dies,

affects the cooling rate, and consequently determines the temperature field distribution in the parts.

Thus, the final formation of parts is directly influenced by the IHTC. Hence, the IHTC solution

and analysis during hot forming have become efficient means of enhancing the formability of

parts.

In heat transfer theory, the IHTC is considered as a constant value under an ideal condition.

However, the ideal condition is difficult to achieve in practice and the IHTC changes in value as

the part is stamped and quenched. Several factors have been considered to affect IHTC, such as

surface roughness, thermophysical properties of the material, thermal contact resistance, and oxide

scale. Ikeuchi and Yanagimoto [12] demonstrated that both the surface roughness and

thermophysical properties of the material would affect the IHTC. Abdulhay et al. [13] obtained

contact pressure and thermal contact resistance at different locations during hot stamping a

U-shaped part by experiments, and further proposed the approximate functional relation between

thermal contact resistance and contact pressure. Blaise et al. [14] explained the variation of

thermal contact resistance of Usibor 1500P steel and the corresponding effects on IHTC during hot

forming. Slowik et al. [15] and Wendelstorf et al. [16] considered that the thermal conductivity of

oxide layer was much less than those of the blank/die and the thicker oxide layer could not be

ignored, and analyzed the effects of oxide layer on the contact heat transfer. Hu et al. [17] studied

the influence of oxide scale on IHTC for 22MnB5 steel. However, insufficient research

Page 3 of 21
information is available about the effect of contact pressure on IHTC and formability,

investigation of crack reasons, and further exploration of solutions for avoiding cracks. Therefore,

how the contact pressure influences IHTC rules remains unclear. In the present study, we aimed at

the actual automotive part, focused on the influence of contact pressure on the IHTC of

hot-formed 22MnB5 parts to reveal the reasons for cracks based on experiments and numerical

analysis. Furthermore, an efficient solution is proposed for avoiding cracks and improving

formability.

2 Experiments and methods

2.1 Experiments

2.1.1 Hot forming process and cracking phenomenon description

A hot-stamped advanced high strength steel B-pillar is utilized as the object in this study. The

B-pillar is produced by the direct hot forming process, as shown in Figure 1. The 22MnB5

pre-cut-blank is heated in a furnace to 900 °C at a heating rate faster than 10 °C/s, kept for 3–5

minutes to achieve full austenitization, and then transferred to a high-speed hydraulic press. It is

subsequently formed in a die with internal water-cooling channels and finally quenched under a

certain holding pressure.

However, the cracking phenomenon occurs after hot forming the B-pillar. Figure 2 shows the

left view and right view of the hot-formed B-pillar. Cracks occur mainly in the border between the

side walls and upper/lower surface. Cracks start between the side walls and upper surface, and

then extend to the border between the side walls and lower surface. Cracks are significant and

obvious, and they harm the overall formability of the part.

Page 4 of 21
2.1.2 Heat-transfer experiment setup

A self-design experimental platform is used for heat transfer test. The experiment setup

consists of a small tonnage hydraulic press, a heat treatment furnace, a two-piece cylindrical die,

four thermocouples, a data acquisition device MX100, and a computer, as presented in Figure 3.

The tested specimen is a circular 22MnB5 blank with 1.80 mm thickness and 75 mm diameter.

Temperature data are collected by four thermocouples, three of which are installed at 2, 4, and 6

mm below the surface inside the lower die and one at the center of the specimen.

The specimen is treated by the direct hot forming process. The collected temperatures are

passed to the computer console. Different contact pressures can be conducted within the range of

0–20 MPa. Table 1 shows the contact pressures of six levels within the scale of the hydraulic press

to investigate the effect of contact pressures on IHTC.

2.2 Methods

2.2.1 Modelling and simplifying of heat transfer procedure

The unsteady heat conductive differential equation in the three-dimensional Cartesian

coordinates inside the die is generally expressed as below Eq. (1).

T   T    T    T 
c       Q
(1)
t x  x  y  y  z  z 

where ρ is the density, c is the specific heat capacity, λ is the heat conductivity coefficient,

x , y , z present the directions of thermal conduction, Q is the heat from inner heat source, T is the

temperature of the thermal conductor, t is the time of heat transfer.

Generally, heat transfer can be considered one-dimensional when the conductive heat

Page 5 of 21
resistance inside the solid is much less than the contact interfacial thermal resistance [18–20]. In

order to ensure the reliability, a one-quarter model of the tested cylinder is built to simulate the

temperature field for confirming the one-dimensional heat transfer in this paper. The temperature

field distributions of both the blank and dies during quenching process are simulated by the

Deform 3D forming software as Figure 4 and Figure 5.

Figure 4(a) shows the temperature difference between the inner center (P1) and the surface

center (P2) of the specimen in the entire quenching procedure. The temperatures at P1 and P2 are

very close because hot-formed 22MnB5 steel has good hardenability and a high heat conductivity

coefficient. The maximum temperature difference is only 3.4%, which occurs at the 1st second of

quenching. Figure 4(b) shows the temperature field of the one-quarter specimen model at the 1st

second of quenching. In addition, the temperature distribution is nearly even within the 10 mm

range of the center. Thus, the specimen center temperature is hardly influenced by the surrounding

areas. These results confirm that the experiment can be considered a one-dimensional heat transfer

problem, i.e., heat transfers only along the normal direction of the blank surface. The temperature

of the inner center can be used similarly to calculate the IHTC for the specimen instead of the

surface center, as expressed in Eq. (2).

CAL M EA
Tt , B  Tt , B (2)

M EA CAL
where, T t , B is the inner-center temperature of measured blank at time t, T t , B means the

surface-center temperature of blank to be used in IHTC calculation at time t.

Figure 5(a) presents the temperature field distribution on the die surface, from the center

along the radius at the 10th second of quenching. Obviously, the surface temperature is uniformly

distributed within 10 mm of the die surface center during quenching. Furthermore, Figure 5(b)

Page 6 of 21
shows the temperature field of the one-quarter specimen model at the 10th second of quenching.

The inner temperature of the specimen is also uniformly distributed within 10 mm around the

center axis. Thus, the temperature of the die center axis is hardly influenced by the surrounding

areas along the radius, which satisfies the one-dimensional heat transfer characteristic [21].

Furthermore, Figure 6(a)–(d) demonstrates the temperature distributions from the center axis

along the radius on different cross sections of the blank and die at the 2nd, 4th, 6th, and 8th seconds

of quenching to validate the one-dimensional heat transfer characteristic. The curves within the 10

mm radius tend to the horizontal line, which indicates that the temperature is distributed evenly

and one-dimensional heat transfer could be assumed.

In summary, the one-dimensional unsteady thermal conduction problem without internal heat

source can be assumed in this paper. Temperature is calculated by solving the partial differential

equation, as presented in Eq. (3), which is simplified from Eq. (1)

T   T
2

 (3)
t  c x
2

For the unsteady thermal conduction problem, the conditions for deterministic solution

consist of two factors: one is the initial condition for providing the temperature distribution of the

initial time, and the other is the second type of boundary condition, known as Neumann condition,

for providing the heat flux on the boundary of the subject.

2.2.2 Beck’s nonlinear inverse estimation method

The IHTC can be calculated by the Newton’s cooling law.

q
h  (4)
TB  TD

Page 7 of 21
where q is the heat flux density between the blank and the die, TB is the surface temperature of the

blank and TD is the surface temperature of the die. Therefore, it is necessary to obtain the

temperatures of the blank and the die, as well as the heat flux density, in order to calculate the

ITHC.

Beck’s nonlinear estimation is an efficient method to solve the inverse heat conduction

M EA
problem [22]. The theoretical model is applied (at any time j), as shown in Figure 7. Here, T j , D x

CAL
means the measured temperature in different positions of die. T j , D x
represents the die

temperatures at the different positions solved by partial differential equation. q is the heat flux

density between the blank and the die. h means the solved IHTC.

The Beck’s nonlinear estimation method is carried out by adjusting the boundary condition,

i.e. the heat flux q, versus each time segment to minimize the difference between the calculated

and measured die temperatures during the whole quenching period. The optimized object function

at a time j is expressed by Eq. (5).

 T 
2
 m in
CAL M EA
d j j , Di
- T j,D (5)
i
i1

where N is the total number of measured points below the die surface, and i = 1, 2, 3 represents the

measured locations at 2, 4, and 6 mm below the die surface, respectively.

Adjusting the heat flux q at time j is described by Eq. (6)

 
N
 T M EA - T CAL  
 i1  j , Di j ,Di j

q j  N (6)
 
2
i1 j

Where  j is the sensitivity coefficient of heat flux and defined by Eq. (7).

 q  1  e    T
CAL CAL
T T (q )
   (7)
q qe

Page 8 of 21
Where e is a small number assumed as 1.0E – 4.

The function Pdepe in Matlab is called to solve Eq. (3) and obtain the temperature field of the

die over time. The iteration procedure yields the optimal heat flux qj and die surface temperature

CAL
T j,D
0
. Furthermore, the IHTC for any time j can be calculated by using Eq. (8) combined with Eq.

(2). Finally, the IHTC h for the entire heat transfer process can be obtained.

qj
hj  (8)
 T j ,D 0
M EA CAL
T j ,B

3 Results and discussion

3.1 Analysis of cracks

3.1.1 Non-uniform temperature field distribution during hot forming the B-pillar

The Finite Element Method (FEM) simulation diagram for temperature field after 3 s of

quenching is shown in Figure 8. The diagram illustrates that the B-pillar temperature field is

non-uniform. The cooling rate is lower on the side walls, which results in higher temperatures in

areas 3 and 4. However, the cooling rate is higher at the fillet corner between the transitional and

upper/lower surfaces, which leads to lower temperatures in areas 1 and 2.

The part is cooled gradually through heat transfer with dies. Martensite transformation will

occur once the temperature of the part reaches the martensite transformation start (Ms) temperature

during quenching. Koistinen and Marburger [23] presented the following relationship between

martensite fraction and part temperature during nonisothermal martensite transformation in 1959.

ξ=1-exp[-θ(Ms-T)] (9)

where ξ is the martensite fraction, Ms is the martensite transformation start temperature, θ is a

constant of martensite transforming rate, and T is a value of quenching temperature below the Ms

Page 9 of 21
temperature.

Eq.(9) have been embedded into hot forming simulation software programs, such as the

famous LS-DYNA, which can be used to calculate the martensite fraction in any automobile part.

It is clear in Eq.(9) that the cooling rate and real-time part temperature directly affect the

martensite fraction. Furthermore, the real-time part temperature depends on the cooling rate, that

is, the cooling rate is the main factor for martensite fraction. A greater cooling rate difference will

lead to a greater martensite fraction difference in the different areas of the part at any moment

during hot forming. The non-uniformity of martensite transformation causes the difference in the

mechanical properties between adjacent areas, which is one of the main causes of cracks.

Figure 8 shows that the temperature difference during forming exceeds 100 °C. Area 1, which

has a higher cooling rate, can reach the Ms temperature earlier, thereby contributing to the early

occurrence of martensite transformation. However, the martensite transformation in area 3 with

lower cooling rate will be delayed relative to that of area 1. Similarly, martensite transformation

occurs in area 2 earlier than that in area 4. Thus, a greater temperature difference results in

martensite transformation non-uniformity of the parts and probably leads to cracking. Cracks are

prone to appear between areas 1 and 3 or between areas 2 and 4, as presented in Figure 8. The

conclusion is consistent with the cracks in experiments.

3.1.2 Effect of contact pressure on temperature field distribution

The B-pillar is a typical automobile part with a complex 3D shape. The contact pressure that

spreads all over the B-pillar surface has been proven less uniform [24]. Through the FEM

simulation method, the distribution of contact pressure on the part’s surface at the 10th and 12th

10

Page 10 of 21
steps during hot forming is presented in Figure 9(a) and 9(b). It shows that the contact pressure

distribution is non-uniform and varies with time. The contact pressures are smaller on the side

walls of the B-pillar because the side walls hardly touch the die surface given the existence of die

clearance. However, the contact pressures at the fillet corners between the transitional surface and

the upper/lower surface are obviously much higher during the entire forming.

During forming and quenching, heat transfer is completed in three ways, namely, conduction,

convention, and radiation. Conduction relies on the effective contact of part and die to realize heat

transfer. Its heat energy transfer is higher than that of the other two ways and is the main way. In

the experiment, different contact pressures change the contact extent between the blank and die

and influence the heat transfer, and IHTC varies with the contact pressure. As a result, varied

cooling rates occur at the different B-pillar areas [Figure 8] following the contact pressure

distribution in Figure 9. The cooling rate is lower on the side walls because of the smaller contact

pressure. The cooling rate is higher at the fillet corner between the transitional surface and the

upper/lower surface because of the higher contact pressure.

In general, the non-uniform distribution of both the cooling rate and part temperature induced

by contact pressure is one of the main causes of cracks. Therefore, investigating the interfacial

heat transfer rules and solving the IHTC under different contact pressures are necessary to reveal

the change procedure of the cooling rate and part temperature. Furthermore, exploring an efficient

solution to narrow the temperature difference throughout the part during hot forming prevents the

occurrence of cracks and enhances the formability of parts.

3.2 Effects of contact pressure on blanks by experiments

11

Page 11 of 21
3.2.1 Effect of contact pressure on surface roughness

The surface of a smooth object is actually rough when scanned by a laser confocal

microscope (LEXT OLS4000), as presented in Figure 10(a). Contact occurs only at some discrete

convex points on the surface of the blank and the die. Heat is primarily transferred through the

heat conduction between these touching discrete points [Figure 10(b)]. Contact pressure changes

the shape of the asperities and consequently affects the thermal contact resistance during

hot-stamping. Figure 10(c) shows that different magnitudes of contact forces cause various

deformations of the asperities, which lead to variations of the contact areas and thermal resistance.

As a result, the IHTC changes over time.

In this experiment, a 22MnB5 steel blank was stamped under a 5 MPa contact pressure, and

the surface was scanned. Figure 11(a) and 11(b) shows the surface topography of the blank before

and after hot-stamping, respectively. The surface roughness changed from 1.032 μm to 0.986 μm.

The stamped blank became smoother than before because the top portion of the cone-shaped

asperities underwent plastic deformation and became relatively flat.

3.2.2 Effect of contact pressure on IHTC

The curves of IHTC versus temperature of the specimen center under different contact

pressures are plotted in Figure 12 by adjusting the hydraulic press. The IHTCs for each curve

increase to the corresponding peak value and then drops. At the former stage of forming and

quenching, the larger temperature difference between the blank and die contributes to the thermal

kinetic force, which drives the IHTC to increase rapidly. At the later stage of quenching, the blank

temperature approaches the die surface temperature gradually, thereby weakening the thermal

12

Page 12 of 21
kinetics and decreasing the IHTC.

Contact pressure has a significant effect on the IHTC. A high contact pressure is beneficial to

increasing the IHTC and speeding up blank-cooling. The peak IHTC reaches nearly

8000 W/(m2•K) under 20 MPa, which is only 340 W/(m2•K) under 0.008 MPa, which indicates

that the non-uniform distribution of contact pressure leads to the non-uniform distribution of

IHTC, influences the interfacial heat transfer rules, and consequently causes various cooling rates

in the part.

3.2.3 Effect of contact pressure on cooling rate and martensite transformation

The cooling curves of the tested steels are obtained corresponding to the contact pressure

from 0.008 MPa to 20 MPa, as presented in Figure 13. The cooling rate affects the martensite

transformation and mechanical properties of the part.

Martensite transformation occurs once the temperature of the part drops to the Ms

temperature. The inflexion point for each curve in Figure 13 represents the Ms temperature.

Obviously, the Ms temperature arrives earlier and the martensite transformation occurs earlier

along with the increase of contact pressure. Therefore, the non-uniform distribution of martensite

microstructure appears as quenching proceeds.

3.2.4 Effect of contact pressure on mechanical properties

Furthermore, the mechanical properties are also affected by the cooling rate. In this paper,

hardness is selected as an example to demonstrate how the cooling rate affects the mechanical

properties. The relationship curve between the hardness and cooling rate is shown in Figure 14.

13

Page 13 of 21
 The variation of hardness is more obvious in the low cooling rate area of curve, which

becomes higher with the increase in cooling rate. The hardness is about 385 HV when the cooling

rate is 27 °C/s, which corresponds to the contact pressure of 0.008 MPa. The hardness increases to

about 462 HV when the cooling rate is 40 °C/s, which corresponds to the contact pressure of

0.02 MPa. However, the hardness approaches the peak value and changes less as the cooling rate

exceeds 80 °C/s. Therefore, different IHTCs cause various cooling rates and further leads to the

non-uniform distribution of hardness for 22MnB5 steel blank.

In general, the relationship of contact pressure-IHTC-cooling rate-martensite

transformation/mechanical properties proves that the non-uniform distribution of contact pressure

in B-pillar is one of the main causes of cracks.

3.3 A solution of avoiding cracks

In this paper, local oil coating is explored as a solution to avoid cracks. Anti-oxidation oil,

which consists of oil-based fatty acid soap, boric acid and phosphate solid lubricants, metallic

soap, and sequestering agent, is selected. The anti-oxidation oil possesses a lower heat

conductivity coefficient than the die and specimen, which can slow down the heat transfer from

the specimen to the dies and consequently decrease the heat transfer efficiency. Oil coating is

considered on some of the high-IHTC areas of the blank before hot forming, and whether the

influence of oil coating on the IHTC is valid for narrowing the temperature difference or not is

investigated. In the experiment, the coating thickness is set to 40, 80, and 120 μm. The IHTC

between the oil-coated blank and die under a contact pressure of 20 MPa is solved as presented in

Figure 15.

14

Page 14 of 21
 The IHTC of oil-coated blank is obviously lower than that of the bare blank in Figure 12. The

peak IHTC of the blank with a 40 μm thick coating is only 2650 W/(m2•K), which is less than

one-third of the IHTC under 20 MPa and even less than that under 1.5 MPa for the bare blank.

Therefore, oil coating can effectively decrease the IHTC.

Furthermore, Figure 16 shows the IHTC versus temperature curves under different contact

pressures (i.e., 5, 10, 20, and 30 MPa) when the tested blank is has an oil coating 50 μm thick.

This finding shows that the IHTC difference under higher and lower contact pressures is narrowed

compared with that in Figure 12. This condition means that the distribution of IHTC may become

uniform through local oil coating treatment, thereby contributing further to the uniform

temperature field in the part.

The oil-coated B-pillar is hot stamped to confirm the actual effect of oil coating treatment

[Figure 17]. A demonstration shows that cracking did not occur. On the one hand, a more uniform

temperature field is obtained. On the other hand, the oil coating layer helps to reduce the friction

coefficient on the fillet corners of the B-pillar and improve the formability further.

4 Conclusions

(1) The distribution of contact pressures is non-uniform on the hot-formed B-pillar with a

complex 3D shape. The difference of contact pressures between the fillet corner and the side wall

results in a higher temperature difference and more non-uniform mechanical properties, which are

more likely to cause cracking.

(2) Beck’s nonlinear inverse estimation model is established to solve the IHTC of

hot-formed 22MnB5 steel. The influence of contact pressure on the IHTC is analyzed, and the

15

Page 15 of 21
conclusion that the IHTC, as well as the cooling rate, increases with the increase of contact

pressure is made.

(3) Oil coating on 22MnB5 steel blank is an effective method to weaken the heat transfer

performance. The temperature difference between the fillet corner and the side wall of a B-pillar

can be narrowed by local oil coating on the areas with high IHTC, such as the fillet corners, which

is beneficial to achieving a more uniform temperature field and mechanical properties, and

avoiding the occurrence of cracks.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (51571048,

11472072), the Fundamental Research Funds for the Central Universities (DUT15QY09) and the

Natural Science Foundation of Liaoning Province (2014028001).

References

[1] M. Naderi, S. Alireza, W. Bleck, The effect of non-isothermal deformation on martensitic

transformation in 22MnB5 steel. Mater. Sci. Eng. A., 487 (2008) 445-455.

[2] F. Li, M. Fu, J. Lin, X. Wang, Experimental and theoretical study on the hot forming limit of

22MnB5 steel. Int. J. Adv. Manuf. Technol., 71 (2014) 297-306.

[3] H. Liu, C. Lei, Z. Xing, Cooling system of hot stamping of quenchable steel BR1500HS:

optimization and manufacturing methods. Int. J. Adv. Manuf. Technol., 69 (2013) 211-223.

[4] Y. Xu, Z. Shan, Design parameter investigation of cooling systems for UHSS hot stamping

dies. Int. J. Adv. Manuf. Technol., 70 (2014) 257-262.

16

Page 16 of 21
[5] R. Kolleck, R. Veit, M. Merklein, J. Lechler, M. Geiger, Investigation on induction heating for

hot stamping of boron alloyed steels, CIRP ANN-MANUF TECHN, 58 (2009) 275-278.

[6] Y. Chang, Z. Meng, L. Ying, X. Li, N. Ma, P. Hu, Influence of hot press forming techniques

on properties of vehicle high strength steels. J. Iron Steel Res. Int. 18 (2011) 59–63.

[7] K. Zhao, Y. Chang, P. Hu, Y.C. Wu, Influence of rapid cooling pretreatment on microstructure

and mechanical property of hot stamped AHSS part, J. Mater. Process. Technol., DOI:

10.1016/j.jmatprotec.2014.09.022

[8] H. Karbasian, A. Tekkaya. A review on hot stamping. J. Mater. Process. Technol. 210 (2010)

2103-2118.

[9] J. Kim, Y. Jeon, B. Kim, C. Kang, Die Design for a center pillar part by process analysis of hot

stamping and its experimental verification. Int. J. Precis. Eng. Man., 13 (2012) 1051-1057.

[10] A Bardelcik, C. Salisbury, S. Winkler, M. Wells, M. Worswick, Effect of cooling rate on the

high strain rate properties of boron steel. Int. J. Impact Eng. 37(2010) 694-702.

[11] A. Turetta, S. Bruschi, A. Ghiotti, Investigation of 22MnB5 formability in hot stamping

operations. J. Mater. Process. Tech., 177 (2006) 396-400.

[12] K. Ikeuchi, J. Yanagimoto, Valuation method for effects of hot stamping process parameters

on product properties using hot forming simulator. J. Mater. Process. Tech., 8 (2011) 1441–1447.

[13] B. Abdulhay, B. Bourouga, C. Dessain, G. Brun, J. Wilsius, Experimental study of heat

transfer in hot stamping process, Int. J. of Mater. Form., 2(2009): 255-257.

[14] A. Blaise, B. Bourouga, B. Abdulhay, C. Dessain, Thermal contact resistance estimation and

metallurgical transformation identification during the hot stamping, Appl. Therm. Eng., 61 (2013)

141-148.

17

Page 17 of 21
[15] J. Slowik, G. Borchardt, C. Köhler, R. Jeschar, R. Scholz, Influence of oxide scales on heat

transfer in secondary cooling zones in the continuous casting process: Part II. Determination of

material properties of oxide scales on the steel under spray-water cooling conditions. Steel Res.

Int., 61(1990): 302-311.

[16] R. Wendelstorf, K. Spitzer, J. Wendelstorf, Effect of oxide layers on spray water cooling heat

transfer at high surface temperatures, Int. J. of Heat Mass Transf., 51(2008): 4892-4901

[17] P. Hu, L. Ying, Y. Li, Z. Liao, Effect of oxide scale on temperature-dependent interfacial heat

transfer in hot stamping process. J. Mater. Process. Tech., 213(9) (2013) 1475-1483.

[18] Q. Bai, J. Lin, L. Zhan, T. Dean, D. Balint, Z. Zhang, An efficient closed-form method for

determining interfacial heat transfer coefficient in metal forming. Int. J. Mach. Tools. Manuf.,

56(2012)102-110.

[19] E. Caron, K. Daun, M. Wells, Experimental heat transfer coefficient measurements during

hot forming die quenching of boron steel at high temperatures, Int. J. Heat Mass Transfer, 71

(2014) 396-404.

[20] H. Wang, W. Yu, Q. Cai, Experimental study of heat transfer coefficient on hot steel plate

during water jet impingement cooling, J. Mater. Process. Tech., 212 (2012) 1825-1831.

[21] T. Hung, P. Tsai, F. Chen, T. Huang, W. Liu, Measurement of heat transfer coefficient of

boron steel in hot stamping, Procedia Engineering, 81 (2014) 1750-1755.

[22] J. Beck, B. Blackwell, C. St Clair, Inverse heat conduction: III posed problems, Wiley

Interscience, New Jersey, USA, 1985.

[23] D. Koistinen, R. Marburger, A general equation prescribing the extent of the

austenite-martensite transformation in pure ironecarbone alloys and plain carbon steels, Acta

18

Page 18 of 21
Metall. 1-7 (1959) 59-60.

[24] Y. Chang, X. Li, K. Zhao, C. Wang, G. Zheng, P. Hu, H. Dong, Influence of stress on

martensitic transformation and mechanical properties of hot stamped AHSS parts, Mater. Sci. Eng.

A., 629 (2015) 1-7

19

Page 19 of 21
Captions

Fig. 1 Direct hot forming process

Fig. 2 Occurrence of crack on the actual hot stamped B-pillar

Fig. 3 Illustration of heat transferring experiment setup

Fig. 4 Temperature field simulation of the blank by Deform 3D. (a) temperature difference

between of the inner center (P1) and surface center (P2) in the entire quenching procedure; (b)

temperature field distribution at 1st second of quenching

Fig. 5 Temperature field simulation of the die at 10th second of quenching by Deform 3D. (a)

curve of Temperatrue versus distance from the center along the radius; (b) temperature field

distribution

Fig. 6 Temperature field distribution from the center axis along the radius on different cross

sections of the blank and die at different quenching moments. (a), (b), (c) and (d) at the 2nd, 4th, 6th,

and 8th seconds of quenching, respectively

Fig. 7 Theoretical model of Beck’s nonlinear estimation method

Fig. 8 Non-uniform distribution of temperature field during hot stamping

Fig. 9 Distribution of contact pressure on the surface of part at different steps during the hot

forming. (a) 10th step; (b) 12th step

Fig. 10 Micromorphology and schematic of actual contact surface between the die and the blank (a)

Micromorphology; (b) Schematic of heat transfer path; (c) Schematic of actual contact surface

under different force

Fig. 11 Surface topography of 22MnB5 blank under 5 MPa contact pressure: (a) Magnified by

4000 times before stamping; (b) Magnified by 8000 times after stamping

20

Page 20 of 21
Fig. 12 IHTC versus temperature curves under different contact pressures

Fig. 13 Cooling curves of the tested steel under different contact pressures

Fig. 14 Curve of hardness versus cooling rate of 22MnB5 steel

Fig. 15 IHTC between oil-coated blanks and die under 20 MPa

Fig. 16 IHTC versus temperature curves under different contact pressures for the tested blank with

50 μm oil coating

Fig. 17 Actual hot-formed B-pillar without the occurrence of crack

Tab. 1 Values of the six selected contact pressures

Tab. 1 Values of the six selected contact pressures

Six levels 1 2 3 4 5 6

The contact pressure(MPa) 0.008 0.02 1.5 3 17 20

21

Page 21 of 21