Applied Thermal Engineering 181 (2020) 115619

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

A general IHTC model for hot/warm aluminium stamping T

a,c a a a b a,⁎
Xiaochuan Liu , Zhaoheng Cai , Yang Zheng , Omer El Fakir , Joao Gandra , LiLiang Wang
a
Department of Mechanical Engineering, Imperial College London, London SW7 2AZ, UK
b
The Welding Institute, Granta Park, Great Abington, Cambridge CB21 6AL, UK
c
School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an, 710049, People’s Republic of China

HIGHLIGHTS

• AA method was developed to identify the critical process parameter in hot/warm stamping.
• Thegeneral aluminium alloy-independent IHTC model was developed.
• Dissimilar
critical contact pressures for AA6082 and AA7075 were identified.
• aluminium alloys were formed to experimentally verify the present work.

ARTICLE INFO ABSTRACT

Keywords: Different hot and warm stamping technologies with particular processing parameters were applied to deform
Alloy-independent IHTC model aluminium alloy sheets to satisfy desired requirements, of which the post-form strength of formed components is
Critical processing parameters one of the most important criteria. In order to save experimental efforts, the present research described an
Hot and warm stamping efficient method to determine the critical processing parameters, i.e. the integration of the finite element (FE)
Dissimilar joining
simulated temperature evolutions with the continuous cooling precipitation (CCP) diagrams of the aluminium
Aluminium alloys
Friction stir welding
alloys. Through the optimisation of the processing parameters, the temperature evolutions and CCP diagrams do
not intersect, indicating that the post-form strength of the aluminium alloys could be fully retained after proper
artificial ageing processes. Therefore, a precise FE simulation of the temperature evolution is of great importance
to this method, which requires the implementation of an accurate interfacial heat transfer coefficient (IHTC) as a
decisive boundary condition. A general aluminium alloy-independent model with one set of fixed model con­
stants was therefore developed to predict the IHTC evolutions as a function of contact pressure, surface
roughness, initial blank temperature, initial blank thickness, tool material, coating material and lubricant ma­
terial. Subsequently, the predicted IHTCs for AA6082 and AA7075 aluminium alloys were used to simulate their
temperature evolutions, which were then integrated with their CCP diagrams to identify the critical processing
parameters in hot/warm stamping processes and thus meet the desired post-form strength of the AA6082 and
AA7075. The developed IHTC model and determined critical processing parameters were then experimentally
verified by the Fast light Alloy Stamping Technology (FAST) of the dissimilar aluminium alloy blanks joined by
Friction Stir Welding (FSW).

1. Introduction strength-to-weight ratio, corrosion resistance, good recyclability and

ease of joining. Furthermore, the application of lightweight aluminium
The growing trend of transport electrification introduces specific alloys is beneficial for reducing carbon dioxide emissions and saving
lightweight challenges, especially in the automotive sector. fuel consumption. It was found that carbon dioxide emissions could be
Manufacturers are working to reduce the weight of the vehicle structure reduced by 10% when a vehicle structure was made from aluminium
to improve its running efficiency and performance. This is typically alloys instead of conventional steels [2]. Therefore, the usage of the
achieved by structural topology optimisation and trading traditional aluminium alloys (predominantly AA5000/6000 series and high
steel fabrication for lightweight materials [1]. Aluminium alloys are strength grades such as AA2000/7000 series) in vehicles is steadily
one of the most commercially viable options due to their attractive increasing in recent years and will dominate in the year 2040, as shown

⁎
Corresponding author.
E-mail address: liliang.wang@imperial.ac.uk (L. Wang).

https://doi.org/10.1016/j.applthermaleng.2020.115619
Received 26 January 2020; Received in revised form 14 May 2020; Accepted 16 June 2020
Available online 23 June 2020
1359-4311/ © 2020 Elsevier Ltd. All rights reserved.
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Fig. 1. Material usage in vehicles from 2010 to 2040 [3].

in Fig. 1 [3]. with increasing heating rate. A similar effect was also described in the
However, the low formability of the aluminium alloys at room research of Zheng et al. [17], in which the effect of forming temperature
temperature significantly limits their applications when forming of on the post-form strength of AA7075 was additionally investigated.
complex-shaped components [4–6]. In order to overcome this draw­ When the heating rate was not sufficient, the coarse particles pre­
back, Chu et al. [7] proposed a disruptive hydro-forging process, which cipitated at a growth rate that increased with decreasing forming
expanded the horizon of lightweight manufacturing for aluminium tube temperature, thus reducing the post-form strength. Although the SHT
components. Meanwhile, hot and warm stamping technologies were temperature activated the precipitates to be dissolved into the alumi­
also developed to enable the forming of aluminium sheet components at nium matrix, the subsequent soaking time and quenching rate de­
elevated temperatures [8–10]. El Fakir et al. [11] investigated how the termined whether a supersaturated solid solution (SSSS) state could be
solution heat treatment, forming and quenching (HFQ) technology obtained and thus the material strength could be fully retained after
could be applied to form AA5754-H111 sheets with a thickness of artificial ageing. Fan et al. [18,19] found that the post-form strength of
1.5 mm. The blank was first heated in a furnace to its solution heat 6A02 aluminium alloy increased from 134.7 to 315.6 MPa when the
treatment (SHT) temperature of 480 °C at 1 °C/s, followed by a transfer soaking time increased from 5 to 50 min. As a result of a better dis­
from the furnace to a press machine within 10 s. Subsequently, the cold solution of the precipitates into the aluminium matrix at a longer
forming tools deformed the hot blank into the desired shape at soaking time, the post-form strength of the aluminium alloy was
250 mm/s. A different hot stamping process was used in the study of therefore larger. In addition, the post-form strength also increased with
Maeno et al. [12]. Specifically, a 1.3 mm thick AA2025-T4 was heated increasing quenching rate because of the rapid freeze of the SSSS state.
by electrical resistance to a temperature below its solution heat treat­ Under the hot stamping conditions, the quenching rate for AA7075 has
ment temperature and subsequently transferred to a press machine to achieve 450 °C/s to prevent secondary phase from being precipitated,
within 0.2 s, followed by deformation in cold forming tools. In Fast light obtaining a high strength after artificial ageing [20,21].
Alloys Stamping Technology (FAST), a blank was rapidly heated to an Currently, the selection and optimisation of decisive processing
elevated temperature, and then deformed and quenched within cold parameters in hot/warm stamping processes heavily relied on abundant
tools, in order to obtain a desired mechanical strength after a proper experiments in previous work. Aiming to expedite experimental process
artificial ageing process. This approach proved to be effective at redu­ development, the present research developed an efficient method to
cing the overall cyle time by eliminating a time-consuming solution determine the critical processing parameters, i.e. the integration of the
heat treatment before forming [13,14]. Additionally, components from finite element (FE) simulated temperature histories with continuous
AA5754, AA6082 and AA7075 were formed in the studies of Cai et al. cooling precipitation (CCP) diagrams, which represent the precipitation
[15] and Palumbo et al. [16] at various forming temperatures from 200 behaviour of the aluminium alloys as a function of temperature and
°C to 500 °C and forming speeds from 75 to 350 mm/s. Different hot and time [22]. When the critical cooling rate is not exceeded, coarse par­
warm aluminium stamping technologies with particular processing ticles are precipitated in the aluminium grains, leading to a non-
parameters were applied to fulfil the structural requirements with homogenous distribution of the primary precipitates, e.g. Fe, Mn, Cr
special focus on optimising the post-form strength of the formed com­ and Si, thereby decreasing the material strength [23,24]. It was found
ponents. that the cooling rate of an aluminium alloy is sufficient to fully retain
In order to achieve a high mechanical strength of the formed the post-form strength only when its temperature evolution and CCP
components, considerable efforts have been made by several re­ diagram do not intersect each other [24]. Therefore, a precise FE si­
searchers to study and optimise the processing parameters in various mulation of the temperature field is of great importance to this method,
forming technologies. Maeno et al. [12] proved that the strength of which requires the implementation of an accurate interfacial heat
AA2024 could be fully retained when using a fast heating rate of ap­ transfer coefficient (IHTC) as a decisive boundary condition [25–27].
proximately 120 °C/s. However, it decreased with decreasing heating Liu et al. [28,29] subsequently developed an experimentally-verified
rate, reaching approximately 75% of that of the as-received material model to predict the IHTC evolutions for aluminium alloys as a function
when a slow heating rate of 3 °C/s was used. The solid solution of the of contact pressure, surface roughness, tool and lubricant materials. The
aluminium alloy was maintained, and only small clusters were dis­ effects of coating material and initial blank temperature were ac­
solved at a fast heating rate. Hence, its post-form strength increased counted into the model in their subsequent research [30,31]. However,

2
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

the model constants were always aluminium alloy-dependent and thus positive linear effect of initial blank temperature on the IHTC consisted
had to be re-calibrated according to each experimental result. of two factors, i.e. the thermal properties and strength of the aluminium
In the present research, a novel efficient method was developed for alloy [31]. The IHTC increases with increasing thermal properties of the
identifying the critical processing parameters in hot and warm blank due to its better heat transfer capability at a higher temperature.
stamping processes for aluminium alloys. A general aluminium alloy- The positive effect of thermal properties of the blank on the IHTC was
independent model with one set of fixed constants was first developed identified by the temperature dependent thermal diffusivity [37], as
to predict the IHTC evolutions as a function of the main governing shown in Eq. (3):
factors, namely contact pressure, surface roughness, initial blank tem­
k s (T )
perature, initial blank thickness, tool material, coating material and = B (T ) ,
(T ) cp (T ) (3)
lubricant material. Subsequently, the predicted IHTC evolutions for
6082 and 7075 aluminium alloys were used to simulate their tem­ where B (T ) is a temperature dependent parameter, ks (T ) , (T ) and
perature evolutions in FAST forming processes, which were then in­ cp (T ) are the thermal conductivity, density and heat capacity of the
tegrated with the CCP diagrams to identify the critical processing aluminium alloy at the target initial blank temperature respectively.
parameters in terms of the desired post-form strength of the aluminium Arrhenius equation is widely used to describe the temperature depen­
alloys. The method was validated experimentally in the FAST forming dence of a reaction rate, which can be either alloy-dependent [38] or
of dissimilar alloy blanks joined by Friction Stir Welding. Post-form -independent [39]. Thus, B (T ) is modelled by Eq. (4) using the Ar­
hardness values were measured to verify the developed IHTC model, rhenius equation.
the determined critical processing parameters and the overall feasibility
Qb
of the method described. B (T ) = b0 exp ,
RT (4)
2. A general aluminium alloy-independent IHTC model where R is the molar gas constant, T is the absolute temperature, b0 and
Qb are model constants. Therefore, the temperature dependent thermal
A general heat transfer model was first introduced by Cetinkale and diffusivity is able to characterise the rate of heat transfer at different
Fishenden [32] as a sum of two parts: (i) the heat transfer across the initial blank temperatures. The effect of material strength of the alu­
interfacial air gaps and (ii) the heat transfer enabled by the solid con­ minium blank on the IHTC was integrated into NP , as shown in Eq. (5):
tact between the blank and tooling. Subsequently, their independences
P
of the overall heat transfer were verified by both Rapier et al. [33] and NP = 1 exp f ,
Cooper et al. [34] respectively. Furthermore, the heat transfers across U (5)
the lubricant and coating layers were also proven as independent where is a model constant, f is a tempering correction factor, P is the
contributions to the overall heat transfer, as stated in the studies of applied pressure, and U is the temperature dependent ultimate
Wilson et al. [35] and Antonetti et. al. [36] respectively. In a hot/warm strength of the blank. Due to the asperities on the contacting surfaces,
stamping process, the heat transfer between two contacting solids the real contact area at the interface is less than the apparent value
mainly depends on four physical mediums, namely air, metallic solid, before compression. When an aluminium blank is heated to elevated
lubricant and tool coating. Based on previous research, the general temperatures, its strength is much lower than that of the steel tools at
IHTC model was therefore considered as a sum of dominant heat room temperature. Consequently, the asperities on the blank contact
transfer mechanisms as a result of these four mediums, as shown in Eq. surface are deformed by the forming tools at a defined contact pressure
(1): during compression, leading to the increased real contact area and
h = ha + hs + hl + hc , (1) IHTC. When the applied pressure reaches its convergent value, the real
contact area approaches its apparent value, leading to the peak IHTC. It
where ha is the air-contact IHTC, hs is the solid-contact IHTC, hl is the was found that the real contact area divides by its apparent value is
lubricant-contact IHTC, and hc is the coating-contact IHTC. Due to as­ equivalent to the applied pressure divides by the ultimate strength of
perities on two contacting surfaces, a large number of vacancies exist at the aluminium blank [40]. Meanwhile, a logarithmical increasing trend
the interface when a blank contacts forming tools. Consequently, the of the real contact area with pressure was identified [41,42]. Therefore,
heat transfer across the air gap becomes the dominant mechanism when NP and thus hs are supposed to logarithmically increase with increasing
the blank is exposed to air before compression by the tools. However, ratio of P to U .
this period is short considering a high stamping speed being applied. As mentioned before, the aluminium strength is the other functional
Furthermore, the heat transfer across the air gap is relatively negligible factor on the effect of initial blank temperature on the IHTC, and it
when compared to the magnitude of heat transfer induced by the decreases with increasing temperature. As a result, more asperities on
contact pressure, lubricant and/or tool coating when the blank is fully the blank contact surface are deformed at a higher initial blank tem­
compressed by forming tools. Therefore, the air-contact IHTC ha is of perature, leading to the increased real contact area and IHTC. Hence,
less interest in the present research and thus assumed as a constant the material strength of the blank negatively influences the IHTC, and
value, determined by the previous experimental results at a contact its temperature dependence was modelled as Eq. (6) using the Ar­
pressure of 0 MPa under dry and uncoated conditions. rhenius equation [43].
When a contact pressure is applied between the blank and forming
tools, the heat transfer between the two metallic solids dominates. The Q
U = 0 exp ,
factors influencing the interfacial conditions, e.g. initial blank tem­ RT (6)
perature, tool material, surface roughness and contact pressure, also where 0 and Q are model constants, identified by the high-tempera­
affect the solid-contact IHTC hs , which is characterised as Eq. (2): ture uniaxial tensile tests. The ultimate strength of the material greatly
Kst depends on tempering. The tempering correction factor f is therefore
hs = NP L,
Rst (2) applied to enable NP to predict the deformation of different tempered
alloys, as shown in Eq. (7), where U (Tx ) and U (T 6) are the ultimate
where is the temperature dependent thermal diffusivity of the blank, strength of the aluminium alloy under the present tempering conditions
L is a blank thickness dependent parameter, Kst is the equivalent and the T6 conditions respectively.
thermal conductivity of the interface between the blank and forming
tools, Rst is the equivalent interfacial surface roughness, and NP is a U (Tx )
f= ,
contact pressure dependent parameter. It has been proven that the U (T 6) (7)

3
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

The amount of heat transfer increases with increasing thermal interface are not changed by the application of the lubricant, and thus
conductivities of the two contacting solids, leading to the increased the interfacial surface roughness Rst is maintained.
IHTC. The equivalent thermal conductivity of the interface between the Furthermore, the IHTC increases with increasing lubricant layer
blank and forming tools Kst was therefore applied to describe the cap­ thickness, as a result of more vacancies at the interface being filled by
ability of the interface to conduct heat, as shown in Eq. (8): the lubricant to enhance the heat transfer. However, when the lubricant
2 thickness achieves a convergent value, the vacancies are fully filled, and
Kst = , the redundant lubricant is squeezed out of the contact surfaces.
ks 1 + kt 1
(8)
Consequently, the further increasing lubricant thickness no longer has
where ks and kt are the thermal conductivities of the aluminium blank an effect on the IHTC. Therefore, N and thus hl have a logarithmical
(specimen) and forming tools at their initial (forming) temperatures. In increasing relationship with the lubricant layer thickness l , as shown in
contrast, the IHTC decreases with increasing surface roughness because Eq. (13).
of the reduced real contact area between the blank and forming tools
[44]. The negative influence of surface roughness was modelled as Eq. N =1 exp( l ), (13)
(9): where is a model parameter. Coatings have been widely used in the
Rst = sin 2
Rs + Rt 2,
(9) hot/warm stamping processes for improving the resistance to oxidation,
corrosion and wear resistance. Differently from a lubricant being ap­
where Rs and Rt are the average (mean) surface roughness of the alu­ plied as an independent heat transfer medium, a tool coating is stably
minium blank (specimen) and forming tools respectively before com­ deposited onto the contact surfaces of the forming tools. Therefore, the
pression, generally describing the height variations in the contact sur­ heat transfer mechanism on the lubricant-contact IHTC hl is different
faces. The root mean squared value was used to describe the roughness from that on the coating-contact IHTC hc , which was modelled as Eq.
condition at the interface. Additionally, is the initial deformation (14):
angle of the blank contact profile, and thus sin describes the mean
modulus of the slope of the blank contact profile [34,45]. The de­ ks
hc = tan ·ln(kc / kl ) c · NP ,
A (14)
formation of the blank by the forming tools has two forms. When Rs is
smaller than Rt , the forming tools coarsen the surface profile of the where is a model parameter, kc is the thermal conductivity of the tool
blank to increasingly mesh the contacting surfaces, thus leading to a coating, c is the layer thickness of the tool coating, and A is the ap­
larger real contact area; when Rs is larger than Rt , the forming tools parent contact area between the blank and forming tools. Because heat
smoothen the surface profile of the blank, resulting in a similar con­ transfers from the hot blank to the cold coated tools across the contact
sequence, i.e. a larger real contact area. However, the initial deforma­ area, the coating-contact IHTC hc is determined by three terms. The first
tion angle is different under these two forms, which is assumed as 20° term of ks tan / A represents the thermal energy at the high potential
when Rs is smaller than Rt , and 70° when Rs is larger than Rt . Therefore, (hot blank), the second term of ln(kc / kl ) c represents the thermal energy
the interfacial surface roughness Rst represents the initial roughness and at the low potential (cold coated tools), and the third term NP de­
deformation conditions at the interface. termines the pressure dependent driving force from the high to low
In the hot/warm stamping industry, blanks with different thick­ potential. Furthermore, the second term ln(kc / kl ) c describes the
nesses are applied to satisfy the desired requirements. Although the thermal performance of the tool coating and the integrated effects of
interfacial conditions are independent of the blank thickness, a thicker coated tools on the IHTC. A positive term value indicates that the ap­
blank will be capable of absorbing and storing a higher internal thermal plied tool coating has a higher thermal conductivity than that of the
energy, which could compensate for the heat loss at the interface. substrate, contributing to a larger IHTC value; while a negative term
Therefore, an engineering IHTC able to define the effect of blank value indicates that the applied tool coating has a lower thermal con­
thickness can be used in the FE model to accurately simulate the tem­ ductivity than that of the substrate, contributing to a smaller IHTC
perature field, while the true IHTC is not changed. The blank thickness value. The effect of tool coating on the IHTC, either positive or nega­
parameter L is able to describe its positive effect on the IHTC and tive, increases with increasing absolute value of this term. Therefore,
modelled as Eq. (10), where l is the blank thickness, m and n are model the second term ln(kc / kl ) c indicates that the thermal conductivity and
constants. layer thickness of the applied tool coating determine its effect on the
L = m ln(l) + n, (10) IHTC.
Therefore, Eqs. (1)–(14) comprise the general model to predict IHTC
Lubricants are widely used in the hot/warm stamping processes to evolutions for different aluminium alloys with contact pressure, initial
increase the drawability of the blank material, as well as decreasing the blank temperature, initial blank thickness, tool material, surface
wear of the forming tools. Due to its importance, the lubricant-contact roughness, lubricant and tool coating. Instead of determining new IHTC
IHTC hl was developed as Eq. (11): results, the present research applied the experimentally verified IHTC
Kslt results in the authors’ previous research [28–31] to calibrate the alloy-
hl = N,
Rst (11) independent model constants, as shown in Table 1, which were opti­
mised by Genetic Algorithm as shown in Fig. 2. In addition to these 10
where is a model constant, Kslt is the equivalent mean thermal con­ model constants, the other 16 material parameters, e.g. thermal con­
ductivity of the interface between the blank, forming tools and lu­ ductivity, surface roughness and lubricant/coating layer thickness, re­
bricant, and N is a lubricant thickness dependent parameter. quire actual measurements to be assigned in the model, which are
3
Kslt = ,
ks 1 + kl 1 + k t 1
(12) Table 1
The IHTC model constants.
where kl is the thermal conductivity of the lubricant. When a lubricant
is used at the interface as a heat transfer medium, the vacancies be­ Parameter b0 (s/m2) Qb (J/mol) R (J/molK) (–) m (–)

tween the blank and forming tools are filled by the lubricant, instead of Value 1.69 −1730 8.314 5 0.64
air. Therefore, the equivalent thermal conductivity of the interface
Parameter n (–) (–) (m−1) (–) A (m2)
between the blank, forming tools and lubricant Kslt is more accurate to
define the capability of the interface to conduct heat under the lu­ Value 0.56 4.2e−5 1.5e5 8.3e3 5e−4
bricated conditions. Meanwhile, the surface roughness conditions at the

4
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Fig. 2. The flow chart of Genetic Algorithm to optimise the model constants.

shown in Table 2. The characterisation of the alloy-independent model As shown in Fig. 3, the IHTC evolutions were always given as a
constants significantly simplifies the model and enhances its flexibility function of contact pressure to cooperate with their implementation in
for different aluminium alloys. the commercial FE software. In order to highlight the effects of other
Fig. 3 shows the comparison between the experimentally verified influential factors, the IHTC evolutions were predicted by the devel­
IHTC results under different conditions determined in the previous re­ oped model as a function of the initial blank temperature shown in
search [28–31] and the IHTC evolutions predicted by the developed Fig. 4 (a), the initial blank thickness shown in Fig. 4 (b), the thermal
model using the actual measured parameters in those research. In order conductivity of tools shown in Fig. 4 (c) and the thermal conductivity of
to demonstrate the comprehensive capability of the developed IHTC coatings shown in Fig. 4 (d). The material conditions shown in Fig. 4,
model, the representative experimentally-verified IHTC results were e.g. thickness, surface roughness, temperature and tool material, were
selected, and the corresponding actual material conditions, e.g. thick­ identical to those applied in the actual experiments shown in Fig. 3.
ness, surface roughness, temperature and tool material, were also
shown in the figure. Specifically, Fig. 3 (a) shows the IHTC for AA6082
with different thicknesses at different contact pressures to demonstrate 3. Critical processing parameters in hot/warm stamping processes
the effects of blank thickness and contact pressure on the IHTC [28],
Fig. 3 (b) shows the IHTC for AA6082 under the dry and lubricated As previously mentioned, the predicted IHTC evolution can be im­
conditions to demonstrate the effect of lubricant on the IHTC [28], plemented in the FE models of hot/ warm stamping processes to si­
Fig. 3 (c) shows the IHTC between AA7075 and steel tool with different mulate the temperature evolutions of the aluminium alloys, which are
coatings to demonstrate the effect of tool coating on the IHTC [30], then compared with the CCP diagrams to identify whether the critical
while Fig. 3 (d) shows the IHTC for AA7075 at different initial tem­ cooling rate and IHTC are reached. Therefore, the critical processing
peratures to demonstrate the effect of initial blank temperature on the parameters can be optimised to meet the desired requirements through
IHTC [31]. Meanwhile, the IHTCs between two different aluminium the integration of the FE simulations with the CCP diagrams. The pre­
alloys and four different tool materials were also shown in this figure, sent research identified the critical contact pressures for 6082 and 7075
demonstrating the effects of blank and tool materials on the IHTC. The aluminium alloys under the FAST forming conditions. The same method
excellent agreements between the experimental and predicted IHTC can be applied to determine other processing parameters, e.g. forming
results suggest a high accuracy of the developed model. temperature, heating rate and soaking time [46].

Table 2
The thermal conductivity, surface roughness and thickness of the materials.
Materials AA6082 [28] AA7075 [29] H13 [29] P20 [29] G3500 [29] D6510 [31]

Thermal conductivity (W/mK) 170 140 24.4 31.5 44 35.2

Surface roughness (nm) 430 340 980 960 810 180

Materials CrN [30] TiN [30] WC-Co [31] Graphite lubricant [28]

Thermal conductivity (kW/mK) 12 19 29.2 24

Thickness (μm) 6 8 2 –

5
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Fig. 3. The comparisons between the experimental and predicted IHTC results for (a) AA6082 with 2 and 5 mm thicknesses under dry conditions at 535 °C when
using uncoated P20 tools [28]; (b) AA6082 with 3 mm thickness under lubricated and dry conditions at 535 °C when using uncoated P20 tools [28]; (c) AA7075 with
2 mm thickness under dry conditions at 490 °C when using uncoated, CrN and AlCrN-coated tools [30]; and (d) AA7075 with 2 mm thickness under dry conditions at
420 °C and 350 °C when using WC-coated tools [31].

3.1. FE simulation setup to identify the critical contact pressure automatically defined as three in the FE model. Due to the heating
process being not simulated, the initial temperatures for the aluminium
In order to characterise the critical contact pressure in the FAST blank and other components had to be assigned as the target heating/
forming of the aluminium alloys, a FE model was developed in PAM- forming temperature and room temperature respectively. In addition,
STAMP software to simulate the heat transfer between an aluminium all six degrees of freedom for the aluminium blank were defined as free,
blank at an elevated temperature and tools at room temperature. This those for the die were restricted, and only freedom in the z-direction
model was composed of seven components; an aluminium blank, two was free for the punch, blankholders and screws.
symmetrical blankholders, two symmetrical screws and a symmetrical A ‘hot forming validation double action’ strategy was used in the FE
punch and die pair, of which the geometries were identical to those in model, in which the aluminium blank was first freely located onto the
the dedicated heat transfer test facility, precisely representing the heat two blankholders and then compressed by the two screws at a pre-de­
transfer test, as shown in Fig. 5 (a) [30]. It has been proven that a mesh fined load/pressure, which was a variable to be identified in this re­
size of 2 × 2 mm2 would ensure accurate simulations while providing search. Subsequently, the punch moved towards the blank along the
reasonable computational times by the mesh sensitivity analysis in the moving direction (z-direction) at a speed of 100 mm/s and compressed
authors’ previous research [31]. Thermal shell elements with a size of it against the die at a load/pressure the same as the blankholding force
2 × 2 mm2 were therefore used to mesh the forming tools and blan­ for 10 s. The temperature evolutions of the blank during compression
kholders, while the elements with a smaller size of 1 × 1 mm2 were was measured and exported. In order to characterise the critical pres­
used to mesh the Al blank, in order to demonstrate a finer quenching sure, the IHTC curve was implemented in the FE simulation, defined as
distribution in Section 3.2. Due to their irregular basements, the punch a function of pressure with the restriction of other processing para­
and die had to be imported to the FE model as rigid bodies that only meters, e.g. heating temperature (initial blank temperature) and tool
their surfaces were meshed by the thermal shell elements. The tem­ materials. Therefore, the heat transfer from the hot aluminium blank to
perature gradient along the thickness direction of a sheet blank is cold forming tools occurred at the IHTC value corresponding to the pre-
negligible in a hot/warm stamping process [47]. Therefore, the number defined contact pressure. In the present research, when a P20 steel was
of the integration point through the Al blank thickness was used as forming tools, the IHTCs for AA6082 and AA7075 under the

6
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Fig. 4. The predicted IHTC evolutions as a function of (a) initial blank temperature; (b) initial blank thickness; (c) thermal conductivity of tool; and (d) thermal
conductivity of coating.

Fig. 5. (a) The FE model in PAM-STAMP to determine the critical processing parameters; (b) The predicted IHTC evolutions for the AA6082 and AA7075 as a
function of contact pressure under the FAST forming conditions.

7
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Table 3 shows the FSW blank geometry, measuring 600 mm long and 300 mm
The material properties defined in the FE model. wide. The FSW tool used combined a concave shoulder of a 15 mm
Materials AA6082 [28] AA7075 [29] P20 [28] diameter with a Triflute™ probe of a 5 mm diameter. The tool was tilted
at 2°, plunged into the material at the joint starting location using a
Young’s modulus (GPa) 70 140 205 rotation speed of 17 rev/s. A dwell time of 1 s was employed once the
Yield strength (MPa) 250 420 840
probe reached its target penetration depth, subsequently initiating the
Density (kg/m3) 2700 2707 7850
Thermal conductivity (W/mK) 170 140 31.5
travel motion at 12 mm/s. The weld cycle was conducted entirely on
Specific heat capacity (J/kgK) 890 1060 473 position control. The welding procedure was selected based on prior
experience at TWI Ltd. FSW blank design and sample extraction was
based on ISO 25239-1:2011.
FAST forming conditions were predicted and implemented in the FE
model, as shown in Fig. 5 (b). The material viscoplastic models for 4.2. FAST forming of panel components
AA6082 [48] and AA7075 [49] were implemented in the FE model,
while the material properties defined in the FE model were shown in Cross weld samples 90 mm long and 10 mm wide were extracted
Table 3. from the dissimilar alloy blanks to form M-shaped panel components
The developed FE model enables symmetrical heat transfer from the under the FAST forming conditions, and their post-form hardness was
hot blank to cold tools as well as the homogeneous distribution of the subsequently measured to verify the determined critical contact pres­
load/pressure on the contact area of the aluminium blank. In addition sures for both aluminium alloys. As shown in Fig. 8 (a), a dedicated
to contact pressure, other processing parameters were also allowed to forming facility was equipped with the M-shaped forming tools and
be assigned in the FE model to reveal their effects on the temperature then assembled in a Gleeble 3800 thermal-mechanical test machine to
evolutions of the blank, in order to determine their critical values. perform the FAST forming processes. Due to its precise automatic
control, this forming facility is beneficial for accurate validation of the
3.2. Integration of the FE simulated results with the CCP diagrams processing parameters under different conditions.
A dissimilar alloy blank was screwed onto the blankholders for each
After the FE simulations, the temperature evolutions of all elements test. A cold forming process at room temperature was first conducted,
on the aluminium blank were exported and then compared with the resulting in a brittle fractured in the AA7075 parent material, as de­
CCP diagrams for 6082 and 7075 aluminium alloys, which were char­ monstrated in Fig. 8 (b). This could be expected due to the lower
acterised in the studies of Milkereit et al. [50,51]. A filter was then ductility of AA7075 compared to that of AA6082. A second dissimilar
applied to distinguish ‘safe’ elements from the entire aluminium blank. aluminium alloy was deformed under the FAST forming conditions,
Elements, for which their temperature evolutions did not intersect the envisaging to increase the formability while preserving its post-form
CCP diagram were defined as ‘safe’ elements shown in green as their strength of the material. A blank was rapidly heated to the target
post-form strength can be fully retained; otherwise, they were defined temperature. Subsequently, the cold punch was activated to move along
as ‘fail’ elements shown in red. It has been proven that the IHTC and the guide pillars towards the cold die and deform the blank into an M-
cooling rate of the aluminium alloy increase with increasing pressure, shaped panel component, as demonstrated in Fig. 8 (c), followed by
consequently leading to an increased number of ‘safe’ elements. Re­ quenching of the component to room temperature at different pressures
garding the contact area on the AA6082 blank, all elements were ‘safe’ of 10, 18 and 28 MPa. Meanwhile, the other processing parameters
when the contact pressure was larger than 18 MPa, as demonstrated in were set as the same as those used in previous FE simulation. After
Fig. 6 (a). This indicated that the critical contact pressure for 6082 appropriate artificial ageing processes were undertaken, the post-form
aluminium alloy is 18 MPa under the FAST forming conditions. hardness of the components formed at the three different contact
Similarly, different contact pressures were applied in the FE simu­ pressures was measured by a Zwick ZHU hardness tester.
lation to identify the critical value for AA7075. The number of ‘safe’ As shown in Fig. 9, the hardness values of the as-received AA7075
elements increased with increasing pressure as well. Until the and AA6082 were 181 and 121 HV respectively. The hardness profile
achievement of 28 MPa, all elements on the contact area of the alu­ across the weld region exhibited a gradient between these two nominal
minium blank were ‘safe’, as shown in Fig. 6 (b). This indicated that the values. This is consistent with the mixing of the dissimilar aluminium
critical contact pressure for 7075 aluminium alloy is 28 MPa under the alloys in the welding zone. The target was to achieve at least 95% of the
same conditions, which is larger than that for 6082 aluminium alloy. parent material original strength following forming and artificial
The identification of the critical contact pressures for different alumi­ ageing. When the contact pressure was 10 MPa, the post-form hardness
nium alloys is of great importance to not only ensure that the post-form of the AA7075 was approximately 164 HV, which was 9.4% lower than
strength of the materials can be fully retained but also prevent the that of the as-received material, while the post-form hardness of the
excessive contact pressure from being applied. It should be noticed that AA6082 was approximately 103 HV with a 14.9% loss in its as-received
the critical pressure would change in different forming processes. Under value. The insufficient post-form hardness of both AA6082 and AA7075
some particular conditions, the critical cooling rate of the aluminium was due to the critical contact pressure being not reached. Increasing to
alloys may not be reached, regardless of the contact pressure applied. In the contact pressure to 18 MPa proved to fulfil the critical value for the
this case, the application of lubricants and tools with higher thermal AA6082 only. As a result, the post-form hardness of the AA6082
conductivities should be considered as potential solutions. Therefore, reached 97% of its as-received hardness, while that of the AA7075 did
the critical processing parameters have to be particularly characterised not meet the target yet. When the contact pressure was 28 MPa, the
according to the applied forming processing windows. post-form hardness of both AA6082 and AA7075 were equivalent to
that of the parent material conditions, reaching 120 and 179 HV re­
4. Experimental validation of the critical contact pressures spectively. The experimental observations agreed well with the pre­
vious deduction, i.e. the post-form hardness/strength of the material
4.1. FSW of dissimilar alloy blanks can be fully retained only when its critical contact pressure is achieved.

The dissimilar alloy blanks were produced by FSW AA6082 and 4.3. FE simulation of FAST forming of dissimilar alloy FSW blanks
AA7075 sheets with a thickness of 2 mm, which were supplied by
Smiths Metal Centres Limited. The FSW technology was deployed using Meanwhile, the FE simulation of the FAST forming of the dissimilar
an AWEA LP 4025Z FSW machine based at TWI Ltd in Cambridge. Fig. 7 alloys was performed in PAM-STAMP to predict the ‘safe/fail’

8
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Fig. 6. The temperature evolutions of all elements, the temperature evolutions of the ‘safe’ elements, after filtering, and the distributions of ‘safe/fail’ elements on (a)
the AA6082 blank at the critical pressure of 18 MPa; and (b) the AA7075 blank at the critical pressure of 28 MPa.

distribution on the formed component under different contact pressure freedom of degrees for all components and the simulation strategy were
conditions. The FE model composed of three components, i.e. a blank identical to previous simulations, i.e. the hot blank C was freely located
made from the dissimilar aluminium alloys, a punch and die made from onto the cold die, and the cold punch instantly moved along the z-
the P20 tool steel, of which geometries were identical to those used in direction at a speed of 100 mm/s to deform the blank into an M-shaped
the experiments, as shown in Fig. 10. Similar to previous FE simulation component, followed by quenching at pressures of 10, 18 and 28 MPa
identifying the processing parameters, the same quadrangle thermal respectively for 10 s. The friction coefficients for AA6082 and AA7075
shell elements with sizes of 1 × 1 mm2 and 2 × 2 mm2 were used for were defined as 0.15 [52] and 0.3 [47] respectively. The heat transfer
the blank and forming tools respectively. Additionally, the definition of from the hot blank to cold forming tools occurred at the IHTC evolution

9
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Fig. 7. Dimensions (in mm) of the FSW sheet.

being assigned. Subsequently, the temperature evolutions of the blank

were exported and then compared with the CCP diagrams.
After filtering, the ‘safe/fail’ distributions on the M-shaped compo­ Fig. 9. The hardness distributions on the M-shaped components at different
nent under different contact pressure conditions were shown in Fig. 11. pressures.
Similarly, the ‘safe’ elements were shown as green colour, while the
‘fail’ elements were shown as red colour. Meanwhile, the shades of red high cooling rate and material strength can be achieved after a proper
colour represent the level of ‘fail’. Due to the insufficient contact on the artificial ageing process.
side vertical walls of the blank, their post-form strength was not able to
achieve a high value under the applied forming conditions. Coincident
5. Conclusion
with the experimental results, the entire blank was failed to achieve a
high post-form strength at 10 MPa, while only the AA6082 part was
The present research developed a novel efficient method for iden­
‘safe’ when the pressure increased to 18 MPa. Apart from the vertical
tifying the critical processing parameters in hot and warm stamping
walls, the rest of the blank was ‘safe’, fully retaining the material
processes for aluminium alloys. A general aluminium alloy-independent
strength for a forming pressure of 28 MPa. Both the experimental and
model was first developed to predict the IHTC evolutions as a function
simulated results verified that the critical pressures for AA6082 and
of contact pressure, surface roughness, initial blank temperature, initial
AA7075 are 18 and 28 MPa respectively under the FAST forming con­
blank thickness, tool material, coating material and lubricant material,
ditions. Therefore, the developed method could enable the accurate
using one set of fixed model constants. Through the integration of the
identification of the processing parameters under different forming
temperature evolutions predicted by the IHTC model with the CCP
conditions. Furthermore, the FE simulation of forming processes can be
diagrams, the critical contact pressures of 18 MPa for AA6082 and
first conducted to optimise the tool design to eliminate insufficient
28 MPa for AA7075 were characterised under the FAST forming con­
contact in particular areas, e.g. side walls and corners, ensuring that
ditions.

Fig. 8. (a) The forming test facility; (b) A M-shaped component formed at room temperature; (c) A M-shaped component formed under the FAST forming conditions.

10
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

Fig. 10. The FE model of a forming process under (a) unloading; and (b) loading conditions.

Subsequently, FSW joining AA6082 and AA7075 were formed into

M-shaped panel components at different contact pressures under the
FAST forming conditions. The post-form hardness of the dissimilar FSW
blanks achieved approximately 97% of the parent material original
hardness at the identified critical contact pressure. Meanwhile, the FE
simulation of the FAST forming of the dissimilar alloys was conducted
to predict whether the post-form strength of the formed components
was reached or not at different contact pressures, verifying the accuracy
of the identified critical contact pressures and developed general IHTC
model as well as the feasibility of the described method. The key
findings were summarised below:

(1) In addition to the critical contact pressure in the FAST forming, the
described efficient method is able to characterise other critical
processing parameters in different hot and warm stamping pro­
cesses for aluminium alloys, dramatically saving the experimental
efforts.
(2) The developed general aluminium alloy-independent IHTC model is
capable of predicting the temperature evolutions under the desired
forming conditions.
(3) Dissimilar FSW blanks joining AA6082 and AA7075 could be
formed using FAST forming, allowing to restore 95% of the original
parent material hardness.

Declaration of Competing Interest

None declared.

Acknowledgement

The strong support from the Institute of Automation, Heilongjiang

Academy of Sciences, for this funded research is much appreciated.

References

[1] A. Smallbone, B. Jia, P. Atkins, A.P. Roskilly, The impact of disruptive powertrain
technologies on energy consumption and carbon dioxide emissions from heavy-duty
vehicles, Energy Convers. Manag. X. 6 (2020) 100030, , https://doi.org/10.1016/j.
ecmx.2020.100030.
[2] C. Ungureanu, S. Das, I.S. Jawahir, Life-cycle cost analysis: aluminum versus steel in
Fig. 11. The ‘safe/fail’ distributions on the M-shaped component made from passenger cars, Miner. Met. Mater. Soc. (2007) 11–24.
[3] S. Brett, A. Spulber, S. Modi, T. Fiorelli, Technology roadmaps: Intelligent mobility
dissimilar aluminium alloys at different contact pressures of 10, 18 and 28 MPa. technology, materials and manufacturing processes, and light duty vehicle pro­
pulsion, Cent. Automot. Res. (2017), https://doi.org/10.1177/1060028017712725.
[4] X. Fan, Z. He, K. Zheng, S. Yuan, Strengthening behavior of Al-Cu-Mg alloy sheet in
hot forming-quenching integrated process with cold-hot dies, Mater. Des. 83 (2015)

11
X. Liu, et al. Applied Thermal Engineering 181 (2020) 115619

557–565, https://doi.org/10.1016/j.matdes.2015.06.058. j.jmatprotec.2017.12.013.

[5] Y. Cai, X. Wang, S. Yuan, Analysis of surface roughening behavior of 6063 alu­ [29] X. Liu, K. Ji, O. El Fakir, H. Fang, M.M. Gharbi, L. Wang, Determination of inter­
minum alloy by tensile testing of a trapezoidal uniaxial specimen, Mater. Sci. Eng. facial heat transfer coefficient for a hot aluminium stamping process, J. Mater.
A. 672 (2016) 184–193, https://doi.org/10.1016/j.msea.2016.07.008. Process. Technol. 247 (2017) 158–170, https://doi.org/10.1016/j.jmatprotec.
[6] J.N. Rasera, K.J. Daun, C.J. Shi, M. D’Souza, Direct contact heating for hot forming 2017.04.005.
die quenching, Appl. Therm. Eng. 98 (2016) 1165–1173, https://doi.org/10.1016/ [30] X. Liu, O. El Fakir, Y. Zheng, M.M. Gharbi, L. Wang, Effect of tool coatings on the
j.applthermaleng.2015.12.142. interfacial heat transfer coefficient in hot stamping of aluminium alloys under
[7] G.N. Chu, G. Chen, Y.L. Lin, S.J. Yuan, Tube hydro-forging − a method to manu­ variable contact pressure conditions, Int. J. Heat Mass Transf. 137 (2019) 74–83,
facture hollow component with varied cross-section perimeters, J. Mater. Process. https://doi.org/10.1016/j.ijheatmasstransfer.2019.03.087.
Technol. 265 (2019) 150–157, https://doi.org/10.1016/j.jmatprotec.2017.11.007. [31] X. Liu, O. El Fakir, Z. Cai, M.M. Gharbi, B. Dalkaya, Development of an interfacial
[8] S. Toros, F. Ozturk, I. Kacar, Review of warm forming of aluminum-magnesium heat transfer coefficient model for the hot and warm aluminium stamping processes
alloys, J. Mater. Process. Technol. 207 (2008) 1–12, https://doi.org/10.1016/j. under different initial blank temperature conditions, J. Mater. Process. Technol.
jmatprotec.2008.03.057. 273 (2019) 116245, , https://doi.org/10.1016/j.jmatprotec.2019.05.026.
[9] H. Karbasian, A.E. Tekkaya, A review on hot stamping, J. Mater. Process. Technol. [32] T.N. Çetinkale, M. Fishenden, Thermal conductance of metal surfaces in contact,
210 (2010) 2103–2118, https://doi.org/10.1016/j.jmatprotec.2010.07.019. Proc. Int. Conf. Heat Transf. Inst. Mech. Eng. (1951) 271–275.
[10] W. Ma, B. Wang, L. Yang, X. Tang, W. Xiao, J. Zhou, Influence of solution heat [33] A.C. Rapier, T.M. Jones, J.E. McIintosh, The thermal conductance of uranium di­
treatment on mechanical response and fracture behaviour of aluminium alloy oxide/stainless steel interfaces, Int. J. Heat Mass Transf. 6 (1963) 397–416, https://
sheets: An experimental study, Mater. Des. 88 (2015) 1119–1126, https://doi.org/ doi.org/10.1016/0017-9310(63)90101-7.
10.1016/j.matdes.2015.09.044. [34] M.G. Cooper, B.B. Mikic, M.M. Yovanovich, Thermal contact conductance, Int. J.
[11] O. El Fakir, L. Wang, D. Balint, J.P. Dear, J. Lin, T.A. Dean, Experimental and nu­ Heat Mass Transf. 153 (1969) 317–323, https://doi.org/10.1016/j.cbpa.2009.03.
merical studies of the solution heat treatment, forming, and in-die quenching (HFQ) 006.
process on AA5754, Int. J. Mach. Tools Manuf. 87 (2014) 39–48, https://doi.org/ [35] W.R.D. Wilson, S.R. Schmid, J. Liu, Advanced simulations for hot forging: Heat
10.1016/j.ijmachtools.2014.07.008. transfer model for use with the finite element method, J. Mater. Process. Technol.
[12] T. Maeno, K. Mori, R. Yachi, Hot stamping of high-strength aluminium alloy aircraft 155–156 (2004) 1912–1917, https://doi.org/10.1016/j.jmatprotec.2004.04.399.
parts using quick heating, CIRP Ann. - Manuf. Technol. 66 (2017) 269–272, https:// [36] V.W. Antonetti, M.M. Yovanovich, Using metallic coatings to enhance thermal
doi.org/10.1016/j.cirp.2017.04.117. contact conductance of electronic packages, Heat Transf. Eng. 9 (1988) 85–92,
[13] Q. Zhang, X. Luan, S. Dhawan, D.J. Politis, Z. Cai, L. Wang, Investigating the quench https://doi.org/10.1080/01457638808939674.
sensitivity of high strength AA6082 aluminium alloy during the new FAST forming [37] A. Salazar, On thermal diffusivity, Eur. J. Phys. 24 (2003) 351–358, https://doi.
process, IOP Conf. Ser. Mater. Sci. Eng. 418 (2018) 012028, , https://doi.org/10. org/10.1088/0143-0807/24/4/353.
1088/1757-899X/418/1/012028. [38] M.S. Mohamed, A.D. Foster, J. Lin, D.S. Balint, T.A. Dean, Investigation of de­
[14] L. Wang, Y. Sun, K. Ji, X. Luan, O. El Fakir, Z. Cai, X. Liu, Fast warm stamping of formation and failure features in hot stamping of AA6082: Experimentation and
ultra-high strength steel sheets, Intellectual Property No.: 1713741.5, 2017. modelling, Int. J. Mach. Tools Manuf. 53 (2012) 27–38, https://doi.org/10.1016/j.
[15] Z. Cai, P. Batthyá, S. Dhawan, Q. Zhang, Y. Sun, X. Luan, Study of springback for ijmachtools.2011.07.005.
high strength aluminium alloys under hot stamping, in: 4th Int. Conf. Adv. High [39] H.R. Shercliff, M.F. Ashby, A process model for age hardening of aluminium alloys-
Strength Steel Press Hardening, 2019. doi: 10.1142/9789813277984_0019. I. The model, Acta Metall. Mater. 38 (1990) 1789–1802, https://doi.org/10.1016/
[16] G. Palumbo, L. Tricarico, Numerical and experimental investigations on the Warm 0956-7151(90)90291-N.
Deep Drawing process of circular aluminum alloy specimens, J. Mater. Process. [40] J.P. Shlykov, E.A. Ganin, S.N. Tsarevskiy, Thermal contact resistance, M. Energy
Technol. 184 (2007) 115–123, https://doi.org/10.1016/j.jmatprotec.2006.11.024. (1977) 328.
[17] K. Zheng, Y. Dong, D. Zheng, J. Lin, T.A. Dean, An experimental investigation on [41] B. Buchner, M. Buchner, B. Buchmayr, Determination of the real contact area for
the deformation and post-formed strength of heat-treatable aluminium alloys using numerical simulation, Tribol. Int. 42 (2009) 897–901, https://doi.org/10.1016/j.
different elevated temperature forming processes, J. Mater. Process. Technol. 268 triboint.2008.12.009.
(2019) 87–96, https://doi.org/10.1016/j.jmatprotec.2018.11.042. [42] M.V. Murashov, S.D. Panin, Numerical modelling of contact heat transfer problem
[18] X. Fan, Z. He, S. Yuan, K. Zheng, Experimental investigation on hot forming- with work hardened rough surfaces, Int. J. Heat Mass Transf. 90 (2015) 72–80,
quenching integrated process of 6A02 aluminum alloy sheet, Mater. Sci. Eng. A. 573 https://doi.org/10.1016/j.ijheatmasstransfer.2015.06.024.
(2013) 154–160, https://doi.org/10.1016/j.msea.2013.02.058. [43] H. Li, L. He, G. Zhao, L. Zhang, Constitutive relationships of hot stamping boron
[19] X. Fan, Z. He, S. Yuan, P. Lin, Investigation on strengthening of 6A02 aluminum steel B1500HS based on the modified Arrhenius and Johnson-Cook model, Mater.
alloy sheet in hot forming-quenching integrated process with warm forming-dies, Sci. Eng. A. 580 (2013) 330–348, https://doi.org/10.1016/j.msea.2013.05.023.
Mater. Sci. Eng. A. 587 (2013) 221–227, https://doi.org/10.1016/j.msea.2013.08. [44] F. Sun, P. Zhang, H. Wang, J. Shen, L. Nie, Experimental study of thermal contact
059. resistance between aluminium alloy and ADP crystal under vacuum environment,
[20] A. Keci, N.R. Harrison, S.G. Luckey, Experimental evaluation of the quench rate of Appl. Therm. Eng. 155 (2019) 563–574, https://doi.org/10.1016/j.
AA7075, SAE Int. J. Mater. Manuf. (2014), https://doi.org/10.4271/2014-01-0984. applthermaleng.2019.04.027.
[21] Y. Zhang, M. Weyland, B. Milkereit, M. Reich, P.A. Rometsch, Precipitation of a new [45] E.J.F.R. Caron, K.J. Daun, M.A. Wells, Experimental heat transfer coefficient mea­
platelet phase during the quenching of an Al-Zn-Mg-Cu alloy, Sci. Rep. (2016) 1–9, surements during hot forming die quenching of boron steel at high temperatures,
https://doi.org/10.1038/srep23109. Int. J. Heat Mass Transf. 71 (2014) 396–404, https://doi.org/10.1016/j.
[22] J. Liu, A. Wang, Y. Zheng, X. Liu, J. Gandra, K. Beamish, Hot stamping of AA6082 ijheatmasstransfer.2013.12.039.
tailor welded blanks for automotive applications, Proc. Eng. 207 (2017) 729–734, [46] Smart Forming, Smart forming technology platform, 2019. http://techtiqdemo.co.
https://doi.org/10.1016/j.proeng.2017.10.820. uk/demonodes/smart-forming/ (accessed May 10, 2019).
[23] B. Milkereit, O. Kessler, C. Schick, Recording of continuous cooling precipitation [47] W. Xiao, B. Wang, K. Zheng, J. Zhou, J. Lin, A study of interfacial heat transfer and
diagrams of aluminium alloys, Thermochim. Acta. 492 (2009) 73–78, https://doi. its effect on quenching when hot stamping AA7075, Arch. Civ. Mech. Eng. 18
org/10.1016/j.tca.2009.01.027. (2018) 723–730, https://doi.org/10.1016/j.acme.2017.12.001.
[24] B.C. Shang, Z.M. Yin, G. Wang, B. Liu, Z.Q. Huang, Investigation of quench sensi­ [48] H. Gao, O. El Fakir, L. Wang, D.J. Politis, Z. Li, Forming limit prediction for hot
tivity and transformation kinetics during isothermal treatment in 6082 aluminum stamping processes featuring non-isothermal and complex loading conditions, Int.
alloy, Mater. Des. 32 (2011) 3818–3822, https://doi.org/10.1016/j.matdes.2011. J. Mech. Sci. 131–132 (2017) 792–810, https://doi.org/10.1016/j.ijmecsci.2017.
03.016. 07.043.
[25] Q. Bai, J. Lin, L. Zhan, T.A. Dean, D.S. Balint, Z. Zhang, An efficient closed-form [49] H. Gao, D.J. Politis, X. Luan, K. Ji, Q. Zhang, Y. Zheng, M. Gharbi, L. Wang, Forming
method for determining interfacial heat transfer coefficient in metal forming, Int. J. limit prediction for AA7075 alloys under hot stamping conditions, J. Phys. Conf.
Mach. Tools Manuf. 56 (2012) 102–110, https://doi.org/10.1016/j.ijmachtools. Ser. 896 (2017), https://doi.org/10.1088/1742-6596/896/1/012089.
2011.12.005. [50] B. Milkereit, N. Wanderka, C. Schick, O. Kessler, Continuous cooling precipitation
[26] V. Norouzifard, M. Hamedi, Experimental determination of the tool–chip thermal diagrams of Al-Mg-Si alloys, Mater. Sci. Eng. A. 550 (2012) 87–96, https://doi.org/
contact conductance in machining process, Int. J. Mach. Tools Manuf. 84 (2014) 10.1016/j.msea.2012.04.033.
45–57, https://doi.org/10.1016/j.ijmachtools.2014.04.003. [51] B. Milkereit, M. Österreich, P. Schuster, G. Kirov, E. Mukeli, O. Kessler, Dissolution
[27] L. Ying, T. Gao, M. Dai, P. Hu, Investigation of interfacial heat transfer mechanism and precipitation behavior for hot forming of 7021 and 7075 aluminum alloys,
for 7075–T6 aluminum alloy in HFQ hot forming process, Appl. Therm. Eng. 118 Metals (Basel). 8 (2018) 531, https://doi.org/10.3390/met8070531.
(2017) 266–282, https://doi.org/10.1016/j.applthermaleng.2017.02.107. [52] A. Wang, O. El Fakir, J. Liu, Q. Zhang, Y. Zheng, L. Wang, Multi-objective finite
[28] X. Liu, O. El Fakir, L. Meng, X. Sun, X. Li, L. Wang, Effects of lubricant on the IHTC element simulations of a sheet metal-forming process via a cloud-based platform,
during the hot stamping of AA6082 aluminium alloy: experimental and modelling Int. J. Adv. Manuf. Technol. 100 (2018) 2753–2765, https://doi.org/10.1007/
studies, J. Mater. Process. Technol. 255 (2018) 175–183, https://doi.org/10.1016/ s00170-018-2877-x.

12