Critical review of automotive hot-stamped sheet steel from an industrial perspective

Blanking

As-delivered hot-rolled (gauge typically >2 mm) or cold-rolled/coated (gauge typically ≤2 mm) sheet steel (width typically 1300–1500 mm) with microstructure of proeutectoid ferrite and pearlite, proof strength <500 MPa, ultimate tensile strength <600 MPa and total elongation >20% (with 22MnB5 steel), is cut to blanks of suitable geometry. Blanking is a cold shearing operation in which the blank is cut and removed from the sheet by applying a sufficient shearing force at ambient temperature. The blanking press consists of a ram, punch and die. The sheet steel is continuously fed into the blanking press with die below and punch above the sheet. The punch and die exhibit geometry of the desired blank, plus an additional clearance margin of 10–20% sheet gauge. Both punch and die are typically fabricated from hardened tool steels and their geometries are customised depending on the desired hot-stamped part geometry. Figure 12(a) [28] presents a blank to be hot stamped into a tunnel reinforcement. To minimise scrap, sheet width is typically specified to equal the longest dimension of the required blank geometry. Thus, sheet supplied by the steel manufacturer may require slitting along the longitudinal centreline before blanking. To increase efficiency and feed multiple hot-stamping tools, multiple sheets may be stacked and blanked simultaneously.

OPEN IN VIEWER

Figure 12

Stages of the direct hot stamping process: (a) blank, (b) blank feeding into furnace, (c) blank transfer from furnace to hot-forming press, (d) punch descent and hot stamping of blank (e) hot-stamped part following ejection from hot-forming press, (f) laser trimming of hot-stamped part (adapted from [28]).

Pre-forming (indirect method)

Applicable only to indirect hot stamping, the blank is cold formed (pre-formed) before furnace heating. Pre-forming is typically conducted to within 95% of the final part geometry. The press consists of a ram, punch, die and blank holder. The ram descends the punch to meet its mating die, with the combined punch and die geometry equal to the desired pre-formed part geometry (the punch is a protrusion that embosses the blank, while the die is a cavity that accepts the blank). Highly intricate part geometries may require a more complex arrangement of punch, die and counter punch; with the counter punch a protrusion rising from the die cavity. With the blank constrained in the blank holder and between punch and die, the blank is rapidly stamped during <2 s into the pre-formed part geometry.

Furnace heating

The blank (via direct hot stamping) or pre-formed part (via indirect hot stamping) is austenised by radiant and convective heating in a continuous electric- or gas-fuelled roller hearth furnace, as illustrated in Figure 12(b) [28]. The mean heating rate is typically 10–12°C/s to a target soak temperature of typically 900–950°C and where the Ac3 temperature of the 22MnB5 steel is typically 800–830°C. The soak temperature is maintained for a soak time of typically 3–8 minutes (depending on gauge and coating) to ensure complete and homogenous austenisation, sufficient austenitic grain growth for hardenability and substrate-coating inter-layer alloying (depending on coating).

The roller hearth furnace includes a multiplicity of rollers extending across typically 30–40 m in length. Rotation of the rollers transports the blank/pre-formed part through the heating section (with these rollers rotating at a relatively high speed) and then into the soaking section (with these rollers rotating at a relatively low speed or even paused so as to achieve the soak time). Total furnace time including heating and soaking is typically <10 minutes. Most furnaces operate in an inert gas atmosphere of nitrogen, hydrogen and/or argon to avoid oxidation and decarburisation of the sheet steel (most relevant to uncoated sheet steel).

Transfer

The blank (via direct hot stamping) or pre-formed part (via indirect hot stamping) at typically 900–950°C is transferred by a robotic system from the furnace to the hot-forming tool in typically <10 s, as illustrated in Figure 12(c) [28]. During the transfer time exposed to atmospheric air, the temperature of the blank/pre-formed part typically decreases by 100–200°C depending on gauge, surface area and furnace temperature. Exposure to atmospheric air at elevated temperature permits oxidation and decarburisation of uncoated sheet steel. Thus, coated sheet steels are favourable even with the utilisation of a furnace operating in an inert gas atmosphere.

It is essential that the transfer time is as short as possible so as to minimise surface oxidation and decarburisation (of uncoated sheet steel) and most importantly so that temperature does not decrease below the Ar3 to ensure that the microstructure remains entirely austenitic (stable or metastable) at the commencement of the hot-stamping and die quenching stages. If the Ar3 temperature is reached before quenching begins, proeutectoid ferrite will form and thus, the microstructure will only partially harden to martensite. The presence of proeutectoid ferrite (as opposed to exclusive austenite) during the hot-stamping stage will also compromise hot formability.

Hot stamping

In common with the cold-forming press, the hot-forming press consists of a ram, punch, die and blank holder. With the blank (via direct hot stamping) or pre-formed part (via indirect hot stamping) constrained in the blank holder and between punch and die, as illustrated in Figure 12(d) [28], the blank/pre-formed part is rapidly stamped during <2 s into the final part geometry.

The temperature of the blank/pre-formed part at the beginning of the hot-stamping stage is dictated by the furnace temperature and degree of cooling during transfer. Typical temperature at the beginning of the hot-stamping stage is 800–850°C, which decreases to typically 600–650°C by the end of the hot-stamping stage. Given that the blank/pre-formed part is maintained in the (metastable) austenite phase and in a narrow temperature range throughout hot forming, proof strength is maintained at typically just 150–300 MPa and with a total elongation of 40–60% (with 22MnB5 steel). Homogenous austenisation which eliminates texture developed during rolling of the sheet steel ensures plastic isotropy.

Die quenching

As the hot blank/pre-formed part systematically contacts the punch and die during hot stamping, quenching effectively begins immediately and intensifies as more of the blank/pre-formed part contacts the punch and die during punch descent and as tool pressure increases. However, the true die quenching stage begins when the hot-stamping stage is complete (defined by the punch reaching the bottom of its decent so as to mate with the die) and with every surface of the part (at typically 600–650°C) in contact with either the punch or the die. Different from the cold-forming tool, the punch and/or die are water cooled, where water circulates the tooling via an intricacy of internal cooling channels. Tool pressure is maintained for typically 8–12 s so as to maintain quenching down to 200°C with a mean cooling rate of more than 30°C/s. Thus, the formed part is rapidly quenched through the Ms–Mf temperature range and hardened to a martensitic microstructure, where the Ms and Mf temperatures of 22MnB5 steel are typically 410 and 230°C, respectively, and the critical cooling rate is typically 25°C/s. Quenching is achieved by conductive heat transfer from the part to the water-cooled tooling. Cooling rates over the critical metastable austenitic temperature range of ∼600–400°C are typically in excess of 100°C/s.

Following quenching, the hot-stamped part is typically entirely martensitic (with the exception of a minute volume fraction, typically <5%, of inter-lath retained austenite). The hot-stamped part is ejected from the tool and air cooled at ambient temperature, as illustrated in Figure 12(e) [28]. Tensile properties in the hot-stamped part (with 22MnB5 steel) are typically proof strength 1000–1250 MPa, ultimate tensile strength 1400–1700 MPa and total elongation 4–8%.

Post-processing

Post-processing includes shot blasting to remove the oxide scale, both for ease of spot welding and for paint adhesion. Machining such as cutting, trimming and/or piercing may be in order to refine geometry, such as removing flanges and the addition of welding points, as illustrated in Figure 12(f) [28]. The ultra-high-strength martensitic microstructure limits mechanical machining and thus, laser machining is typically conducted. Via indirect hot stamping, machining to refine geometry may be conducted between the pre-forming and furnace heating stages when the pre-formed part exhibits the relatively soft ferritic–pearlitic microstructure.

Heating rate

Figure 24 illustrates the influence of heating rate on the final part hardness [38]. Samples of 22MnB5 steel with a gauge of 1.5 mm and a chemical composition (wt-% Fe balanced) of 0.25C, 0.02Si, 2.50Mn, 0.04Al, 0.015Ti and 0.002B were heated to various soak temperatures with heating rates of 10 and 200°C/s, followed by immediate water quenching to ambient temperature. Up to the critical soak temperature of ∼800°C (corresponding to the Ac3 temperature), a higher heating rate gave rise to a lower hardness following quenching due to greater superheating raising the Ac3 temperature and rendering less austenite to harden to martensite on quenching, as illustrated in Figure 25. Above the critical soak temperature of ∼800°C, a higher heating rate gave rise to a higher hardness following quenching due to greater superheating raising the austenite recrystallisation and grain growth temperatures and giving rise to a finer austenite grain size (and in turn a finer martensite packet size following quenching). Cooperative results to the above have been presented, where samples of 22MnB5 steel with a gauge of 1.6 mm and a chemical composition (wt-% Fe balanced) of 0.25C, 0.40Si, 1.40Mn, 0.05Ti and 0.005B were heated to various soak temperatures with heating rates of 1, 5 and 10°C/s, followed by soaking for 15 minutes and quenching to ambient temperature [39]. With a higher heating rate, a shorter soak time, but a higher soak temperature, was required to attain a given volume fraction of austenite. This can be interpreted to be due to the effect of superheating.

OPEN IN VIEWER

Figure 24

Influence of heating rate on hardness (adapted from [38]).

OPEN IN VIEWER

Figure 25

Influence of heating rate on martensite percentage (adapted from [38]).

Soak temperature and time

Soak temperature and time are closely related. Soak temperature is also closely related to tailored heating. Here, soak temperature is considered in the absence of tailored heating. Figure 26 illustrates the time–temperature–transformation diagram on heating for 22MnB5 steel with a chemical composition (wt-% Fe balanced) of 0.22C, 1.20Mn, 0.025Ti and 0.003B, where samples were heated to various soak temperatures at 10°C/s, soaked for various soak times and rapidly quenched to ambient temperature with subsequent microstructural analysis estimating the percentage of austenisation under the various soak temperatures and time conditions [40]. Given the soak temperature of 1000, 900, 850, 825 and 800°C, the soak time of 1, 10, 100, 1000 and 10,000 s was required, respectively, for complete transformation to austenite.

OPEN IN VIEWER

Figure 26

Time–temperature–transformation diagram on heating for 22MnB5 steel (adapted from [40]).

Once complete austenisation has been achieved, it is then necessary to consider the consequences of a higher soak temperature and/or a longer time in the austenitic phase field. A higher soak temperature or a longer soak time gives rise to greater austenitic grain growth, smaller austenite grain boundary surface area, fewer heterogeneous nucleation sites for ferrite formation during slow cooling (transfer stage) and greater hardenability to ensure a complete transformation to martensite during rapid cooling (die quenching stage). Insufficient austenitic grain growth resulting from low soak temperature and/or short soak time gives rise to ferrite (and/or bainite) in the final microstructure, and in turn, lower tensile strength, as illustrated for 22MnB5 steel of 1.5 mm gauge by Figure 27 [41]. Excessive austenitic grain growth (over and above that required to enable a complete transformation to martensite) gives rise to a coarse martensitic packet and lath structure and in turn, lower tensile strength due to loss of the grain boundary strengthening effect, as illustrated for 22MnB5 steel of 1.5 mm gauge by Figure 28 [42]. The optimal soak temperature and time combination required for complete martensite formation and maximum tensile strength in the final part depends on steel chemistry, gauge and transfer time. Chemistries richer in hardenability raising alloying elements such as carbon are optimised with lower soak temperatures and/or shorter times, as illustrated by comparing 22MnB5 steel (0.22 wt-% C) and 38MnB5 steel (0.38 wt-% C) of Figure 27 [41]. Lighter gauge and shorter transfer time are optimised with a lower soak temperature and/or a shorter time.

OPEN IN VIEWER

Figure 27

Influence of soak temperature and time on ultimate tensile strength of 22MnB5 and 38MnB5 steel (adapted from [41]).

OPEN IN VIEWER

Figure 28

Influence of soak temperature and time on austenite grain size of 22MnB5 steel (adapted from [42]).

Deformation rate

Deformation rate, or strain rate, is not directly controllable but is dependent on: (1) punch decent rate; (2) material specific true stress–true strain (flow) behaviour; (3) absolute strain; (4) gauge and (5) temperature. Strain rate can vary across a single blank, giving rise to numerous local strain rates rather than one universal strain rate. Figure 29 illustrates the influence of strain rate on a tensile strength of 22MnB5 steel with a gauge of 1.5 mm and a chemical composition (wt-% Fe balanced) of 0.23C, 0.22Si, 1.18Mn, 0.04Ti and 0.002B, where samples were heated at 10°C/s to a soak temperature of 950°C, held for a soak time of 180 s, rapidly cooled at 80°C/s to deformation temperatures of 500, 650 and 800°C and isothermally tensile tested across strain rates of 0.01, 0.1 and 1 s⁻¹ while in the stable or metastable austenite phase [29]. There was a consistent increase in austenite tensile strength (and loss of formability) with increasing strain rate due to dynamic strain aging. Similar results to the above have been presented elsewhere [43]. However, in the latter research, a more advanced analysis method was developed, in which correlative correction factors were established to account for necking of the deformed region in the tensile test (and to accurately measure strain at the deformed region) and to account for the temperature rise at the deformed region due to the conversion of mechanical energy to thermal energy (and to accurately measure stress at the deformed region). Note that the true stress–true strain curves presented in Figure 29 have been truncated at the commencement of necking and, therefore, do not provide a genuine representation of maximum true stress or true strain. This truncation method, which is recommended in the standardised tensile testing procedures [44], was necessary in order to eliminate the inaccuracies of true stress and true strain measurement that exist beyond the point of necking. The correction factors thus seek to eliminate these inaccuracies. The correction factors gave rise to a maximum increase in a true stress of 50 MPa and a maximum increase in a true strain of 0.1.

OPEN IN VIEWER

Figure 29

Influence of strain rate on tensile strength of 22MnB5 steel (adapted from [29]).

Deformation temperature

Deformation temperature can be controlled by soak temperature, transfer time, residual tool temperature and deformation rate. Deformation temperature is closely related to tailored heating. Here, deformation temperature is considered in the absence of tailored heating. Figure 30 illustrates the influence of deformation temperature on compressive strength of 22MnB5 steel with a gauge of 1.5 mm and a chemical composition (wt-% Fe balanced) of 0.23C, 0.27Si, 1.20Mn, 0.04Ti and 0.004B, where samples were heated at 10°C/s to a soak temperature of 950°C, held for a soak time of 180 s, rapidly cooled at 80°C/s to deformation temperatures of 550–900°C and isothermally compression tested with a strain rate of 1 s⁻¹[45]. There was a consistent decrease in compressive strength (improved formability) with increasing deformation temperature due to a greater rate of dynamic recrystallisation. Moreover, there was a significant increase in compressive strength (loss of formability) from 650 to 600°C due to austenite to ferrite–pearlite transformations. Similar results have been presented previously under tensile conditions, as demonstrated in Figure 29 [29]. Comparing Figure 29 to Figure 30, for the range of comparable deformation temperatures (500, 650 and 800°C) and constant strain rate of 1 s⁻¹, compressive strength was consistently in the order of 50–75 MPa higher than tensile strength (chemical composition between each investigation was very similar). Hot tensile testing of 22MnB5 steel has been a substantially publicised (and exhausted) area of research over the past 10 years, with comparable results to the above presented by Merklein and Lechler [46], Chang et al. [47] and Abspoel et al. [48] to cite just a few examples.

OPEN IN VIEWER

Figure 30

Influence of deformation temperature on compressive strength of 22MnB5 steel (adapted from [45]).

Residual tool temperature

Residual tool temperature influences deformation temperature and most of all, cooling rate. Residual tool temperature is closely related to tailored quenching. Here, residual tool temperature is considered in the absence of tailored quenching. Figure 31(a) illustrates the evolution of residual tool temperature over 20 consecutive hot-stamping cycles (strokes), where samples of 22MnB5 steel with a gauge of 1.5 mm and a chemical composition (wt-% Fe balanced) of 0.22C, 0.15Si, 1.20Mn, 0.03Ti and 0.003B were heated with a mean rate of 7°C/s to a soak temperature of 925°C, held for a soak time of 360 s, transferred to forming tool in 8 s and hot-stamped with tool pressure applied for 9 s [40]. Part and punch temperatures were measured immediately at part ejection from the tool, with temperatures measured at three select points of the part and corresponding points of the punch (Figure 31(b)). Part temperature increased locally by a maximum of 50°C, with the maximum local temperature reaching 180°C. Punch temperature increased locally by a maximum of 100°C, with the maximum local temperature reaching 140°C. Temperature increase was highest at the point corresponding to the upper section of the part (Part-t1) and lowest at the point corresponding to the lower section of the part (Part-t3) due to different local geometry, mass conditions and in turn, heat transfer rates. The upper section of the part (Part-t1) exhibited a narrow geometry and thus, a relatively low mass where heat is transferred to the tool more rapidly compared to the lower section. Punch temperature at the point corresponding to the narrow upper section of the part (Punch-t1) increased more rapidly compared to the point corresponding to the wider lower section (Part-t3). A more rapid temperature increase in the punch (Punch-t1) then resulted in a lower cooling rate and higher temperature in the part (Part-t1) over successive strokes. Figure 32 illustrates the influence of residual tool temperature on the tensile strength of the parts [40]. Lower tensile strength was generally observed with increasing part ejection temperature (and residual tool temperature), attributed to a lower cooling rate and a greater auto-tempering of martensite.

OPEN IN VIEWER

Figure 31

(a) Evolution of residual tool temperature over 20 consecutive hot stamping strokes (adapted from [40]). (b) Temperature measurement points of part (adapted from [40]).

OPEN IN VIEWER

Figure 32

Influence of residual tool temperature on tensile strength of 22MnB5 steel (adapted from [40]).

Die quench rate

It is essential that the die quench rate is in excess of the critical cooling rate in order to achieve a maximum tensile strength. Figure 33 illustrates the influence of cooling rate on phase transformations in 22MnB5 steel with a gauge of 1.5 mm and a chemical composition (wt-% Fe balanced) of 0.23C, 0.21Si, 1.23Mn, 0.04Ti and 0.004%B [49]. Samples were heated with a mean rate of 7°C/s at a soak temperature of 900°C, held for a soak time of 300 s to ensure complete austenisation and cooled at ambient temperature with various mediums of furnace cooling, natural air cooling, compressed air cooling and die quenching (hot stamping); and each with associated mean cooling rates (between 800 and 400°C) of 0.26, 5.17, 48.00 and 341.51°C/s. Phase transformations can be seen from inflections in the cooling curves due to the release of latent heat of phase transformation.

OPEN IN VIEWER

Figure 33

Influence of cooling rate on phase transformations in 22MnB5 steel [49].

Hot stamping in the above investigation was conducted with a standard cold-forming tool. Additional research was conducted (with 22MnB5 steel) with a hot-forming tool exhibiting internal cooling channels [50]. Hot stamping investigations were conducted with both water and liquid nitrogen purged through the cooling channels, giving rise to residual tool temperatures at the commencement of hot stamping equal to 20 and −50°C, respectively. The water-cooled tool reportedly gave rise to a microstructure of 97% martensite and 3% bainite, while the liquid nitrogen-cooled tool gave rise to a microstructure of 100% martensite (although differentiation between martensite and bainite is ambiguous and subjective). What is objective, however, are mechanical properties, where the liquid nitrogen cooled tool gave rise to a higher proof strength, comparable ultimate tensile strength and lower total elongation. This can be attributed primarily to less auto-tempering of martensite with a higher cooling rate of the liquid nitrogen cooled tool.

Further research saw samples of 22MnB5 steel heated to 900°C, soaked for 300 s, compressed to a strain of 0.4 with a strain rate of 7 s⁻¹ and quenched with cooling rates from 0.01 to 100°C/s [21]. The same experiments were also conducted without hot deformation. The addition of hot deformation raised the critical cooling rate from 30 to 60°C/s. Similar results have been presented for 22MnB5 steel by [51] and for a manganese–chromium gear steel [52]. Hot austenitic deformation increases the austenite grain boundary surface area which in turn, increases the number of ferrite nucleation sites. Deformation of austenite also increases dislocation density, raises free energy of austenite and increases the thermodynamic driving force for transformation to ferrite. It has also been suggested that the ferrite growth rate is accelerated, particularly with higher cooling rates (but below the critical cooling rate), due to the enhancement of atomic diffusion rates along dislocations in austenite, as demonstrated in the manganese–chromium gear steel [53]. However, with lower cooling rates, time is permitted for greater dynamic recrystallisation. Therefore, the formation of ferrite and pearlite is decreased. Thus, it can be concluded that if the cooling rate is below a given threshold dependent on dynamic recrystallisation (irrespective of the critical cooling rate), the effect of hot austenitic deformation on hardenability is negligible.

Part/tool boundary conditions

Given that die quenching takes place predominantly by conductive heat transfer from the part to the surfaces of the forming tool, boundary conditions play an important role in determining cooling rate and microstructural evolution. Hot austenitic deformation sustained at the formed regions of the part can influence microstructural evolution. Figure 34 illustrates the formed part geometry and locations of in situ thermocouple data logging at the part/die interface when hot-stamping 38MnB5 steel with a gauge of 1.5 mm and a chemical composition (wt-% Fe balanced) of 0.38C, 1.20Mn, 0.024Ti and 0.003%B [33]. Temperature at a maximum rate of transformation of austenite to martensite (analogous to the martensite formation temperature) increased from the flat (un-deformed) regions of the part geometry to the radii (deformed regions) of the part geometry. There was a consistent inverse correlation between temperature at a maximum rate of transformation and cooling rate (Figure 35). Thus, hot austenitic deformation and a lower cooling rate raise the martensite formation temperature. Higher volume fractions of retained austenite were measured at the radii of the part geometry and correlated to lower cooling rates (Figure 36), where lower cooling rates can permit more time for carbon partitioning from martensite to austenite simultaneously during martensite formation, enabling more austenite to be stabilised at ambient temperature. Higher volume fractions of retained austenite were correlated to hot austenitic deformation, where dislocations introduced can provide more transformation nucleation sites, a finer martensite lath structure and in turn, more lath boundaries across which carbon partitioning from martensite to austenite can occur. More transformation nucleation sites can also accelerate the transformation to higher temperatures (Figure 35), at which the carbon partitioning rate is increased. Vickers hardness number was significantly higher at the radii of part geometry than at the flat regions of part geometry, which was correlated to a finer martensite lath size originating from hot austenitic deformation giving rise to more transformation nucleation sites (Figure 37). The results similar to the above, marketed by lower cooling rates at the radii (deformed regions) of the part geometry, have been demonstrated for 22MnB5 steel [50]. However, in the case of the 22MnB5 steel, exhibiting lower hardenability than the 38MnB5 steel discussed above, lower cooling rates at the radii of the part geometry gave rise to bainite formation rather than homogenous martensite formation.

OPEN IN VIEWER

Figure 34

Locations of in situ thermocouple data logging at the part/die interface when hot-stamping 38MnB5 steel (adapted from [33]).

OPEN IN VIEWER

Figure 35

Temperature at a maximum rate of transformation against mean cooling rate at different locations of the part/die interface [33]

OPEN IN VIEWER

Figure 36

Retained austenite percentage against mean cooling rate at different locations of part/die interface [33].

OPEN IN VIEWER

Figure 37

Martensite lath size against Vickers hardness number at different locations of part/die interface [33].

The conductive heat transfer coefficient from part to tool when hot-stamping 22MnB5 steel was found to be influenced by residual tool (surface) temperature, part (surface) temperature, part height-to-diameter ratio and most significantly by contact pressure between part and tool, with an approximately linear relationship between contact pressure and conductive heat transfer coefficient [54]. The effect of contact pressure on microstructural evolution and mechanical properties in the final part has been investigated when hot-stamping 22MnB5 steel, where the critical contact pressure for a complete transformation to martensite during die quenching was found to be 0.4 MPa [55]. However, in this research, the 22MnB5 steel exhibited a gauge of 4.0 mm, which is significantly greater than that typically used in automotive BiW applications and, therefore, the critical contact pressure cited is of limited relevance. Further research on 22MnB5 steel has demonstrated the influence of surface topography on the conductive heat transfer coefficient, where both contact pressure and part roughness impact surface topography and with a positive power function correlating contact pressure to the conductive heat transfer coefficient [56].

Optimisation of cooling channel design is a topic of much interest in relation to part/tool boundary conditions. One of the earliest investigations into optimisation of cooling channel design was conducted by Hoffmann et al. [57]. Longitudinal cooling channels have since been developed in order to optimise cooling performance, with most significant improvements resulting in the radii (deformed regions) of the part geometry [58], with these regions being more susceptible to inefficient cooling, as noted above. The same researchers further investigated the cooling performance of various cooling channel designs, including the longitudinal design cited above, with computational flow dynamics numerical analysis [59]. It was demonstrated that turbulent flow of the coolant fluid is the major source of inefficient cooling performance, where the latter can be evaluated by maximum coolant temperature, average coolant temperature, coolant temperature uniformity, coolant flow velocity and coolant flow pressure. With the increase in the coolant flow velocity (rate), maximum coolant temperature and average coolant temperature generally decrease, increasing cooling performance. Thus, cooling performance mainly depends on the cooling channel geometry, aided by geometries that limit turbulence. It is for this reason that the longitudinal cooling channel design exhibited the most efficient cooling performance.

Body in White

The safety cell of the BiW represents the original and greatest application of hot-stamped sheet steel. The safety cell is designed to resist deformation during impact in order to maintain a survival space for vehicle occupants. Crumple zones represent more recent applications of hot-stamped sheet steel. Crumple zones are designed to deform during impact in order to absorb impact energy in a controlled and progressive manner, maximising deceleration time and minimising peak load to vehicle occupants. The further development has been engineering the combination of anti-intrusive crashworthiness and impact-energy absorptive crashworthiness into a single part (tailored properties) to optimise load management. Figure 38 illustrates a schematic representation of the generic BiW and parts to which hot-stamped sheet steel has been applied.

OPEN IN VIEWER

Figure 38

Schematic representation of the Body in White and parts to which hot-stamped sheet steel has been applied (adapted from [14]).

By way of case studies to demonstrate the evolving application of hot-stamped sheet steel to the BiW, Figure 39 compares the BiW of the 2003 Volvo XC90 to that of the 2014 Volvo XC90. The 2003 Volvo XC90 (with hot-stamped sheet steel accounting for 7% of the BiW) was one of the first production passenger vehicles to move from a few select isolated hot-stamped sheet steel parts towards the integrated hot-stamped sheet steel safety cell that is now commonplace. In addition to rear bumper and door reinforcements, novel applications of hot-stamped sheet steel included B-pillar reinforcements and a central roof cross member [60]. The hot-stamped sheet steel was exclusively uncoated 22MnB5 steel exhibiting a martensitic microstructure and typical properties of 1100 MPa proof strength, 1500 MPa ultimate tensile strength and 6% total elongation in the final part.

OPEN IN VIEWER

Figure 39

Evolving application of hot-stamped sheet steel to the Body in White of the 2003 Volvo XC90 (adapted from [60]) and 2014 Volvo XC90 (adapted from [61]).

The 2014 Volvo XC90 became the most hot-stamped sheet steel intensive production passenger vehicle to date with hot-stamped sheet steel accounting for 38% of the BiW [61]. The extensive application of hot-stamped sheet steel gave rise to a direct weight reduction of 22 kg compared to the 2003 XC90 and a weight reduction of 40 kg compared to what the 2014 XC90 would have weighed if hot-stamped sheet steel was not utilised. The hot-stamped sheet steel was aluminised 22MnB5 steel exhibiting a martensitic microstructure and typical properties of 1100 MPa proof strength, 1500 MPa ultimate tensile strength and 6% total elongation in the final part; uncoated 38MnB5 steel exhibiting a martensitic microstructure and typical properties of 1200 MPa proof strength, 1900 MPa ultimate tensile strength and 4% total elongation in the final part; and aluminised carbon–manganese–niobium steel exhibiting a ferritic–martensitic microstructure and typical properties of 350 MPa proof strength, 500 MPa ultimate tensile strength and 25% total elongation in the final part. Process technology applications included tailor-welded blanks in the front and rear side longitudinal members and B-pillar reinforcements, tailor-rolled blanks in the B-pillar reinforcements, tailored quenching in the roof rail reinforcements and patch reinforcements in the A-pillar and C-pillar reinforcements, sill reinforcements and floor cross members.

In common with the original concept of Bela Barényi patented in the 1950s [62], the BiW of the 2014 Volvo XC90 can broadly be divided into two functions: crumple zones and safety cell. From the front of the vehicle considering a front impact (Figure 40), the crumple zone starts with the bumper reinforcement. The bumper reinforcement itself is not designed to deform, but rather to resist deformation so as to distribute impact load across the deformable members, enabling the latter to deform along a defined load path. To optimise this, the bumper reinforcement is a wide and high section. Although the ultra-high-strength properties of hot-stamped martensitic steel are suited to this application, in order to minimise weight given the bumper reinforcement's relatively large geometry, the bumper reinforcement is manufactured from heavy gauge aluminium alloy (note that in many smaller segment vehicles with a smaller bumper reinforcement, this part is manufactured from hot-stamped martensitic steel). Immediately behind and connecting to the bumper reinforcement at each end is a crash box, which attaches to the front side longitudinal member. The crash box is a hollow thin-walled aluminium section designed to undergo progressive bending, folding and collapse initiated by folding triggers and thereby absorb all impact load in a low-speed collision (typically up to 16 km/h – required for insurance reparability). Immediately behind and connecting to the crash box is the front side longitudinal member. The front side longitudinal member is manufactured from a hot-stamped tailor-welded blank, with ferritic–martensitic carbon–manganese–niobium steel constituting the majority of the member (and connected to the crash box) and martensitic 22MnB5 steel constituting the connection to the front bulkhead reinforcement (safety cell). The front side member (which resides in parallel above the front side longitudinal member) is manufactured from cold-formed low-strength ferritic carbon–manganese steel. The front side longitudinal member and front side member exhibit pre-defined bend lines and curvature, which combined with the relatively low yield strength and high ductility of the constituting materials and relatively light gauge, promote progressive axial collapse followed by inward bending, folding and collapse along a pre-defined load path so as to maximise deceleration time, maximise impact-energy absorption and minimise peak load to the safety cell and vehicle occupants. The safety cell then starts with the front bulkhead reinforcement, which is the barrier between the crumple zone and the passenger compartment. Extending from the front bulkhead reinforcement is the A-pillar reinforcement and roof rail reinforcement (to the upper side), sill reinforcement (to the lower side) and tunnel reinforcement (to the lower centre). While the A-pillar reinforcement and roof rail reinforcement are commonly separate parts, in the 2014 Volvo XC90, these two parts are in fact one continuous monolithic hot-stamped part ∼2.5 m long, taking advantage of the part consolidation characteristic of hot-stamped sheet steel. The absence of joints ensures maximum structural strength, stiffness and anti-intrusive crashworthiness (particularly in roll over and side impact). The load management strategies described above are also applicable to the rear of the vehicle in rear impact.

OPEN IN VIEWER

Figure 40

Load management during front impact of the 2014 Volvo XC90 (adapted from [61]).

Figure 41 illustrates a conventional BiW assembly sequence. The safety cell is brought together by combining the numerous sub-assembly elements sequentially. The same is true for each crumple zone at the front and rear of the vehicle. The final stage is fitting of the bolted and hinged outer closures. This assembly sequence is used for all BiW constructions, irrespective of material choice. However, as noted above in the discussion of the consolidated A-pillar and roof rail reinforcements of the 2014 Volvo XC90, an advantage of hot-stamped sheet steel is the ability to consolidate parts. While hot-stamped sheet steel and part consolidation has been applied to the safety cell (the most notable example to date being the ‘door ring’ of the 2014 Acura MDX, in which the A-pillar, roof rail, B-pillar and sill reinforcements were consolidated in one continuous monolithic hot-stamped part [63]), opportunities exist for part consolidation to be applied to crumple zones.

OPEN IN VIEWER

Figure 41

Conventional Body in White assembly sequence.

Chassis and suspension

There are chiefly two types of chassis design distinguished by a different utilisation of the BiW: unibody and body on frame. The vast majority of modern passenger vehicles utilise the unibody which has largely superseded the body on frame since the 1950s (in Europe) and 1980s (in the USA) due to lower manufacturing costs and lightweighting. The unibody integrates the BiW into a single unit which then represents the chassis. Thus, in the unibody, the chassis and BiW are equivalent.

The body on frame was the original chassis design. The chassis is the frame element of ‘body on frame’ and is most commonly a ladder frame. The minority of modern passenger vehicles utilises the body on frame (the exception being in the USA, where the fashion is for a traditional body on frame ladder frame for select sports utility vehicles and pick-up trucks endures). In the ladder frame, two longitudinal beams run the length and periphery of the vehicle, with numerous cross beams welded perpendicularly giving rise to a ladder-like structure (Figure 42). The ladder frame serves as the structure that provides the vehicle with strength and stiffness. All other components, including those of the BiW, are attached to the frame. The body on frame ladder frame remains the preferred chassis design for many commercial and heavy goods vehicles since compared to the unibody, it has increased stiffness (particularly in the longitudinal axis and in turn, increased towing capacity) and increased static load carrying capacity; while heavy gauge sheet steel sections are used for the direct attachment of powertrain and suspension parts (hence sub-frames can be eliminated). Articulated tractor units have separate cabs to enable the entire cab to tilt for servicing and to allow for the cab to be isolated from chassis vibration and noise via springs and dampers. Many rigid chassis trucks are still sold in the form of a semi-finished ‘running chassis’ (as were passenger vehicles in the early twentieth century) so that independent body specialists can build them into tankers, military vehicles, horse boxes, etc. Buses can be built with or without a separate chassis frame depending on the sales territory and end use. The large and heavy gauge sections used in truck and bus chassis construction are not producible by the hot-stamping processes currently available. However, there is a need for higher strength and lightweighting in some of the upper structures (e.g. a bus or coach frame).

OPEN IN VIEWER

Figure 42

Schematic representation of the body on frame ladder frame.

The term ‘chassis’ is still used in modern unibody vehicle manufacture but nowadays refers to sub-frames that are used to pre-assemble the suspension and attach to the BiW, enabling road loads to be transferred to the BiW.

Front suspension systems

The most common front suspension system design, in use in over 90% of passenger vehicles and light commercial vehicles, is the MacPherson strut type (Figure 43), since it is simple, low cost and packages efficiently for both front and rear wheel drive vehicles. High-end vehicles and sports cars tend to use the double wishbone-type suspension. In both suspension designs, the suspension links can be manufactured from stamped sheet steel, forged steel or cast aluminium. The road wheel is connected to the strut or suspension via a bearing housed in a front upright, also known as a steering knuckle. This part is usually of cast iron or cast aluminium. The front sub-frame (also presented in Figure 43) can have a number of functions, such as carrying the steering rack, anti-roll bar and lower engine mounts, while also contributing to the crumple zone in a front impact.

OPEN IN VIEWER

Figure 43

General Motors Astra front suspension (MacPherson strut) assembly mounted on the sub-frame.

Rear suspension systems

Rear suspension system designs are more varied, dependant on whether the vehicle is front or rear wheel drive and on ride/handling expectations. Most designs fall into one of the three categories: twist beam (Figure 44), trailing arm or multi-link.

OPEN IN VIEWER

Figure 44

General Motors Astra rear suspension (twist beam) assembly (with Watts linkage).

The twist beam is so-called because of the transverse twisting element that acts as an anti-roll bar as well as reacting side loads from cornering and kerb strike. This element is therefore often manufactured from UHSS – in many cases, a normalised 22MnB5 boron steel tube that is cold formed into a C-section (Figure 44) before being heat treated and water or oil quenched. Rear suspension systems of this type are usually mounted directly to the BiW and therefore, seldom need a sub-frame.

Chassis and suspension requirements

There are many structural parts in the various suspension systems (besides the springs and shock absorbers) that are manufactured from sheet materials and bear major loads. Indeed, chassis and suspension parts generally require high stiffness, tensile strength and fatigue strength in order to tolerate repeated torsional, vibrational, damping and braking loads. Chassis and suspension parts are generally designed not to deform; hence, there is a particular demand for yield strength and fatigue strength, while ductility and toughness are less important. However, many chassis and suspension parts exhibit intricate geometries, including tight radii and punched holes (Figure 45). Thus, the sheet materials may have a requirement for high edge ductility (hole expansion coefficient) and edge fatigue strength. This is one of the reasons that HSLA, CP and martensitic boron (tube) steels have been successful in chassis and suspension engineering. In common with BiW engineering, the sheet materials require formability (with an increased demand for formability in order to downgauge without compromising forming limits) and high strength (with an increased demand for strength in order to downgauge without compromising structural strength). Again in common with BiW engineering, both downgauging and increased yield strength lead to increased springback. Based on the requirements for formability, high strength and eliminated springback, hot-stamped sheet steel has great potential for application to chassis and suspension engineering. This is evidenced by the relatively recent application of hot-stamped boron steel tubes to twist beams. The boron steel tube is hot formed into the desired geometry of the twist beam (e.g. C-section) and die quenched in a process similar to hot-stamped sheet steel. In hot stamping of sheets, double-sided tool contact of the sheet gauge is maintained, but when hot-stamping tubes, there is just one-sided tool contact. To compensate for missing cooling on the inside of the tube, internal water–air spray cooling has been developed.

OPEN IN VIEWER

Figure 45

Lower control arm of a front suspension system demonstrating the intricate geometry typical of chassis and suspension parts.

The leading reason why hot-stamped sheet steel has not been applied to chassis and suspension engineering is the absence of cathodic corrosion protection. Chassis and suspension parts are exposed to the most corrosive environments of salt water, combined with mechanical degradation from stone chipping which accelerates corrosion. Therefore, corrosion is a major consideration and often the limiting factor when attempting to downgauge chassis and suspension parts (with part stiffness another common limitation). Loss of structural strength (and stiffness) as a result of corrosion becomes more significant as downgauging is practised as a given degree of corrosion represents a greater percentage of the original part gauge.

To successfully downgauge chassis and suspension parts, there are three fundamental requirements: (1) higher (particularly yield and fatigue strength) strength of the sheet material (dependent on chemistry and microstructural evolution); (2) higher strength and stiffness of the part (dependent on part geometry) and (3) cathodic corrosion protection of the part (dependent on sheet material chemistry or galvanic coating). The current and most common manufacturing method for chassis and suspension parts is cold forming followed by coating in multiple layers of organic paint that provides barrier corrosion protection (but not cathodic corrosion protection). Moreover, the organic coating typically starts to deteriorate owing to stone chipping and edge cracking within 1–2 years of service, exposing the steel substrate and leading to corrosion. However, chassis and suspension parts are not put at risk since the parts are sufficiently thick (typically >2.0 mm) to allow for a degree of corrosion without significantly impairing structural strength and stiffness. It is for this reason that most automotive manufacturers specify a minimum sheet steel gauge of 2.0 mm for chassis and suspension applications. Note that galvanised sheet steel is rarely used in chassis and suspension applications, since the Metal Inert Gas (MIG)/Metal Active Gas (MAG) welding processes used during assembly (as opposed to resistance spot welding, commonly used in the BiW assembly, where MIG/MAG welding provides continuous welds for greater joint strength and stiffness) evaporate the zinc coating in large areas. Thus, cathodic corrosion protection in these areas is not provided. It is for this reason that the development of higher strength (>800 MPa tensile strength) cold formable sheet steel products (both uncoated and galvanised) specifically for chassis and suspension applications by certain steel manufacturers, add little value, since the downgauging threshold is still limited to the current 2.0 mm.

Development of technologies to produce (direct) hot-stamped sheet steel parts exhibiting cathodic corrosion protection has great potential for application to chassis and suspension engineering. This would enable the downgauging opportunities provided by hot-stamped sheet steel, which have been exploited in BiW engineering, to be exploited in chassis and suspension engineering.

Impact-energy absorptive sheet steels

Hot-stamped 22MnB5 steel exhibiting a predominantly martensitic microstructure, characterised by ultra high tensile strength of typically 1100 MPa proof strength and 1500 MPa ultimate tensile strength, provides excellent anti-intrusive crashworthiness. However, with limited ductility of typically 6% total elongation, application to impact-energy absorptive parts of the BiW is limited. Thus, the advantages of hot stamping (e.g. formability, downgauging, part consolidation and absence of springback) cannot be exploited in impact-energy absorptive parts. Hence, sheet steel developments have included improving the impact-energy absorptive crashworthiness of hot-stamped steel, broadly defined by lower yield strength, higher work hardenability, higher ductility and higher toughness in the final hot-stamped part. Compared to the mild carbon content (0.15–0.30 wt-%) and martensitic microstructure of 22MnB5 steel, sheet steels targeting improved impact-energy absorptive crashworthiness have generally been characterised by low carbon content (<0.15 wt-%) and multiphase ferrite–martensite, ferrite–bainite or ferrite–pearlite microstructures. The greatest challenge with these multiphase impact-energy absorptive parts has been ensuring a broad hot-stamping process window so that the final part delivers consistent microstructure, mechanical properties and crashworthiness in different regions of the part and with different part geometries. In contrast to the traditional 22MnB5 steel which is by comparison simplistically austenised and quenched to a homogenous martensitic microstructure with relatively low process sensitivity, the multiphase microstructures are highly dependent on inter-critical soak conditions and/or carefully controlled cooling during the transfer stage and/or die quenching stage, which introduces much greater process sensitivity. ThyssenKrupp Steel AG introduced MBW500 to the hot-stamping market in 2007. MBW500 is a low-carbon–manganese–niobium steel with a chemical composition comparable to that of an HSLA (HC340LA) product. MBW500 is designed to be completely austenised and then quench hardened to exhibit a microstructure of ferrite and martensite, with the die quench rate below the critical cooling rate. The partially martensitic microstructure exhibits a proof strength of ∼350 MPa, an ultimate tensile strength of ∼500 MPa and a total elongation of ∼25%. ArcelorMittal introduced Ductibor500 to the hot-stamping market in 2009. Ductibor500 is also a low-carbon–manganese–niobium steel, but with a chemical composition closer to that of a DP600 product. Ductibor500 is also designed to be completely austenised and then quench hardened to exhibit a microstructure of ferrite and martensite, giving rise to a proof strength of ∼400 MPa, an ultimate tensile strength of ∼550 MPa and a total elongation of ∼20%. The higher ductility products such as MBW500 and Ductibor500 may be used in monolithic hot-stamped parts requiring impact-energy absorption, but are mainly intended for use in hot-stamped tailor-welded blanks in combination with 22MnB5.

Many sheet steel developments achieving higher ductility than 22MnB5 in the final part have emerged from academic research. The carbon–manganese–chromium steel with a chemical composition (wt-% Fe balanced) of 0.15C, 1.66Mn and 1.26Cr was developed by Zhu et al. [64]. The critical cooling rate of just −14°C/s gave rise to a martensitic microstructure exhibiting a proof strength of 1100 MPa, an ultimate tensile strength of 1570 MPa and a total elongation of 9.3%. The tensile strength is similar to that of 22MnB5, yet with notably higher total elongation. Moreover, the lower critical cooling rate than 22MnB5 encourages homogenous martensite formation across the part geometry. Similarly, the carbon–manganese–aluminium steel with a chemical composition (wt-% Fe balanced) of 0.40C, 2.02Mn and 2.50Al was developed by Yi et al. [65]. Inter-critical annealing (at typical hot-stamping soak temperatures of 900–950°C, where the aluminium content rises the A3 temperature) and quenching gave rise to a DP ferritic–martensitic microstructure exhibiting a proof strength of 600 MPa, an ultimate tensile strength of 1550 MPa and a total elongation of 8.8%. The ultimate tensile strength is similar to that of 22MnB5, yet with notably lower proof strength, higher proof-to-ultimate tensile strength ratio and marginally higher total elongation.

Medium to high manganese, carbon–manganese–silicon steels have gained much attention for automotive body engineering applications over the past 10–20 years owing to excellent combinations of strength and ductility arising from the stress/strain-induced transformation (of retained austenite to martensite) and Twinning-induced plasticity (TWIP) effects. These steels have more recently gained attention for hot-stamping applications. Criteria based on the Ashby method has been proposed to identify the most suitable sheet metals for hot-stamping applications from the perspective of formability, with results indicating the potential of high stacking fault energy (SFE) austenitic high manganese, carbon–manganese–silicon steels [66]. Subsequently, steels exhibiting 5–12 wt-% Mn were developed for hot stamping, giving rise to a proof strength of up to 915 MPa, an ultimate tensile strength of up to 1800 MPa and more than 10% total elongation in the hot-stamped part [67]. The steel exhibiting 9.57 wt-% Mn was developed for hot stamping, giving rise to a proof strength of up to 791 MPa, an ultimate tensile strength of up to 1488 MPa and a total elongation of up to 25.3% in the hot-stamped part [68]. Steels exhibiting 5–8 wt-% Mn were developed for hot stamping, giving rise to a proof strength of up to 1400 MPa, an ultimate tensile strength of up to 1880 MPa and a total elongation of up to 16% in the hot-stamped part [69]. Steels exhibiting 4–7 wt-% Mn were developed for hot stamping, giving rise to a proof strength of up to 1220 MPa, an ultimate tensile strength of up to 1420 MPa and a total elongation of up to 12% in the hot-stamped part [47]. All of these steels were developed with hot-stamping soak temperatures of 650–850°C, notably lower than the typical soak temperatures of 900–950°C employed with 22MnB5 steel. The lower soak temperatures translate into significant energy savings on the hot-stamping line.

While the above developments from academic research exhibit very impressive tensile properties in the final part, the industrial feasibility is limited due to the relatively rich chemical compositions. Alloy additions of 1.26 wt-% Cr or 2.50 wt-% Al or certainly 4–12 wt-% Mn are unfavourable (even impossible) with standard steelmaking and sheet steel processing practices, due to problems including ladle temperature control during secondary steelmaking, ladle nozzle blockage from calcium aluminate formation during casting, substitutional alloy segregation during casting (which does not homogenise during subsequent heat treatment), slab cracking if permitted to cool (and thus, slabs must be hot rolled within 24 hours of casting, limiting freedom of scheduling), solid solution strengthening increasing rolling loads (and decreasing dimensional matrices and shape control) and hardenability delaying phase transformations on the run-out table during hot rolling (giving rise to excessively hard hotband which increases subsequent cold rolling loads) and during continuous annealing (giving rise to excessively hard as-delivered mechanical properties that are not suitable for blanking ahead of hot stamping). There are also the limitations of coatability and weldability.

From the perspective of industrial feasibility, developments that have focused on existing low-to-mild carbon low-alloy steel chemistries that are already in commercial production and automotive application are more promising. Three carbon–manganese–chromium steels with a chemical composition (wt-% Fe balanced) of 0.14–0.19C, 1.45–1.71Mn and 0.01–0.55Cr were developed by Naderi et al. [50]. Following hot stamping, the three steels gave rise to multiphase microstructures of ferrite, bainite and martensite exhibiting a proof strength of up to 600 MPa, an ultimate tensile strength of up to 910 MPa and a total elongation of up to 9.3%. Three carbon–manganese–chromium steels with a chemical compositions (wt-% Fe balanced) of 0.09–0.14C, 1.70–2.10Mn and 0.55–0.62Cr were developed by Taylor et al. [70]. Following simulated hot stamping, the three steels gave rise to dual-phase microstructures of ferrite and martensite exhibiting a proof strength of up to 474 MPa, an ultimate tensile strength of up to 816 MPa and a total elongation of up to 16.4%.

Anti-intrusive sheet steels

Sheet steel developments have also included improved anti-intrusive crashworthiness of hot-stamped steel, broadly defined by yet higher tensile (particularly yield) strength, but without compromise of ductility, in the final hot-stamped part. Compared to the mild carbon content (0.15–0.30 wt-%) and martensitic microstructure of 22MnB5 steel, sheet steels targeting improved anti-intrusive crashworthiness have generally been characterised by medium carbon content (0.30–0.60 wt-%) and tempered martensitic microstr-uctures. The greatest challenge with these highest strength anti-intrusive parts have been ensuring the richer chemistries are compatible with standard steelmaking and sheet steel processing practices (as described above), ensuring adequate ductility and toughness in the final part, ensuring resistance to hydrogen embrittlement, ensuring compatibility with standard automotive welding practices and ensuring coatability.

Sumitomo Metal Industries (which merged with Nippon Steel Corporation in 2012 to form Nippon Steel & Sumitomo Metal Corporation) introduced PHS1800 to the hot-stamping market in 2008. PHS1800 was developed with Aishin Takaoka Corporation (tier 1 part manufacturer) and Mazda Motor Corporation for the 2013 Mazda CX-5. PHS1800 was the first commercially utilised hot-stamped steel to approach 1800 MPa ultimate tensile strength and thus, make a significant improvement on 22MnB5 steel. PHS1800 is a medium carbon–manganese–niobium steel, exhibiting a chemical composition (wt-% Fe balanced) of approximately 0.30C, 1.70Mn and 0.08Nb. PHS1800 is designed to be completely austenised and then quench hardened to a martensitic microstructure exhibiting a proof strength of ∼1350 MPa, an ultimate tensile strength of ∼1750 MPa and a total elongation of ∼4%. The niobium addition (formation of niobium carbides) encourages austenitic grain size refinement, increasing martensitic strength (and toughness) via a grain boundary strengthening effect. Niobium carbides also potentially minimise hydrogen embrittlement susceptibility by providing hydrogen traps. Boron is eliminated as associated hardenability is not required given the raised carbon and manganese additions (primarily present for solid solution strengthening). ThyssenKrupp Steel AG introduced MBW1900 to the hot-stamping market in 2016. MBW1900 is a medium carbon–manganese–boron steel, exhibiting a chemical composition (wt-% Fe balanced) of approximately 0.38C, 1.40Mn and 0.003B (38MnB5 steel). MBW1900 is also designed to be completely austenised and then quench hardened to a martensitic microstructure exhibiting a proof strength of ∼1200 MPa, an ultimate tensile strength of ∼1900 MPa and a total elongation of ∼4%.

From academic research, a 37MnB5 steel exhibiting a proof strength in excess of 1600 MPa, an ultimate tensile strength in excess of 1800 MPa and a total elongation in excess of 9.0% following hot stamping and post-quench tempering has been developed [71]. Post-quench tempering is not desirable due to the additional processing cost presented to the tier 1 part manufacturer. Although tempering could be incorporated within the forming tool (as in tailored quenching), this increases cycle time. 28MnB5 steel (with 3 wt-% Mn and 0.1 wt-% Ti) exhibiting an ultimate tensile strength in excess of 2200 MPa following hot stamping has been developed [72]; However, as mentioned previously, the relatively high manganese and titanium additions are not desirable from the perspective of industrial feasibility.

From the perspective of industrial feasibility, developments that have focused on the existing medium carbon–manganese–boron steels that are already in commercial production, or with minor microalloy additions, are more promising. Three mild carbon–manganese–boron (25MnMoB5, 25MnVB5 and 25MnNiB5) steels with chemical compositions (wt-% Fe balanced) of approximately 0.25C, 1.20Mn, 0.003B and independent additions of 0.20Mo, 0.01V and 0.50Ni; and a medium carbon–manganese–boron (38MnB5) steel with a chemical composition (wt-% Fe balanced) of 0.38C, 1.20Mn and 0.005B, were developed [73]. Following hot stamping, the three microalloyed boron steels gave rise to martensitic microstructures exhibiting a proof strength of 1179–1270 MPa, an ultimate tensile strength of 1664–1684 MPa and a total elongation of 9.7–11.6%, while 38MnB5 gave rise to a martensitic microstructure exhibiting a proof strength of 1419 MPa, ultimate tensile strength of 2066 MPa and a total elongation of 6.2%. Dynamic (high strain rate) tensile properties of 25MnVB5 and 38MnB5, in addition to the benchmark 22MnB5, were subsequently evaluated [74]. While 22MnB5 and 25MnVB5 generally exhibited increased tensile strength with increased strain rate from 0.01 to 1 s⁻¹ and then decreased tensile strength with increased strain rate from 1 to 100 and to 200 s⁻¹; 38MnB5 exhibited consistently increased tensile strength with increased strain rate, giving rise to a maximum true stress of 2443 MPa under a strain rate of 200 s⁻¹. The evaluation of crashworthiness and downgauging opportunities of 25MnVB5 and 38MnB5 relative to 22MnB5 were also subsequently evaluated [75]. Applied to the B-pillar reinforcement of a 50 km/h full vehicle side impact simulation, 38MnB5 gave rise to an 11 mm reduction in the final displacement (intrusion) of the B-pillar reinforcement. It was also demonstrated that the 38MnB5 B-pillar reinforcement of 1.4 mm resulted in a similar intrusion as the 22MnB5 B-pillar reinforcement of 1.6 mm, thereby giving rise to a 12.5% downgauging opportunity.

Composite sheet steels

ThyssenKrupp Steel AG revealed TriBond1200 and TriBond1400 to the hot-stamping market in 2015. Both TriBond products exhibit a multi-layered structure, consisting of a sheet of MBW1500 sandwiched and laser welded between two sheets of MBW500. In the TriBond1200 product, the MBW1500 sheet accounts for 60% of total thickness, while in the TriBond1400 product, the MBW1500 sheet accounts for 80% of total thickness. When applied to hot stamping, the inner layer of MBW1500 is hardened to martensite, while the outer layers of MBW500 form ferrite and martensite. ThyssenKrupp Steel AG claim that the three-layer structure is available with different material combinations and different layer thicknesses to tailor the product to the required part properties. The TriBond products supposedly exhibit increased bendability and thus, hot formability compared to the referenced monolithic MBW1500 product, while exhibiting superior combinations of anti-intrusive and impact-energy absorptive crashworthiness in the final part. Further research is required to understand the true improvements to hot formability and crashworthiness, while the potential for delamination between the laser-welded sheets also requires understanding. The TriBond products are currently listed as ‘special mill grades’, while no OEM currently uses the TriBond products. The ability to upscale is worth consideration.

Functions

Coatings applied to the sheet steel are designed to serve numerous functions during and after hot stamping. During hot stamping, coatings protect the sheet steel from oxidation and decarburisation at elevated temperature and exposure to an oxidising atmosphere in the furnace heating, transfer and/or hot-stamping stages. The oxide scale typically exhibits a thickness of 2–5 µm and increases tool wear due to excessive friction between tool and hard oxide scale, reduces formability (potentially giving rise to splitting of the part) and reduces die quenching efficiency due to a lower heat transfer coefficient of the oxide scale (relative to the steel substrate) and also due to the heterogeneous contact between tool and part owing to associated surface roughness of the oxide scale. The residue oxide scale accumulated on the tooling can also lead to scoring of subsequently formed parts. The oxide scale requires removal (at extra cost), typically by shot blasting following hot stamping to provide acceptable weldability and paintability. Decarburisation gives rise to heterogeneous hardenability (typically marked by ferrite formation within 5 µm of the surface) and resulting mechanical properties through part gauge. After hot stamping, coatings provide barrier and/or cathodic corrosion protection. It is reported that in the earliest hot-stamping process of the 1970s and 1980s, the oxide scale was removed by chromium shot blasting, which left a thin film of chromium on the part surface to provide barrier corrosion protection [19].

Aluminium–silicon

Hot-dip-aluminised (aluminium–silicon coated) sheet steels have been used for decades in high-temperature applications, such as automotive exhaust systems (designed to operate at up to 900°C). ArcelorMittal patented hot-dip-aluminised boron steel for hot stamping in 1999, while commercial production by ArcelorMittal began in 2006 and the first commercial application took place in 2007 [12]. In addition to the first patent which stipulates the application of aluminising coatings to sheet boron steels [76], ArcelorMittal has a second patent which stipulates the use of aluminised sheet steels in the hot-stamping process [77].

Aluminised boron steel dominates the hot-stamping market and is considered the benchmark. Aluminised boron steel for hot stamping is supplied by the patent originator ArcelorMittal, in addition to ThyssenKrupp Steel AG and Nippon Steel & Sumitomo Metal Corporation under licence by ArcelorMittal, to collectively dominate the European, North American and Japanese markets. POSCO supplies aluminised boron steel for hot stamping to much of the East and South-East Asian markets, including China and South Korea; in addition to the South American market, as well as Mexico, which are industrially and technologically significant automotive regional markets that ArcelorMittal's patents do not cover (Figure 46). Kobe Steel Corporation and several Chinese steel manufacturers, including Baosteel Corporation, also supply aluminised boron steel for hot stamping to China.

OPEN IN VIEWER

Figure 46

Regions covered by ArcelorMittal's patents pertaining to aluminised sheet steels for hot stamping.

World production of aluminised boron steel for hot stamping increased from 5 kt in 2006 to approximately 480 kt in 2011 [78]. While aluminised boron steel dominates the hot-stamping market and is secured under lucrative patents by ArcelorMittal, it is interesting to note that neither boron steels nor aluminising coatings represent new technologies. The ingenuity behind ArcelorMittal's patents was to combine these two well-known technologies and to apply the result to the highly lucrative hot-stamping market.

In accordance with the widespread use, of all the coatings applied to hot-stamped sheet steel, aluminising coatings are by far the most abundantly documented in the literature pertaining to hot stamping, for example by [19]. The continuous hot-dip aluminising coating is applied with a pot chemistry equal to or close to the eutectic aluminium–silicon composition, typically 8–12 wt-% Si (and sometimes with up to 3 wt-% Fe) and maintained at approximately 675°C (with a melting point of approximately 575°C). The coating weight is typically 80–150 g/m² per side (25–30 μm thickness) depending on customer specification. Two distinctive layers are present, a ‘coating layer’ and an ‘interface layer’. The coating (upper) layer, typically 6 µm in thickness, is an AlSi solid solution containing between 65 and 95 wt-% Al (equal to or close to the eutectic composition), while the interface (lower) layer, typically 20–22 µm in thickness, is predominantly an inter-metallic compound of Fe₂SiAl₇, although a finer sub-interface layer of Fe₂Al₅ and FeAl₃ inter-metallic compounds is also common [19]. Iron diffuses into the aluminium–silicon coating from the substrate to form the inter-metallic compound(s)/interface layer(s) during the hot-dip process. Typical average chemistry of the total coating is (wt-%) 83Al–14Si–3Fe [79].

During the furnace heating stage of the hot-stamping process, with furnace temperature far greater than the melting point of the (near-) eutectic coating layer (∼575°C), the coating layer melts. Iron from the substrate and interface layer(s) diffuses into and alloys with the molten coating layer to progressively form a sequence of inter-metallic compounds, initially Fe₂Al₅ and Al₈Fe₂Si relatively early in the heating stage, which subsequently transform into a multi-layered (typically five layered) structure of numerous solid solution and inter-metallic compound phases on prolonged heating. It has been demonstrated that the five sub-layers that make up the coating differ by aluminium content: approximately 50 wt-% in the first (upper) and third layers, 30 wt-% in the second and fourth layers and 10 wt-% in the fifth (lower) layer closest to the interface layer(s) [79]. Two of these sub-layers are FeAl₂ (inter-metallic compound) and FeAl (solid solution) with melting points in excess of 1100°C. The complete coating is thus solid by the end of the furnace heating stage, with a typical total coating thickness of 15 µm [78], although it has been suggested that a greater coating thickness of 40 µm is optimal for weldability and (barrier) corrosion protection of the final part [80]. Coating thickness can be increased by lower heating rates, higher soak temperatures and/or longer soak times, as these conditions give rise to greater intra-layer diffusion, alloying and inter-metallic compound formation. Recommended furnace conditions for the optimal evolution of the coating as described above include a heating rate of <12°C/s, a soak temperature of 890–940°C and a soak time of 4–8 minutes.

When exposed to elevated temperature in an oxidising atmosphere (e.g. during the transfer stage), the surface of the aluminising coating oxidises to form an oxide layer of Al₂O₃. However, this is a thin, compact, adherent and passivating oxide layer which prevents further oxidation (and decarburisation) of the coating and substrate beneath. This is the advantage of aluminising coatings and why they have become dominant in hot stamping. However, aluminising coatings present numerous disadvantages.

Aluminising coatings are hard and brittle when cold and therefore cannot be used with the indirect hot-stamping process due to cracking of the coating and excessive (pre-forming) tool wear. This means that for a steel manufacturer to serve both the direct and indirect hot-stamping markets, two different coated products are necessary. Even when formed at elevated temperature through the direct hot-stamping process, cracking is common. It has been demonstrated that cracks form during hot stamping due to differing thermal expansion coefficients among the various inter-metallic compounds, solid solution phases and steel substrate [81]. It has been demonstrated that under applied tensile strain, cracks propagate along the direction perpendicular to the coating/substrate interface. When cracks reach the interface layer, the coating layer is divided into segments by through-thickness cracks. With increasing deformation temperature, crack density decreases, while cracks become wider. It has been suggested that with a higher deformation temperature, the aluminising coating may withstand more deformation before the steel substrate is exposed [81]. Alternatively, so-called aluminide coatings have evolved from the traditional aluminising coatings. One of the earliest publications pertaining to aluminide coatings for hot stamping reported annealing of a typical aluminising coating with a relatively thin initial coating thickness of 4–7 µm [82]. After the annealing stage conducted at 950°C for 5–10 minutes, the aluminising coating consisted of exclusively Fe₃Al and FeAl phases – a so-called aluminide coating. The aluminide coating was reported to exhibit much greater ductility and crack resistance under subsequent hot-stamping conditions than the traditional aluminising coating consisting of up to five layers of various inter-metallic compounds and solid solution phases. However, practical application is limited, as the annealing process window for producing the aluminide coating is very narrow, requiring very precise coating chemistry (controlled through Fe diffusion from the substrate) and temperature.

The range of furnace heating conditions are constrained in accordance with optimal evolution of the aluminising coating (as described above), where higher heating rates, lower soak temperatures and shorter soak times than those required for the optimal evolution of the coating can give rise to improved microstructural and mechanical property evolution of the sheet steel substrate and improved process efficiency.

Melting of the aluminising coating layer during furnace heating can result in residue accumulation on the rollers of the roller hearth furnace, with residue then scoring blanks and/or causing blanks to slide in the furnace. Sliding of blanks in the furnace contributes to scoring and also requires greater spacing between consecutive blanks in order to avoid contact, which reduces process efficiency. Prolonged accumulation of the molten coating layer can lead to a thermo-chemical attack on the rollers, ultimately resulting in failure [83]. It has been demonstrated that even very low heating rates of less than 1°C/s are insufficient to prevent melting of the coating layer. In order to overcome melting and resulting residue accumulation, a two-stage heat treatment has been developed, in which the blank is initially (batch) furnace heated to 650°C where the formation of the solid phases is accelerated; and then in the second stage, the blank is subjected to the conventional furnace heating regime, during which the pre-treated coating does not melt [79]. It may be suggested that an improvement to the above could be to conduct the pre-heat treatment by induction heating immediately ahead of the roller hearth furnace. This would provide a localised form of heating so as to heat treat the coating as desired, but without interfering with microstructural evolution of the sheet steel substrate, which has been raised as one concern with the method proposed above. The induction heating method would also be more efficient and compatible with the hot-stamping line. It is also worth considering that with the pre-heat treatment concept, the problem of melting of the aluminising coating and subsequent residue accumulation on the heating equipment is merely moved from the hot-stamping line to the pre-heat treatment furnace (thus the problem is moved, rather than solved). Thus, as a further improvement, the pre-heat treatment could be combined with the continuous hot-dip aluminising process in the production of the sheet steel, in a so-called alumannealing process (comparable to galvannealing). Such a process may provide the means to overcoming melting and resulting residue accumulation, in addition to cracking, as discussed above.

Aluminising coatings are relatively expensive owing to the high raw material cost of aluminium. Further, while it has been argued that aluminising coatings provide cathodic corrosion protection to the sheet steel substrate, any such cathodic corrosion protection has been demonstrated to be negligible [78]. Therefore, aluminised hot-stamped sheet steel is mostly confined to the dry zone of the BiW, namely internal and concealed structural parts. It is interesting to note that many OEMs specify cold-formed galvanised (GI)/galvannealed (GA) sheet steels for such applications, where such galvanic (cathodic) corrosion protection is not strictly necessary due to concealment of moisture. Nonetheless, in view of expanding applications of hot-stamped sheet steel to the wet zone of the BiW and most of all, to chassis and suspension, cathodic corrosion protection is in demand. Moreover, the mentioned technical disadvantages of aluminising coatings are in addition to the political disadvantage, namely the widespread monopolisation of the aluminised hot-stamped sheet steel market by ArcelorMittal. This maintains the market price at a premium and thus, tier 1 part manufacturers and OEMs pay a premium for an aluminised sheet steel product for hot stamping. Combining the technical disadvantages with the political disadvantage, there is a huge demand for alternative coatings for hot-stamped sheet steel.

Zinc/zinc–iron

The two most common forms of zinc coating applied to sheet steel are hot-dip GI and GA. Both GI and GA coatings have been used for traditional cold-stamped sheet steel automotive parts for decades, primarily due to the provision of cathodic corrosion protection. Cathodic corrosion protection means that the coating is sacrificial to the steel substrate so that in the event of the coating becoming breached and exposing the steel substrate, the coating will sacrificially corrode to preserve the steel substrate. Moreover, the zinc oxide corrosion product that provides cathodic (sacrificial) corrosion protection is a strongly adherent, thin, optically invisible and passivating layer.

GI and GA coatings are often referred to as zinc and zinc–iron coatings, respectively. The continuous hot-dip GI process involves dipping the sheet steel into a molten pot of zinc immediately after the continuous annealing cycle. The zinc pot chemistry is usually commercially pure (containing ∼0.2 wt-% Al) and maintained at 460°C. Following hot dipping, the zinc-coated sheet steel is cooled at ambient temperature and the zinc coating solidifies. This process is very similar to continuous hot-dip aluminising. In contrast, the GA process involves an additional heating stage immediately following hot dipping, in which the zinc-coated sheet steel is heated to and maintained at 480–520°C to induce diffusion of iron from the steel substrate into the zinc coating and formation of iron–zinc phases.

Similar to the aluminising coating, the GI coating consists of two layers: an interface layer of the Fe₂Al₅Zn inter-metallic compound, typically 1–2 µm in thickness, where aluminium in the zinc pot chemistry segregates to the interface layer to form the inter-metallic compound; and a coating layer of pure zinc, typically 10–20 µm in thickness. Total coating weight is typically 30–60 g/m² per side (10–20 μm thickness) depending on customer specification. For GA coatings, the aluminium content of the zinc pot chemistry is typically lowered to ∼0.1 wt-% in order to suppress the formation of the Fe₂Al₅Zn inter-metallic compound and to permit the immediate formation of iron–zinc inter-metallic compounds at the interface. This accelerates the formation of iron–zinc phases throughout the coating during the subsequent heating stage. GI coatings exhibit greater cold formability than GA coatings owing to the enhanced ductility of the interface layer (Fe₂Al₅Zn inter-metallic compound) and the coating layer (pure zinc), whereas the continuous layer of iron–zinc phases that characterises GA coatings is relatively brittle.

In an attempt to provide hot-stamped sheet steel parts with cathodic corrosion protection, the first commercial application of zinc-coated (GI) sheet steel to (indirect) hot stamping took place in 2008 [12]. However, the limitation of zinc coatings for hot stamping is the melting point of zinc and of the evolved phases is far below the furnace temperatures of hot stamping. During furnace heating, the coating layer evolves to form an oxide layer (at the surface), a liquid zinc layer and an α-Fe(Zn) solid solution layer (at the interface). The oxide layer in GA coatings is typically ZnO which is highly susceptible to evaporation. The oxide layer in GI coatings is typically Al₂O₃ (where aluminium in the interface layer diffuses to the surface) which prevents evaporation of zinc. During the transfer and hot-stamping stages, the liquid zinc layer reacts with the α-Fe(Zn) solid solution layer to form the Γ-ZnFe inter-metallic compound at ∼782°C (according to the equilibrium phase diagram). With a lower heating rate, a higher soak temperature and/or a longer soak time, the α-Fe(Zn) solid solution layer gradually consumes the liquid zinc layer to give rise to a thinner Γ-ZnFe inter-metallic compound layer following hot stamping. Total coating thickness also increases. Total thickness after a typical hot-stamping cycle is 20–25 μm. Both the α-Fe(Zn) solid solution and Γ-ZnFe inter-metallic compound phases provide cathodic corrosion protection of the steel substrate. However, the Γ-ZnFe inter-metallic compound phase provides the highest cathodic corrosion protection and is therefore preferred in the final part. Obtaining the preferred Γ-ZnFe inter-metallic compound requires melting of the coating layer (given that liquid zinc is a necessary precursor to the formation of the Γ-ZnFe inter-metallic compound). Moreover, co-existence of liquid zinc and the Γ-ZnFe inter-metallic compound is typical under hot-stamping conditions, owing to the temperature (∼782°C) at which the Γ-ZnFe inter-metallic compound forms (which is in the typical hot-stamping temperature range). The presence of liquid zinc during hot stamping potentially leads to LME, as liquid zinc initiates and leaches into micro cracks at the substrate surface under hot deformation. The indirect method of hot stamping was developed for zinc-coated sheet steel. By the indirect method, very little hot deformation takes place, with the majority of deformation taking place cold before furnace heating. The comparative absence of hot deformation prevents LME.

In order to overcome LME when hot-stamping zinc-coated sheet steel via the direct method, the pre-cooling technology has been developed [84]. The blank is transferred from the furnace to a pre-cooling unit and rapidly cooled to a temperature below 700°C. The blank is then transferred to the forming tool, stamped into the desired part geometry and die quenched in the forming tool to near-ambient temperature. The pre-cooling stage enables liquid zinc to solidify to the Γ-ZnFe inter-metallic compound so that during the hot-stamping stage, no liquid zinc is present and LME is avoided. Given that the blank is maintained in the metastable austenite phase field throughout pre-cooling, transfer and hot-stamping stages; transformation to martensite while die quenched, as in the traditional hot-stamping process, is maintained. It is also claimed that the process permits the use of leaner sheet steel chemistries of lower hardenability [85]. While the pre-cooling technology solves the problem of LME, formability will be impaired at lower deformation temperatures. The practicalities of up-scaling the technology remain to be demonstrated. Additionally, liquid zinc residue is still permitted to accumulate on the rollers of the roller hearth furnace and also on the forming tool, in a similar fashion as described for aluminising coatings.

Hybrid sol–gel

There have been three generations of hybrid sol–gel coatings. All three generations have been based on organo-silane or organic monomer precursors that are synthesised to produce a cold applied sol–gel coating. The sol–gel coating may be applied at the end of a continuous annealing line by dipping, rolling or spraying and then dried in a hot air atmosphere. The first-generation sol–gel coatings contained wax and graphite additions, with wax additions aiding tribological properties at ambient temperature (for indirect hot stamping) and with graphite additions aiding tribological properties at elevated temperature (for direct hot stamping). The first-generation sol–gel coatings provided oxidation and decarburisation resistance during hot stamping but could not be resistance spot welded due to poor electrical conductivity and did not provide cathodic corrosion protection of the final part. The second-generation sol–gel coatings exhibited aluminium additions in order to provide sufficient electrical conductivity for resistance spot welding. The third-generation sol–gel coatings exhibited magnesium additions in order to provide a degree of cathodic corrosion protection of the final part.

One of the latest and most technically successful variants of hybrid sol–gel coatings is Hydrosol [86]. Hydrosol can be classified as a third-generation hybrid sol–gel coating, containing an organic solvent, a curable organic binder and a combination of metallic particles, where the metallic particles are constituted by aluminium (at least 50%) and a balance (50% or less) of non-aluminium metal or metallic alloy. The non-aluminium particles are composed of zinc, magnesium, copper and tin at 10–20%. The optimal particle size is said to be 5–30 µm, which may be achieved from powders, flakes or a combination of both. The presence of the non-aluminium particles is said to reduce the formation of the oxide scale at the steel substrate surface by forming a diffusion layer with iron above 700°C; and by forming an inter-metallic layer at temperatures above 900°C. Zinc and magnesium particles provide cathodic corrosion protection. The curable organic binder comprises polyamic acid. When cured, the polyimide binds with the other constituents of the coating and provides barrier corrosion protection. The coating is said to exhibit an electrical resistivity of just 0.1–2.0 mOhm and thus, suitability for resistance spot welding.

The hybrid sol–gel coatings have demonstrated excellent combinations of decarburisation resistance, oxidation resistance, cathodic corrosion protection and weldability. However, residue accumulation, tribology and up-scaling opportunities are areas of interest that need further understanding.

Lubricants and organic thin films

As with all forming processes, the coefficient of friction is an important parameter and should be optimised for formability. In hot stamping, the elevated temperature and corresponding evolution of tribological properties of the sheet steel further complicate friction. The coefficient of friction of aluminised sheet steel in hot stamping has been demonstrated to be in the order of 0.4 [87] to 0.5 [88] under similar test conditions. The coefficient of friction of aluminised sheet steel during hot stamping is not helped by the coating layer adhering to the forming tool. It has been demonstrated that the coefficient of friction of galvanised sheet steel in hot stamping is lower than that of aluminised sheet steel in hot stamping under equivalent test conditions [89].

Lubricants are commonly used in forming processes, including hot forging. The effect of a water-based commercial hot forging lubricant (suspected to consist of a hydrophilic polymer dispersed at 20% in water) when hot-stamping-aluminised 22MnB5 steel has been evaluated at 720°C, where the coefficient of friction decreased from 0.55 under the dry condition to 0.20–0.35 under the lubricated condition [90]. Under dry conditions, the coefficient of friction was independent of tool surface roughness (which can be attributed to the sheet steel sliding against the relatively thick adhered aluminium layer), whereas under lubricated conditions, the coefficient of friction was dependent on tool surface roughness (which can be attributed to the sheet steel sliding against the thinner adhered aluminium layer and tool surface beneath, where the adhered aluminium layer is minimised by the presence of lubricant and where the reduced thickness of the adhered aluminium layer enables the tool surface roughness to impart an effect), as illustrated in Figure 47.

OPEN IN VIEWER

Figure 47

Effect of tool surface roughness on coefficient of friction of aluminised 22MnB5 steel during hot stamping dry and with hot forging lubricant (adapted from [90]).

The coefficient of friction when hot-stamping-aluminised 22MnB5 steel has been further investigated with five commercial hot forging lubricants in addition to five novel experimental lubricants [91]. The five commercial lubricants were different variations of a hydrophilic polymer water-based lubricant, while the five experimental lubricants were based on one of the commercial lubricants (containing a mineral salt), with independent solid additions (each at 5%) of swellable mica, non-swellable mica, melamine–cyanuric acid, potassium titanate and cellulose powder. The five commercial lubricants decreased the coefficient of friction from 0.56 under dry conditions to a minimum of 0.27, which can be attributed to the same reasoning as above. The five experimental lubricants, each containing a solid additive (as listed above), were rationalised by the theory that to improve the lubricity of a hot forging lubricant under hot-stamping conditions, it must be considered that the contact pressure in hot stamping is lower than that in hot forging (thus solid additives can increase the local contact pressure). The five experimental lubricants decreased the coefficient of friction from 0.27 under the condition of the referenced commercial lubricant, to a minimum of 0.16. This was attributed to the layered crystal structure of the swellable mica additive giving rise to a low shear stress and in turn breakdown during forming. It can be suggested that this behaviour gives rise to a dynamic lubricating effect with increasing forming pressure. However, the limitations of this lubricating mechanism may be the accumulation of solid residue from the lubricant on the tool surface.

Once used exclusively as lubricants, organic thin films have gained attention for their attributes as coatings during hot stamping in order to prevent oxidation and decarburisation and thus, replace the conventional metallic coatings. One of the earliest investigations into organic thin films as coatings during hot stamping included two oil-based commercial hot forging lubricants, the first containing boric acid, a fatty acid and a sequestering agent; the second containing the same as above, plus phosphate and calcium additions [92]. Applied to the sheet steel in liquid form and then cured to a solid thin film by drying, during furnace heating the solid thin film liquefied to provide an oxidation retardant at an elevated furnace temperature. The thin film containing the phosphate and calcium additions was most effective at minimising oxidation, although some oxide scale was still present on the hot-stamped part while the effect on decarburisation was inconclusive.

Blank heating

The conventional heating method is an electric or gas (depending on region and thus local price) fuelled roller hearth furnace (Figure 48(a)). However, the roller heath furnace exhibits several disadvantages.

Extending across 30–40 m in length, the roller hearth furnace occupies a large area.

OPEN IN VIEWER

Figure 48

Blank heating methods (a) roller hearth furnace, (b) double-decker roller hearth furnace, (c) multi-chamber roller hearth furnace, (d) rotary hearth furnace, (e) conduction, (f) in-tool conduction, (g) longitudinal induction, (h) transverse induction and (i) face induction.

To increase the speed of work from a conventional roller hearth furnace, the concept of a two-floor (double-decker) roller hearth furnace (Figure 48(b)) allows multiple blanks to be heated simultaneously (to feed multiple forming tools) while saving on floor space [93]. The double-decker roller hearth furnace also enables for continuous production during furnace maintenance, as one layer of the furnace can be serviced while the other is still in operation. Double-decker roller hearth furnaces are now commonplace in industrial hot-stamping plants, particularly in North America.

Further efforts to increase the speed of work include multi-chamber roller hearth furnaces (Figure 48(c)), with a direct-fired (gas fuelled) section at the entry end of the furnace and with the temperature here typically set at 1050–1100°C. This provides a relatively rapid heating rate of the blank to the desired soak temperature of typically 900–950°C. The second chamber of the furnace is then composed of radiant tubes (electric or gas fuelled) set at the desired soak temperature. In addition to increased speed of work and reduced floor space, the higher heating rates provide metallurgical advantages, such as a finer martensitic microstructure and in turn, increased strength and toughness in the final part [72]. Multi-chamber furnaces are also in commercial use.

The rotary hearth furnace (Figure 48(d)) has been in existence for many years, but has only recently been considered for hot stamping. The rotary hearth furnace consists of multiple isolated chambers that rotate around a central axis. The leading advantage of the rotary hearth furnace is that each chamber can be maintained in isolation and thus, a staggered maintenance schedule for each chamber means that the entire furnace does not have to be stopped. The rotary hearth furnace can be compact and thus, save on floor space compared to the conventional roller hearth furnace. Rotary hearth furnaces are not in commercial use (for hot stamping) as of yet.

Conduction (resistance) heating involves connecting the blank in series between electrodes (Figure 48(e)). As current is passed through the blank, electrical resistivity gives rise to heat generation by Joule's law. The energy efficiency of conduction heating is directly dependent on blank geometry, being most efficient for geometries with a large length-to-width (and thickness) ratio such as wires and rods, reaching as high as 87%. Heating rates can reach as high as 450°C/s, thus increasing the speed of work compared to the conventional convection and radiation heating. The equipment for conduction heating is much more compact and comparatively inexpensive to both install and maintain compared to the roller hearth furnace, while there is also much greater flexibility to suit production demand. Conduction heating is also benefited by heat generated directly inside the sheet steel blank and not in an external heating device. This assists energy efficiency and safety.

An additional benefit of conduction heating is that it can be carried out in the forming tool (Figure 48(f)), as demonstrated in one of the earliest published example by Mori et al. [94]. The leading advantage of in-tool conduction heating is that the transfer step from heating apparatus to forming tool is eliminated, where the temperature of the sheet steel blank typically decreases by 100–200°C. In-tool conduction heating thus enables the use of leaner chemistry sheet steels of lesser hardenability while still achieving a martensitic microstructure in the final part. Due to the rapid heating rate followed by a minimal soak time and eliminated transfer time, it is permissible to significantly reduce oxide scale formation with the use of uncoated sheet steel blanks. It has also been demonstrated that in-tool conduction heating can give rise to greater homogeneity of temperature distribution in the sheet steel blank and in turn, enhanced formability, as during the conventional hot-stamping process of transferring the sheet steel blank from furnace to forming tool, initial contact between localised regions of the blank and the forming tool ahead of the forming step leads to localised cooling of the blank, in turn giving rise to heterogeneous temperature distributions across the blank [95]. In-tool conduction heating also enables additional heat treatment following die quenching.

In later examples of in-tool conduction heating, such as the movable conduction heating device on die [96] and so-called 1-shot hot stamping [97], the concept of in-tool conduction heating has been developed further. In the ‘1-shot hot stamping’ concept, the sheet steel blank was heated, formed, sheared and die quenched all in the forming tool. The advantage of this process compared to the standard hot-stamping process is the addition of shearing (i.e. conventionally conducted as post-processing to refine geometry and remove flanges) while the part is still in the austenite phase and maintained in the forming tool immediately following the forming step and ahead of the die quenching step.

Given that resistivity greatly depends on cross-sectional area and profile, conduction heating of shaped (non-square or non-rectangular) blanks is highly heterogeneous (local temperature is inversely proportional to local cross-sectional area). Thus, conduction heating has received limited industrial uptake. In the only example, Toyota Motor Corporation is believed to have used conduction heating with rectangular blanks for hot-stamping door beam reinforcements since 2013 [98]. Figure 49 illustrates a laboratory simulation of the process. However, it is not clear if the process is used industrially.

OPEN IN VIEWER

Figure 49

Conduction heating of a blank on a laboratory hot-stamping line (adapted from [98]).

Induction heating involves an induction coil (inductor) that is positioned in close proximity to the sheet steel blank (conductor) and is supplied by a high-frequency alternating current by a generator. The rapidly alternating magnetic field penetrates the sheet steel blank, generating electric (eddy) current. The current flowing against the electrical resistance of the sheet steel blank then generates heat by Joule's law in a similar manner to conduction heating. Heat is also generated up to the Curie temperature by magnetic hysteresis losses. Induction heating depends on the operating frequency of the inductor, electrical conductivity of the sheet steel blank and magnetic permeability of the sheet steel blank. In common with conduction heating, one of the advantages of induction heating is that heat is generated directly inside the sheet steel blank and not in an external heating device. Again in common with conduction heating, induction heating can be carried out in the forming tool, as demonstrated in one of the earliest published example by [99]. The further advantage of induction heating (which is not true for conduction heating) is that there is no direct contact between the sheet steel blank and the heating device. This can in theory provide greater flexibility. The further advantage of induction heating over convection, radiation and conduction heating is that the heating effect is independent of the surface and coating, where coatings of low thermal conductivity limit heating by convection; where coatings of low emissivity limit heating by radiation; and where coatings of low electrical conductivity limit heating by conduction.

There are three main types of inductor, defined by orientation of the induction coil to the sheet steel blank. The longitudinal inductor (Figure 48(g)), defined by an induction coil that loops around the sheet steel blank, offers the highest energy efficiency of up to 98%, but is susceptible to heterogeneous heating and moreover, is difficult to heat past the Curie temperature, requiring an adjustment to the operating frequency. This is due to a sharp increase in the penetration depth of the induced magnetic field at the Curie temperature (where the magnetic permeability of the sheet steel blank suddenly decreases), resulting in the penetration depth becoming greater than the sheet steel gauge and thus, resulting in a loss of heating. The limited penetration depth (by tailoring the operating frequency) is commonly exploited in several industrial applications to provide localised surface heat treatment, such as surface hardening and curing of organic coatings. The transverse inductor (Figure 48(h)), defined by an induction coil that traverses the sheet steel blank on each face, and face inductor (Figure 48(i)), defined by an induction coil that traverses just one face of the sheet steel blank, are not susceptible to varying penetration depth with operating frequency or magnetic permeability and therefore, are not affected by heating beyond the Curie temperature. Thus, even very thin sheet steel blanks can be heated to the melting point with a fixed operating frequency. Transverse inductors exhibit higher energy efficiency than face inductors (95% against 60%), but face inductors benefit from far superior homogenisation of temperature with shaped (non-uniform) blanks.

One of the most promising developments in induction heating technology for uncoated hot-stamped sheet steel involved a combination of a longitudinal inductor and face inductor [100]. The longitudinal inductor was used to initially heat the blank to the Curie temperature of 741°C, while the face inductor was used to immediately continue heating of the blank by another 150–200°C to the desired soak temperature. The dual longitudinal-face inductor process combined the high energy efficiency (98%) of the longitudinal inductor with the high temperature (greater than Curie temperature) and homogenous heating capabilities of the face inductor. While the face inductor has an energy efficiency of just 60%, since it was applied for only a small heating segment (150–200°C), the overall efficiency of the dual configuration was still significantly higher than that of conventional roller hearth furnaces.

While the above process may be considered advantageous for heating uncoated sheet steel blanks, the process is unsuitable for coated sheet steel blanks, especially zinc-coated and aluminium–silicon-coated sheet steel blanks, which are most prominent. This is because melting of the metallic coatings that takes place during blank heating can readily result in liquefied coating dripping onto and contaminating the induction coil. Magnetic field forces can also eject liquefied coating from the substrate. In order to overcome the limitation of induction heating applied to aluminium–silicon-coated sheet steel blanks, a four-stage heating process has been developed [101]. The sheet steel blank was initially heated in a longitudinal inductor to (or just below) the coating's melting temperature of ∼675°C. The blank was transferred to a muffle furnace, where heating was continued to the desired soak temperature by the conventional radiation and convection methods. Liquefied coating was thus not ejected from the substrate by magnetic field forces, nor did the liquefied coating contaminate the inductor. Once the liquefied coating was re-solidified (by the formation of inter-metallic and solid solution phases with continued temperature increase and time), the blank was maintained at the soak temperature by a face inductor. In the final stage, conduction heating was used to homogenise temperature distribution throughout the blank ahead of forming. This process thus combines the advantages of each heating method, namely the high heating rate and energy efficiency of longitudinal induction heating, the absence of magnetic field forces from radiation and convection heating when liquefied coating is present, the high temperature (greater than Curie temperature) and homogenous heating capabilities of face induction heating with greater energy efficiency than the furnace; and finally, the relative homogenisation of conduction heating. However, this process is still restricted to square or rectangular blanks with uniform cross-sectional profile and moreover, the inclusion of four heating methods makes the process complicated and thus, industrial up-scaling is likely to be unfavourable.

In recent years, continued development of induction heating technologies for hot-stamped sheet steel has been scarce. Conduction heating would appear to be the most promising alternative heating technology with the greatest opportunity for up-scaling.

Heating atmosphere

Most furnaces operate in an inert reducing gas atmosphere of nitrogen, hydrogen and argon. With increasing risk of hydrogen embrittlement occurring in steels exhibiting higher tensile strength, the inert reducing gas atmosphere of nitrogen, carbon monoxide and propane has been developed in order to reduce susceptibility to hydrogen uptake during blank heating [102]. However, the extent of oxidation and decarburisation prevention and moreover, the extent of hydrogen embrittlement prevention require further understanding.

Blank transfer

In the early hot-stamping process of the 1970s, the austenised blank was transferred from the furnace to forming tool manually with the use of tongs clamped against an extended flange. This manual process is still commonly used in many laboratories for physical simulation of hot stamping. However, in common with widespread automotive manufacturing since the 1960s and especially since the 1990s, robotics have become commonplace to automate manufacture, reduce cost and improve consistency.

While (cold) sheet steel blanks are usually transferred via magnets or rubber suction pads, hot sheet steel (up to 950°C) needs to be handled with mechanical grips since rubber would disintegrate at this temperature and steel becomes non-magnetic as it changes phase to austenite.

Consistency provided by robotics is of upmost importance in the transfer step of hot stamping. Different transfer times from one hot-stamping cycle to the next will give rise to a different degree of cooling to the blank and in turn, a different deformation temperature and different phase transformation kinetics. Transfer robotics used in hot stamping is typically based on the multi-axis servo-mechanical motion control system. The most advanced examples are capable of completing the transfer step in as little as 4 s. High-temperature-resistant pressure pads grip the blank. Yet more advanced robotics include integrated temperature sensors that monitor the temperature of the blank and can tailor the transfer time to achieve a specified temperature at placement in the forming tool.

Contrary to the implementation of advanced robotics, the vertical hot press (VHP) concept places the furnace discharge door directly above the forming tool [103]. The blank enters the forming tool from above and vertically (assisted by gravity) and is then stamped into the part geometry with lateral motion of the forming tool. Tool opening distance and therefore cycle time is minimised since there is no requirement for large tool openings to accommodate a robot arm. The VHP concept enables transfer times to be minimised to less than 5 s, does not require expensive robotics and given that the blank can be maintained in a controlled atmosphere throughout blank heating, transfer, hot stamping and die quenching, surface oxidation and decarburisation can be eliminated while using uncoated sheet steel.

Pre-cooling

Rapid pre-cooling of austenised blanks by means of compressed air or water during the transfer stage and immediately ahead of hot stamping was originally patented by Voestalpine GmbH and Schuler GmbH in order to prevent LME [104]. The effect of pre-cooling on mechanical behaviour and subsequent phase transformation kinetics has been investigated, where it has been demonstrated that rapid pre-cooling to 700°C before hot stamping increases the work hardening exponent (n-value), which can reduce the probability of cracking defects and introduce more nucleation sites for martensite formation and thereby produce a finer martensitic lath structure, increasing strength and toughness [105].

Pre-cooling may offer the advantages described and is certainly of merit for the application of zinc-coated sheet steels to hot stamping. However, improvement to formability through increasing the n-value is negated by increased flow stress and reduced ductility at the lower part forming temperature. Up-scaling to industrial hot-stamping lines is also likely to be problematic and unfavourable with tier 1 part manufacturers and OEMs alike, as the process window to complete hot stamping in the austenite phase is reduced and thereby potentially introduces infeasibility across a range of different part geometries with different cooling rates.

Press design

The essential components of the hot-stamping press are the ram (descends the punch), punch and die (tools which make direct contact with the sheet steel blank/part), internal cooling channels purged with water (network through the punch and/or die to conduct heat away from the part) and blank holder (maintains the blank/part stable throughout hot stamping).

In contrast to conventional cold stamping, there is additional requirement for a minimal closing time in order to minimise time that the blank is subjected to a slow rate of air cooling. Once the tool has closed, there is then the added requirement of maintained tool pressure until the formed part has been sufficiently die quenched to less than 200°C. For these reasons, the conventional servo-hydraulic- and servo-mechanical-driven presses used in cold stamping are insufficient due to a slow punch decent rate and inability to maintain a constant tool pressure at the dead point, respectively. Thus, presses used for hot stamping are generally high-speed servo-mechanical, servo-pneumatic or most commonly, servo-hydraulic exhibiting multiple actuators, developed specifically for hot stamping. Due to enhanced formability of hot stamping against cold stamping, load capacity of the hot-stamping press is typically much lower than that of the cold stamping press, at 400–1200 tonnes rather than 2500–3000 tonnes.

In order to increase productivity, multiple part production within a given press has become commonplace, with up to four parts hot stamped simultaneously (each part in its own tooling, but all contained in the one press). While increasing production efficiency in theory, this can attribute disadvantages, as raised residual temperature of the tooling (resulting from a greater mass of hot sheet steel) can negate any cycle time saved from producing multiple parts simultaneously, as more time is required for residual temperature to decrease between cycles; load distribution across the different tools within the press can be heterogeneous; and various degrees of wear between the different tools in the press can make tool changes irregular and increase press downtime.

The development of so-called pressure controlled hardening (PCH) technology from Schuler Presses GmbH is claimed to control pressure during the hot stamping and die quenching stages; and distribute the press load evenly over a single part and across multiple parts. The centre of the PCH technology is a flexible bed cushion in the tooling, which dynamically reacts to different local geometries and sheet steel gauges to provide homogeneous contact pressure and thus, conductive heat transfer to the tooling. The technology is also claimed to reduce the typical die quench time of 8–12 s to 6 s per part and thereby increase productivity yet further.

The tool steels for hot stamping require a unique combination of properties compared to cold stamping. The tool steels must exhibit corrosion resistance, hardness, toughness and fatigue resistance; at both low and high temperatures of up to 850°C in order to tolerate contact with the heated blanks. Additionally, the tool steels must exhibit a low thermal expansion coefficient and high thermal conductivity. It is common for the tool steel to be selected depending on the specific part to be formed, with for example, a tool steel of particularly high thermal conductivity preferable when hot stamping a relatively large part geometry, of relatively thick gauge, but of relatively simple complexity with limited forming depth; but on the other hand, a tool steel of particularly high hardness preferable when hot stamping a complex part geometry with large forming depth which generates high friction against the tools. Chemical compositions of the tool steels are typically (wt-% Fe balanced) 0.38C, 0.20Mn, 0.30Cr, 5.00Mo and 1.00V. The tool steels are typically surface treated with plasma particle vapour deposition carbide or oxide ceramic coatings or plasma nitrating in order to increase surface hardness, thermal conductivity and lubricity.

The die quench rate is perhaps the most important parameter since a higher quench rate permits complete martensite formation for sheet steel chemistries of lower hardenability; and also reduces cycle time. The die quench rate is considered the weak link in this respect and dictates the minimum cycle time achievable.

The press was developed containing an intrinsic direct oil quenching system [106]. Oil was injected from a centre hole in the die and circulated with direct contact with the tooling and the part. To enable the oil to achieve this circulation and contact, the punch and die exhibited 5 mm deep, concave-shaped dimples which collected with oil. The quench rate was controlled by injection pressure of the oil in addition to the initial temperature of the oil. It is surprising that no further research would appear to be available in this area of integrated direct quenching in hot stamping. This would appear to be a successful means of significantly increasing the die quench rate, while simultaneously aiding lubrication.

An alternative means of increasing productivity is through ‘air hardenable’ hot-stamped sheet steels. One of the first publications on this topic emerged in 2008 [107]. The chemical composition of the sheet steel was (wt-% Fe balanced) 0.11–0.18C, 0.10–0.30Si, 1.80–2.20Mn, 1.00–2.00Cr, 0.001–0.050Ti and 0.001–0.004B. The air hardenable steels can be hot stamped and then immediately ejected from the forming tool so as to eliminate the die quenching step. The hot-stamped part naturally air cools at ambient temperature, yet still hardens to a martensitic microstructure. While this approach significantly increases productivity, highly rich chemical compositions are required in order to achieve adequate (air) hardenability (as in the example above, containing up to 2 wt-% chromium). Moreover, air hardening eliminates geometric constraint during quenching as provided by press hardening. Air hardening thus has significant potential to introduce springback. Therefore, it is difficult to envisage that air hardenable hot-stamped steels have opportunity for up-scaling. However, press manufacturers are proposing that air hardenable steels are the next generation of hot-stamped sheet steels.

Post-processing

The post-processing step of geometry refinement is typically conducted by laser machining after the die quenched part has been removed from the forming tool. This carries relatively high capital cost, maintenance cost and low speed of work. To increase productivity, geometry refinement has been achieved by localised in-tool conduction heating [108]. While maintained in the forming tool following die quenching, the part is re-heated locally to 600°C by passing current through electrode pins in the blank holder, but which are not in contact with the punch or die. Thus, the part is heated locally in the regions in contact with the electrode pins. The shearing and punching operations then take place in the forming tool at the locally heated regions. Local heating gives rise to much lower strength and thus, easier refinement of geometry. Punch load at the typical shearing temperature of 500°C is cited to be approximately one-third of that at ambient temperature. The process is also faster and the capital and maintenance costs are lower compared to laser machining. As a further development, so-called 1-shot hot stamping has been developed, in which the geometry refinement operations are conducted in the forming tool before die quenching, while the part is still in the austenite phase [97]. In-tool post-processing technology is considered to be on the verge of up-scaling to industrial hot-stamping lines.

Data management

The so-called Industry 4.0 and the digital revolution have not been escaped by hot-stamped sheet steel. Given that hot stamping is a relatively energy-intensive sheet metal forming process, optimisation of energy consumption through effective data management has great potential to increase productivity and profitability as much as the ‘hardware’ advances discussed previously. Moreover, when it is considered that most tier 1 part manufacturers and OEMs are vast organisations operating numerous plants around the world, the potential energy savings are significant. As tier 1 part manufacturers and OEMs are installing more hot-stamping lines in favour of cold-forming lines, the necessity to optimise energy consumption grows ever stronger.

Tier 1 part manufacturer Gestamp implemented a cloud-based energy management platform in partnership with Siemens AG in 2014, which has led to a 15% reduction in energy consumption [109]. The system works by monitoring the behaviour of equipment, from furnace temperatures, to the movement of transfer robotics, to press loads, to compressors and even to lighting. Energy consumption can then be charted over days, hours, minutes or even seconds. Analysis of data permits inefficient energy usage to be identified (e.g. analysing the behaviour of fixed and variable compressors to see how they are working in tandem illustrates any redundant sequencing, which can then be addressed and optimised to create significant energy savings). Algorithms based on energy consumption patterns can be generated which can be used to raise alerts to energy malfunctions of equipment, to predict efficient maintenance schedules, to manage production schedules and to forecast energy demand based on future production schedules.

In addition to energy efficiency, effective data management enables for improvement of product quality, productivity and customer service, for example through monitoring stock throughout the process chain (from purchased sheet steel coil, to blank and to dispatched hot-stamped part), monitoring warehouse space availability (so that excess sheet steel coil is not ordered) and optimisation of plant layouts (so that stock can move through plant efficiently). Systems can also extend to enabling the customer to track its order throughout manufacturing and dispatch, while integration with the systems of the steel manufacturer can enable the tier 1 part manufacturer to track its sheet steel order throughout manufacturing and dispatch. This creates a seamless record for traceability, ensuring that the correct stock and in the correct volume is delivered on time, or if it is not going to be, alternative suppliers can be contacted; while defects can be traced back to the origin for a root-cause analysis. The system also allows for greater flexibility in manufacturing processes, as processes can be automated to meet customer-specific orders.

Origins

Thyssen Stahl AG patented the tailor-welded blank technology in 1986 with application intended for cold forming [110]. The principle of the original tailor-welded blank concept was to laser butt weld sheet steels of different gauges to form a blank which was subsequently formed into the desired part geometry. Later applied to hot stamping, there are the added thermo-mechanical variables of the hot-stamping process, which can be manipulated for tailored microstructural evolution. Thus, there are two broad categories of tailored technologies: tailored blank technologies in which different sheet steels (chemistries and/or gauges) are exploited to produce different mechanical (or structural) properties from a common hot-stamping process; and tailored process technologies in which different heat treatment regimes within the hot-stamping process are exploited to produce different microstructures and mechanical properties from a common sheet steel. Figure 50 provides a schematic overview.

OPEN IN VIEWER

Figure 50

Overview of tailored technologies.

Tailor-welded blank

In the tailor-welded blank applied to hot stamping, two or more sheet steels of different gauges or crucially chemistry (hardenability) are laser welded and then subsequently hot stamped. With the use of different chemistries, regions (of the blank) of lower hardenability give rise to predominantly ferritic microstructures (soft zones). The first commercial application of the tailor-welded blank to hot stamping was the B-pillar reinforcement of the 2007 Audi A4 (Figure 51). The centre and upper regions of the B-pillar reinforcement were constituted by aluminised 22MnB5 steel exhibiting a martensitic microstructure, while the lower region of the B-pillar reinforcement was constituted by aluminised carbon–manganese–niobium steel exhibiting a ferritic–martensitic microstructure. The tailor-welded blank concept may be the oldest of the tailored technologies and comprises the added financial cost of manufacturing the tailored blank. However, it is a relatively simple technique, can be applied to a standard hot-stamping line, does not increase cycle time, is applicable to both direct and indirect hot stamping and almost guarantees reproducibility across a range of part geometries.

OPEN IN VIEWER

Figure 51

Tailor-welded blank in the B-pillar reinforcement of the 2007 Audi A4 (adapted from [12]).

Tailor-rolled blank

Benteler Automotive Engineering GmbH patented the tailor-rolled blank technology for hot stamping in 2002 [111]. In the tailor-rolled blank, different regions of the blank are locally rolled to different gauges before hot stamping. This permits the allocation of thicker gauges at regions of the part requiring maximum stiffness and buckling resistance for anti-intrusive crashworthiness; and thinner gauges at regions of the part requiring lower stiffness and collapse for impact-energy absorptive crashworthiness (simultaneously while lightweighting). Tailored rolling can also be performed on the tailor-welded blank utilising different sheet steel chemistries for further versatility. The first commercial application of the tailor-rolled blank in hot stamping was to the B-pillar reinforcement of the 2012 VW Golf (Figure 52). In a side impact, there is a pendulum effect of the B-pillar reinforcement initiated by the difference in gauge and crash behaviour, with the thinner upper and lower sections of the B-pillar reinforcement undergoing buckling, while the centre section of the B-pillar reinforcement resists buckling. This transfers impact loads into the roof and sill structures, with loads dissipated to the opposite side of the vehicle through the cross members. The tailor-rolled blank presents excellent opportunities for optimised crashworthiness, especially when combined with the tailor-welded blank. However, the process adds significant financial expense and moreover, different local blank gauges can complicate blank heating and die quenching, as regions of different gauge heat and cool at different rates, as demonstrated by [112].

OPEN IN VIEWER

Figure 52

Tailor-rolled blank in the B-pillar reinforcement of the 2012 VW Golf (adapted from [12]).

Patchwork blank

The patchwork blank may be considered to fall under the tailor-welded blank technology patented by Thyssen Stahl AG in 1986, essentially following the same principles. In the patchwork blank, however, local (patch) reinforcements are layered on top and joined (usually by spot welding or laser welding) to the blank before stamping. One of the first commercial applications of the patchwork blank to automotive, although through cold stamping, was the front side longitudinal member of the 1998 Peugeot 206. In common with the tailor-rolled blank, the patchwork blank permits the allocation of thicker gauges at specific regions of the part for increased local strength and stiffness. Moreover, reinforcements may be of the same sheet steel as the base blank or may be of a higher strength sheet steel for a twofold local strength increase (although this has not been applied to hot stamping to date). One of the first commercial applications of the patchwork blank in hot stamping was the 2014 Volvo XC90 (Figure 53). The patchwork blank presents excellent opportunities for optimised crashworthiness, even more so than the tailor-rolled blank and tailor-welded blank, owing to the combined gauge and sheet steel variables within the single technology. Moreover, the process is more economical than tailor-rolled blanks. However, different local blank gauges can complicate blank heating and die quenching, as demonstrated by Lei et al. [113].

OPEN IN VIEWER

Figure 53

Patchwork blanks in the 2014 Volvo XC90 (adapted from [61]).

Tailored heating

ThyssenKrupp Steel AG patented the tailored heating technology in 2005, with the first commercial application to hot stamping being the B-pillar reinforcement of the 2006 VW Tiguan. The blank was aluminised 22MnB5 boron steel throughout. While the centre and upper regions of the blank were heated to above the Ac3 temperature in a conventional manner, the lower region was shielded during furnace heating so as to maintain the temperature below the Ac3 (Figure 54). During die quenching, the austenised region of the blank hardens to martensite, while the partially austenised region of the blank forms a ferritic–martensitic microstructure (soft-zone).

OPEN IN VIEWER

Figure 54

Tailored heating of a B-pillar reinforcement (adapted from [32]).

While shielding (with ceramic or stainless steels masks) is the only industrially practised method of tailored heating to date, alternative methods have been considered. Chamber furnaces with multiple chambers in parallel (separated by an insulating wall) have been suggested [114]. To implement on an industrial scale would require major investment as the existing roller hearth furnace would have to be heavily modified or even completely changed. Although, contrary to the suggestion by others [20], the method is suitable for a large range of part and blank geometries, offering flexibility. The concept of tailored heating by conduction heating has been developed by so-called by-pass conduction heating [115]. Localised regions of the sheet steel blank are non-electrified and thus, not heated, by using localised copper by-passes. The current method of tailored heating with shielding is not very popular due to the sensitivity of such a process and difficulty to precisely control temperatures of the blank. Moreover, when forming at a temperature below the Ac3, formability of the blank is compromised. The concept of tailored heating with conduction heating would provide much more control and precision. Thus, if conduction heating can be up-scaled to industrial hot-stamping lines, this would present greater scope for tailored heating.

Tailored quenching

ThyssenKrupp Steel AG patented the tailored quenching technology in 2006 [116], with the first commercial application to hot stamping being the B-pillar reinforcement of the 2011 Audi A6 (Figure 55). The part was cooled differentially, with local regions of the part exposed to retarded cooling rates (by heating elements integrated within the forming tool) in order to promote localised auto-tempering of martensite, bainite formation or even ferrite formation (soft zones). Tailored quenching by this method is the most common industrially practised tailored process technology, owing to (1) there are no special requirements of blank heating, enabling conventional furnace heating to be used; (2) there is no additional process step in comparison to the standard hot-stamping process; (3) it can be used for the direct and indirect process; (4) the mechanical properties can be tailored over a wide range; (5) it is highly flexibility concerning the size, shape and position of the soft-zone; and (6) formability is not compromised.

OPEN IN VIEWER

Figure 55

Tailored quenching to the B-pillar reinforcement of the 2011 Audi A6 (adapted from [12]).

While integrated heating elements within the forming tool is the only industrially practised method of tailored quenching to date, alternative methods have been considered. Different tool materials that exhibit different thermal conductivities have been exploited [117]. Tools exhibiting grooves were used to locally reduce thermal (conductive) contact between part and die and in turn, to locally retard cooling rates [118]. Localised insulation from ceramics in order to reduce the thermal transfer coefficient to the forming tool has also been trialled [118]. Combining the concepts of tailored heating by conduction [115] and in-tool conduction heating [94] may present an alternative tailored quenching method, where localised regions of the part could be treated to retarded cooling rates with localised in-tool resistance heating. However, such a method would be grossly more complicated than the conventional method of tailored quenching with integrated heating elements and would only be considered if in-tool conduction heating can be up-scaled to industrial hot-stamping lines.

Tempering and partitioning

Additional heat treatment following die quenching has been trialled in numerous configurations, including tempering and partitioning, of complete parts and locally to tailor properties. Localised tempering can be conducted by flame, induction or laser. The quench and partition concept was introduced to hot stamping, where a forming tool exhibiting integrated heating elements was used to raise the temperature from 300 to 400°C before resumed die quenching to near-ambient temperature [119]. The quench and partition method has potential to retain austenite in the final part and thereby increase toughness. If applied to a tailor-welded blank comprising a hardenable boron steel and a sheet steel chemistry exhibiting sufficient silicon, aluminium and/or phosphorous additions so as to make full use of the quench and partition concept, this method would present significant potential.

Cold-formed advanced high-strength steel and UHSS

The choice between cold-formed advanced high-strength steel (AHSS)/UHSS and hot-stamped steel is complex and depends on several factors including part geometry, part application and production volume. For isolated parts (which present difficulty to consolidate, even via hot stamping) and of limited geometric complexity, such as a bumper reinforcement, the formability enhancement of hot stamping may not be utilised and thus, lower cost cold forming (particularly roll forming) may be preferable, whereas for geometrically complex parts (especially those which can be consolidated, such as the A-pillar, roof rail, B-pillar, C-pillar and sill reinforcements), hot stamping may be preferable. For parts requiring relatively heavy gauge for buckling resistance, cold forming may be preferable as downgauging opportunities are limited by the application rather than formability, whereas for parts requiring collapse and with crash behaviour benefiting from lighter gauge, hot stamping may be preferable. For a relatively small production volume, investment in a hot-stamping line may not be returned by the cost savings from part consolidation, whereas for a large production volume, hot stamping may be preferable.

Hot-rolled and cold-rolled martensitic UHSS products exhibiting up to and even above 1400 MPa tensile strength have been available from numerous steel manufacturers worldwide since the early 1990s. Manufacturing of cold-rolled martensitic products can place significant demands on the steel manufacturer, marked by continuous annealing cycles that are beyond the capabilities of most lines originally designed for AHSS such as DP, CP and TRansformation Induced Plasticity (TRIP) steels. SSAB was one of the first steel manufacturers to install a continuous annealing line with direct water quench capability, facilitating cold-rolled martensitic products. In the early 2000s, tier 1 part manufacturer Trim Trends Ltd and SSAB applied a 1200 MPa tensile strength martensitic (M1200) steel to the battery holder of the 2004 Ford Escape Hybrid, with the battery holder manufactured through a modified cold stamping process in which the blank was permitted to move in the forming tool without constraint. This demonstrates the potential for cold-formed UHSS in geometrically complex automotive body engineering application and the potential even for cold-formed UHSS to challenge hot-stamped steel, if the forming process is developed appropriately around the sheet steel product.

Hot-rolled carbon–manganese–silicon TRIP steels were under development by steel manufacturers, such as British Steel, for automotive body engineering applications in the early 1990s. With the advance of continuous annealing lines, cold-rolled variants soon followed during the late 1990s. TRIP steels gained much attention for automotive body engineering applications during the 2000s owing to their impressive combinations of strength and ductility. The stress/strain-induced transformation effect, commonly referred to as the TRIP effect, characterises cold formable TRIP steels typically exhibiting chemical compositions (wt-% Fe balanced) of approximately 0.15–0.25C, 1–2Mn and 1.5Si. Cold formable TRIP steels are processed to exhibit a multiphase microstructure of necessarily bainitic ferrite and retained austenite, with proeutectoid ferrite and martensite also possible. The silicon content retards iron carbide precipitation during bainite formation, giving rise to carbide-free bainite, with the excess carbon over and above the ferrite saturation limit partitioned to the metastable austenite solid solution. The carbon enriched austenite is then stabilised at ambient temperature, with the local Ms temperature depressed below ambient temperature. The austenite content (typically 10–15%) may be stress/strain induced transformed to martensite during plastic deformation. The plastic strain at which austenite is transformed to martensite depends on the stability of the austenite, in turn depending most notably on austenite grain size, austenite grain morphology and austenite carbon content. While it is commonly believed that the excellent combinations of strength and ductility (typically characterised by a product of ultimate tensile strength and a total elongation of 25,000–30,000 MPa%) arising from the stress/strain-induced transformation owes to the named TRIP effect (where crystallographic dilatation resulting from the austenite to martensite transformation gives rise to the internal microscopic plastic strain which is translated into macroscopic plastic strain), it has been demonstrated that the TRIP effect is insignificant, with the maximum plastic strain obtained from the TRIP effect of 100% austenite transforming to 100% martensite equal to just 0.15 [126]. When it is considered that the austenite content in TRIP steels is much lower than 100% (typically just 10–15%) and that total elongation values of TRIP steels are typically in the order of 25–30%, the TRIP effect alone cannot account for the high elongation values. Instead, it has been suggested that the excellent combinations of strength and ductility owe to the multiphase character of TRIP steels and particularly, the nature in which the microstructural composition changes during plastic deformation, starting with a relatively soft and ductile ferrite–austenite microstructure and ending with a relatively hard and brittle ferrite–martensite microstructure owing to the austenite to martensite stress/strain-induced transformation, giving rise to so-called composite deformation behaviour. The ductility of the austenite can be exploited early in the deformation process and then the hardness of the martensite can be exploited late in the deformation process.

Cold formable TWIP steels, based on the principles of Hadfield steels, also gained much attention for automotive body engineering applications during the 2000s. TWIP steels typically exhibit chemical compositions (wt-% Fe balance) of 0.3–0.6C, 15–30Mn, 1.5–3.0Si and 1.5–3.0Al. TWIP steels are processed to exhibit a single phase austenitic microstructure. The manganese content depresses the Bs and Ms temperatures below ambient temperature, enabling austenite to be stabilised at ambient temperature. During plastic deformation, mechanical twins are introduced within the austenite grains. The mechanical twinning process is unique to the FCC crystal structure of austenite in steel and moreover, depends on the occurrence of stacking faults within the crystal. The mechanical twins are analogous to grain boundaries, giving rise to a dynamic grain boundary strengthening effect. Thus, diffusional dislocation glide/slip (the conventional process of plastic deformation in crystalline metals) becomes increasingly inhibited, giving rise to a significant work hardening exponent. Austenitic mechanical twinning and austenite to martensite stress/strain-induced transformation (displacive dislocation slip) are competitive processes. The SFE, which is a measure of the atomic mismatch in a solid solution and influenced most strongly by chemistry and temperature, dictates which process dominates, with lower SFE lowering the critical stress for mechanical twinning and thus, favouring mechanical twinning rather than stress/strain-induced transformation. Manganese decreases SFE. Silicon decreases SFE. Aluminium increases SFE. Thus, combinations of these three elemental additions can give rise to tailored combinations of austenitic mechanical twinning and austenite to martensite stress/strain-induced transformation.

Despite their impressive combinations of strength and ductility, both cold formable TRIP and TWIP steels have experienced limited uptake by the steel and automotive industries. This has been due to numerous manufacturing limitations. Quench & Partition (Q&P) steels and Transformation-induced plasticity Bainitic Ferrite (TBF) steels have recently been developed for automotive body engineering applications and are now available from numerous steel manufacturers [127]. Q&P and TBF steels seek to find a balance between the relative simplicity of DP and martensitic steels; and the impressive mechanical properties of TRIP and TWIP steels. Q&P and TBF steels share similar chemistries to TRIP steels, based on carbon–manganese–silicon. The microstructure of TRIP steels consists of a predominantly (proeutectoid) ferritic matrix with bainitic ferrite and retained austenite dispersions, produced by isothermal holding in the bainitic region following rapid cooling from a predominantly (proeutectoid) ferritic microstructure (produced from either inter-critical annealing or slow cooling from the austenite phase field into the proeutectoid ferrite phase field). The microstructure of TBF steels consists of a predominantly bainitic ferrite matrix with retained austenite dispersions, produced by isothermal holding in the bainitic region following rapid cooling from a fully austenitic microstructure. As a result of the harder bainitic ferrite in place of the softer proeutectoid ferrite, TBF steels exhibit higher proof-to-ultimate tensile strength ratio and lower elongation than TRIP steels. The microstructure of Q&P steels consists of a predominantly tempered martensitic matrix with retained austenite and/or bainitic ferrite dispersions, produced by isothermal holding in the bainitic region following partial martensite formation, or by isothermal holding below the Ms temperature. As a result of the harder tempered martensite in place of the softer bainitic ferrite, Q&P steels exhibit yet higher proof-to-ultimate tensile strength ratio and lower elongation than TBF steels.

Such a 1200 MPa tensile strength TBF steel formed by cold stamping has been co-developed between Nippon Steel & Sumitomo Metal Corporation and Nissan with intended application to the B-pillar reinforcement [128]. The development of TBF and Q&P steels for cold stamping and roll forming is an alternative approach to not only hot stamping, but also to developing cold-forming technologies around conventional AHSS and UHSS such as martensitic products.

Non-ferrous alloys

While sheet steel remains the dominant material in automotive body engineering, the application of non-ferrous sheet materials is increasing. Aluminium alloys (typically magnesium and magnesium-silicon alloyed 5xxx and 6xxx series respectively) experienced gradual uptake for automotive body engineering during the 1990s and 2000s, where the 1990 Honda NSX was the first production vehicle to exhibit an all-aluminium alloy (monocoque) body, which was followed by the aluminium space-frame technology of the 1994 Audi A8. Subsequently, the application of aluminium alloys has continued to grow (for example, with Jaguar Land Rover). 5xxx and 6xxx series aluminium alloys used in automotive body engineering typically exhibit ultimate tensile strength of just 200 and 300 MPa, respectively. The poor cold formability of these alloys and especially 7xxx series (zinc alloyed) aluminium alloys has previously limited their application to automotive body engineering. With the development of the Hot Form Quench (HQF) technology [129], which is essentially the hot-stamping process as applied to sheet steel, adapted to sheet aluminium alloy, the application of sheet aluminium alloys to automotive body engineering has an opportunity to grow. However, 7xxx series aluminium alloys typically exhibit an ultimate tensile strength of just 600 MPa. Thus, aluminium alloys, even 7xxx series, cannot match hot-stamped sheet steel for anti-intrusive crashworthiness. Moreover, aluminium alloys exhibit approximately a third the elastic modulus of hot-stamped steel. Thus, in order to engineer sufficient stiffness into the part, heavier gauges must often be used which negate the lower density of aluminium alloy and renders the hot-stamped steel part lighter. This can be exemplified by the 2018 Audi A8, where numerous aluminium alloy parts of the safety cell, including A-pillar, B-pillar and roof rail reinforcements as used in the previous generation A8, were replaced by hot-stamped steel. For impact-energy absorptive parts, aluminium alloys have merit. This is especially true for low-to-medium volume performance vehicles which place less focus on environmental sustainability and more focus on performance attributed to lightweighting. However, when selecting aluminium alloy over steel for the purpose of environmental benefit attributed to lightweighting, it is worth considering the complete lifecycle. Extracting aluminium from its ore involves an energy-intensive electrolysis process. Further, aluminium alloys pose limited recycling opportunities. These factors often negate any environmental benefit brought about by lightweighting in application.

Magnesium alloy and, to a lesser extent, titanium alloy represent materials that have more recently begun to experience uptake for automotive body engineering. Development of a sheet magnesium alloy manufacturing process that is claimed to be economical and energy efficient may lead to the introduction of sheet magnesium alloy to structural automotive engineering [130]. The lightweighting and crashworthiness potential of a magnesium body structure has been proposed [4], although the steel grades used in the baseline steel body for comparison were not specified and thus, it is not possible to draw an objective conclusion. Moreover, these concepts would appear to be a long way from high volume industrialisation. The limitations of aluminium alloys across the complete lifecycle mentioned above also apply to magnesium and titanium alloys.

Carbon fibre-reinforced polymer

Two of the leading attractions of CFRP for lightweight structural automotive engineering are high specific strength and high specific stiffness. Standard modulus unidirectional CFRP exhibits approximately four and three times the specific strength and specific stiffness, respectively, of the highest strength hot-stamped steel currently applied to the production vehicle. With a focus on crumple zones, CFRP exhibits high impact-energy absorbing capacity. While CFRP is considered to be a brittle material, exhibiting very low tensile elongation values of typically less than 1%, due to numerous microscopic impact-energy absorbing mechanisms, including fracture of the polymer matrix, fracture of the carbon fibre reinforcements, de-bonding between matrix and reinforcement at the interface; and delamination between plies, CFRP parts can exhibit very high impact-energy absorption. Further advantages of CFRP are directionality, corrosion resistance and environmental inertness (particularly applicable to exposed chassis and suspension parts) and high formability.

Despite the numerous technical advantages, high financial costs associated with the traditional manufacturing processes render CFRP unaffordable to high volume automotive manufacturing. Traditional CFRP part manufacturing involves manual lay up of dry carbon fibre plies with inter-layer wetting with resin, typically in a mould or around a mandrel in order to achieve the desired part geometry; and then curing of the resin by exposing the part to elevated temperature and pressure in an autoclave for up to 24 h. BMW AG made a significant progression towards an automatable, rapid, environmentally sustainable, material efficient, financially economical and repeatable CFRP part manufacturing process with suitability for high volume automotive engineering with laying up of pre-resin-impregnated carbon fibre plies (prepreg), induction heating to partially cure the resin, cold forming (stamping) to the approximate final part geometry, injection of additional resin under high pressure simultaneously during further cold forming (stamping) to the final part geometry and then, induction heating to fully cure the resin [131]. However, this process is still highly energy and financially intensive owing to the numerous heating, stamping and pressurised resin injection stages. For this reason, uptake of CFRP has been limited to medium volume production vehicles such as the 2013 BMW i3. Moreover, the energy, financial and environmental limitations of producing the carbon fibre plies (through a complex pyrolysis process) render CFRP even less attractive than non-ferrous alloys when considering the complete lifecycle. CFRP is also even more difficult to recycle than non-ferrous alloys. Thus, these factors may negate any environmental benefit brought about by lightweighting in application. However, the development of CFRP manufacturing technologies with suitability for high volume structural automotive engineering, such as warm forming of cured CFRP sheets [132], certainly have great potential to play a role in the multi-material vehicle of the modern era. The 2018 Audi A8 is a prime example, where a combination of cold-formed AHSS and UHSS, hot-stamped steel, aluminium alloy, CFRP and even (cast) magnesium alloy has been applied simultaneously. This should be embraced by the various material industries, with technologies such as multi-material structural design and joining, developed to optimise lightweight hybrid construction.