Keywords

1 Introduction

Ultra-high strength steel (UHSS) is increasingly used in automotive manufacturing for reducing vehicle weight and improving safety and crashworthiness [1,2,3]. Press hardened steel (PHS), as a type of UHSS, is typically used for automotive crash-resistant structural parts such as A-pillars, B-pillars, and roof rails [4, 5]. The most widely used steel grade in PHS is 22MnB5, which has a martensitic microstructure after hot stamping and a tensile strength of approximately 1500 MPa [1]. Tailor welded blanks (TWBs) are used to produce parts with the required geometry, shape, and mechanical properties while reducing weight and improving crash behavior due to the absence of reinforcing blanks and fewer joining elements necessary [6, 7]. For the TWBs of PHSs, the most common method of joining sheets with varying thicknesses or different materials is laser welding to create a laser welded blank (LWB), which is then subjected to hot stamping [8].

To avoid surface oxidation as well as decarburization of sheet metal during hot stamping, the surface is typically pre-coated with an additional protective layer [1, 9]. Windmann et al. [10] discovered that the Al-Si layer has high-temperature oxidation resistance and excellent corrosion resistance. At present, Al-Si coating is commonly used as a protective coating for PHSs. However, many study results showed that the Al-Si layer deteriorates the final mechanical performances of the joints. Sun et al. [11] revealed that Al-10 wt.% Si coating led to approximately 15% ferrite and 85% martensite generated in the weld seam of the coated joints, whereas martensite is predominantly formed in the weld seam of the uncoated joints. Saha et al. [12] examined the microstructures of FZ at different conditions of pre- and post-press hardening. Results showed that the pre-press-hardened FZ microstructures mainly consisted of δ-ferrite, martensite, and a minimal quantity of lower bainite. On the other hand, the post-press-hardened FZ microstructures were primarily composed of martensite and α-ferrite. Chen et al. [13] investigated the microstructure characteristic of laser-welded Al-Si-coated boron steel joints. They discovered that the Al-Si coating fused during the welding process, forming a Fe-Al phase, which was determined to be a solid solution of α-Fe and Al. This study discovered that the Al-Si layer had a substantial effect on the microstructures and mechanical performances of the welded joints. Vierstraete et al. [14] indicated that laser ablation before welding is a useful approach for forming Al-Si-coated tailor-welded blanks. The approach has been patented by ArcelorMittal. Lin et al. [8] found that the use of filled wire during the welding process can dilute the content of aluminum in the weld, thereby reducing the amount of δ-ferrite phase.

Therefore, partial ablation welding (PAW) and filler wire welding (FWW), in comparison with the traditional self-fusion welded (SFW), were used to enhance the mechanical performances of Al-Si-coated 22MnB5 steel welded joints in this work. The mechanical properties of the PAW, FWW, and SFW joints were compared. Meanwhile, the Al content of the various phases formed in the weld was investigated. Based on these analyses, the influences of welding processes on the microstructure evolution of welded joints as well as the mechanism of performance enhancement were investigated.

2 Experimental Procedure

2.1 Materials

Laser-welded Al-Si-coated 22MnB5 steel sheets with a full martensite microstructure were used in this work. The chemical composition (wt.%) of the base material (BM) is shown in Table 1. The ultimate tensile strength of BM was 1446 MPa, and the elongation was about 6.7%. Figure 1 shows the microstructure and the energy-dispersive X-ray spectroscopy (EDS) line scanning result across the Al-Si layer. The overall thickness of the Al-Si layer was approximately 45 µm and consisted of FeAl, Fe2Al5, and α-Fe layers. As shown in Fig. 1b, two platforms were distinguished by the element variation and were determined as FeAl and Fe2Al5 phases. The crossing position represented the elemental change between α-Fe and BM.

Table 1 Chemical compositions of 22MnB5 steel (wt.%)
Full size table
Fig. 1
figure 1

Microstructure of the Al-Si coating: a microstructure, b EDS line scanning result across the Al-Si coating

Full size image

2.2 Laser Welding Processes

Compared with traditional self-fusion welding (SFW), partial ablation welding (PAW) and filler wire welding (FWW) were used to enhance the mechanical performances of welded joints of Al-Si-coated 22MnB5 steel. The process parameters of the three laser welding processes are listed in Table 2. The laser power and welding speed applied in the PAW process were 3 kW and 6.5 m/min, respectively. 20 µm of Al-Si layer on one side was removed by laser ablation before laser welding. The experimental laser power and welding speed for the FWW and SFW processes were 4 kW and 2.8 m/min, respectively, and the defocus distance was 3 mm above the upper surface of the specimen.

Table 2 Process parameters of three laser welding processes
Full size table

2.3 Mechanical and Microstructural Characterization

Uniaxial tensile tests were conducted at a tensile speed of 3 mm/min using a universal testing machine (MTS E45.105-ATBC). The full-field strain distribution on the specimen's surface was recorded using digital image correlation (DIC) technology. Three specimens were tested for each laser welding process to ensure the repeatability of the experiment. Figure 2 depicts the geometry of a tensile test sample in accordance with the ASTM: E8/E8M standard. Vickers microhardness was determined using a force of 200 g and a dwell duration of 10 s.

Fig. 2
figure 2

Schematic diagram of tensile specimen (unit: mm)

Full size image

Metallography specimens were cut by wire cutting in the direction perpendicular to the weld. The microstructure of the welded joint was observed by optical microscope (OM) and scanning electron microscopy (SEM) after the specimens were inlaid, ground, polished, and etched. The content of aluminum was determined by EDS.

3 Results and Discussion

3.1 Tensile Properties

Figure 3 presents the engineering stress–strain curves of the BM and PAW, FWW, and SFW joints. The corresponding mechanical properties are shown in Table 3. The PAW and FWW joints fractured at the base metal, while the SFW joints fractured at the fusion zone. The ultimate tensile strength (UTS) of the PAW and FWW joints was 1427 MPa and 1431 MPa respectively, slightly lower than the BM (1442 MPa). The total elongation of the PAW and FWW joints was also slightly lower than that of the BM. However, the UTS and elongation of SFW joints reduced to 1276 MPa and 1.4%. The UTS of PAW and FWW joints was 12.75% and 12.93% higher than that of SFW joints, respectively, and the elongation was 77.8% and 76.7% higher. Therefore, compared with the SFW process, PAW and FWW processes can significantly increase the tensile strength and elongation of welded joints.

Fig. 3
figure 3

Engineering stress–strain curves of BM and PAW, FWW, and SFW joints

Full size image
Table 3 Mechanical properties of BM and PAW, FWW, and SFW joints
Full size table

The local ductility of a material is determined by its resistance to crack initiation and propagation, which is responsible for bending, stretch flanging, and crash resistance. In order to determine local ductility, fracture surfaces need to be evaluated [15, 16]. Figure 4 shows the fracture locations and fracture macro-morphology of PAW, FWW, and SFW joints. The three welded joints did not show any significant necking near their failure locations. As shown in Fig. 5, the width and thickness of the fracture surface were obtained by an optical microscope equipped with a digital microscope system, and the fracture surface was perpendicular to the microscope's observation direction. ASTM E8 was used to measure the fracture surfaces’ width and thickness. According to the corresponding equations, the reduction of area at fracture and local fracture strain of PAW, FWW, and SFW joints was about 0.50%, 0.61%, and 0.08%, respectively. SFW joints fractured at the FZ, and the reduction of area at fracture was significantly lower than that of PAW and FWW joints. All three joints showed less than 5% reduction in area at fracture, indicating brittle fractures.

Fig. 4
figure 4

Failure macro-morphology in different tensile specimens: a PAW, b FWW, c SFW

Full size image
Fig. 5
figure 5

Measurement of width and thickness of fracture surfaces: a, d PAW, b, e FWW, c SFW

Full size image

3.2 Microhardness

The microhardness test indentation points were spaced at 0.2 mm, and 10 points were tested on the left and right sides of the weld seam. Figure 6 presents the microhardness profile of PAW, FWW, and SFW joints. The microhardness distribution of PAW and FWW joints was uniform without a softening zone, and the average microhardness values were 491 and 503 HV, respectively. However, in the FZ of the SFW joint, the microhardness declined significantly due to the presence of ferrite structures. The average microhardness of the FZ of the SFW joint decreased to 308 HV, which was about 68% of the BM, corresponding to the phenomenon of the joint fracturing at the fusion zone.

Fig. 6
figure 6

Microhardness profile of PAW, FWW, and SFW joints

Full size image

3.3 Microstructure

Figure 7 presents the morphology of PAW, FWW, and SFW joint cross-sections, respectively. The welds were all hourglass-shaped without obvious defects, such as cracks and porosity. The FZ of PAW and FWW joints was uniform and had a less white phase. In contrast, a light-colored phase was produced in the SFW joint because the entire Al-Si layer was incorporated into the FZ from both sides of the weld. Figure 8 shows the microstructures of the upper and lower regions of the PAW, FWW, and SFW joints. As shown in Fig. 8a–d, the lath martensite microstructures were clearly demonstrated in the FZ of the PAW and FWW joints. In contrast, α-ferrite was generated in the upper and lower areas of the SFW joint, while martensite and a substantial amount of α-ferrite phase existed in the FZ, as shown in Fig. 8e, f. EDS spot scanning was conducted on five spots in each phase to determine the cause of α-ferrite formation and to calculate the average. As presented in Table 4, the content of Al element in α-ferrite (1.73–1.82 wt.%) was considerably higher compared to that in martensite (0.42–0.54 wt.%).

Fig. 7
figure 7

Cross-sectioned morphology of the welded joints: a PAW, b FWW, c SFW

Full size image
Fig. 8
figure 8

Microstructure of the upper and lower regions of the FZ: a, b PAW, c, d FWW, e, f SFW

Full size image
Table 4 EDS spot scanning results of the Al content (wt.%) for different phases in the FZ
Full size table

Saha et al. [17] revealed that ferrite was responsible for the decrease in tensile properties in welded joints. Al contents above 1.2 wt.% have been reported to prevent the transformation of ferrite to austenite, resulting in the stabilization and maintenance of the ferrite phase throughout the welding process [17, 18]. Dippenaar et al. [19] discovered that the presence of Al components reduces free energy change, accelerates ferrite phase transition kinetics, and suppresses austenite phase transition kinetics. The high Al element content is the primary reason for δ-ferrite formation, which recrystallizes to form α-ferrite after the hot stamping process [20]. Wang et al. [21] identified that the existence of δ-ferrite reduces the initial energy and propagation energy of cracks, which leads to easy ductile cracking in δ-ferrite and initiates brittle cracking in the tempered martensitic matrix.

For the SFW joints, the Al-Si layer melts and spreads into the weld seam during the laser welding process, causing the Al content in the weld to increase. The strong stabilizing effect of Al elements on ferrite prevents the transformation of the primary δ-ferrite phase [11, 22], thus producing α-ferrite phases in the SFW joints after hot stamping. The presence of α-ferrite is where failure cracks can easily propagate, reducing the strength and hardness of the SFW joints.

For the PAW joints, the martensite has a low Al content of less than 0.20 wt.%, as shown in Table 4. The PAW process removed the 20 µm Al-Si layer on one side by laser ablation, and the remaining α-Fe (Al, Si) layer was melted into the PAW joints throughout the welding process. However, no ferrite was found in the PAW joints due to the whole martensite structure, which explains the high strength and hardness of the weld, as observed in the mechanical properties.

Furthermore, the Al content in the FWW joints was approximately 0.50 wt.%, as shown in Table 4. Lin et al. [8] revealed that the use of carbon steel filler wire during laser welding can reduce the probability of δ-ferrite generation. As a result, the initial cracks were difficult to interconnect to form long cracks. Thus, the addition of carbon steel filler wire improved the strength and elongation of the FWW joints.

4 Conclusions

In this study, the effect of different welding processes on the mechanical properties, microstructural evolution, and Al content of laser-welded Al-Si-coated 22MnB5 steel was studied. The welded joints were produced using three different welding processes, namely, PAW, FWW, and SFW, and their mechanical properties were evaluated. The main conclusions of this study were summarized as follows:

  1. 1.

    The results presented that the SFW joint had significantly lower strength and elongation compared to the PAW and FWW joints. Moreover, a brittle fracture occurred at the weld seam of the SFW joint. The ultimate tensile strength of the SFW joint decreased from 1442 to 1245 MPa, and the elongation decreased from 6.7 to 1.4%. In contrast, the ultimate tensile strength of PAW and FWW joints showed an improvement of approximately 12%. The fracture locations of the PAW and FWW joints were consistently at the BM.

  2. 2.

    The formation of α-ferrite in the weld caused softening of the SFW joints. The average microhardness of the SFW joints decreased to 308 HV, which was about 68% of the BM. No softening occurred in the FZ of the PAW and FWW joints, and the microhardness distribution was uniform. The average microhardness of PAW and FWW joints was 491 and 503 HV, respectively.

  3. 3.

    The PAW process removes one side of the Al-Si layer by laser ablation, which can effectively control the content of the Al element. The FWW process dilutes the Al content and inhibits the formation of α-ferrite in the weld seam through the use of carbon-steel filler wire. Al-Si coating provided the Al element during the SFW process, leading to a large amount of α-ferrite produced in the weld. In comparison, the predominant martensite was formed in the PAW and FWW joints.