Influence of oscillating fiber laser welding process parameters on the fatigue response and mechanical performance of butt-jointed TWIP980 steels

2.1 Materials

The TWIP steels are austenitic steels with a face-centered cubic structure, which gain strength with the twinning mechanism. In all experiments, 1.4 thickness uncoated TWIP980 steel was used. TWIP steels are high strength and elongation steels. Additionally, it contains high manganese. The TWIP (Twinning-Induced Plasticity) steels used in this study were obtained from POSCO Steel Co., a prominent manufacturer based in Pohang, South Korea, known for producing high-quality advanced steels. The chemical compositions are given Table 1. The mechanical properties of TWIP980 steel, which is listed as follows: tensile strength 972 MPa, yield strength 537 MPa, elongation 50 %.

2.2 Oscillating fiber laser welding process

The oscillating fiber laser welding has superior features compared to standard laser welding methods. Welding process parameters are critical for weld quality. The different experimental sets were designed to analyze the effects of oscillating laser welding parameters and to determine the optimum oscillating laser welding parameters. Laser power, welding speed, and linear heat input were determined as the basic input parameters. The sheet plates were cut in 80 mm × 140 mm sample sizes. The samples were joined by continuous oscillating laser welding in the butt configuration. The circular oscillation mode laser was used. Welding operations were carried out on water-cooled fiber laser welding device. The welding process was carried out on an LRW420 model single-axis oscillating 70° inclined laser head, at 1,080 ± 10 nm laser wavelength, 100 mm collimator distance, 200 mm focal distance, under 15 l min⁻¹ argon shielding gas, 1 mm oscillation amplitude. The process schematic visual and test sample dimensions are shown in Figure 1.

Figure 1:

Process schematic visual and test sample dimensions.

Within the scope of the study, the effects of laser power, welding speed and linear heat input were investigated. Since laser power and welding speed directly affect the heat input to the material surface, the linear heat input effect was also investigated as a parameter to be studied. Linear heat input was calculated according to Equation (1). The welding parameters used during oscillating laser welding are shown in Table 2.

2.3 Mechanical and metallurgical investigations

Samples were precisely cut to size and placed in hot bakelite for metallurgical examination. Bakelite samples were sanded from rough to fine on an ATA-Saphir 520 brand automatic sander to remove scratches on the surface. It was polished with diamond solution and etched with 2 % nital solution. Microstructural examination was carried out on a Nikon brand optical microscope. In addition, scanning electron microscope (SEM) analysis was performed on the FEI brand Quanta FEG 250 model device for detailed examinations and SEM-EDS analysis. Tensile test specimens were cut to size according to ASTM-E8 standards. Dimensions of sub-size test samples are shown in Figure 1. In order to avoid the effect of extra heat during cutting of the samples, the cutting process was carried out with water-jet. The tensile testing was applied to determine the maximum strength and elongation at break of welded joints. All tensile tests were carried out on a U-Test brand tensile testing machine at strain rate of 0.001 s⁻¹. Tensile tests were carried out at. In addition, computer software was used to compare areas under the tensile test curve for toughness. Force-controlled fatigue tests were performed on a Shimadzu fatigue testing machine to determine the fatigue behavior of the welds. A stress ratio R (σ _min/σ _max) of 0.1 and a frequency of 10 Hz were used in all experiments. The micro-hardness measurements were carried out on the Metkon brand Duroline-M micro hardness device with load of 200 g and dwell time of 10 s.

3.1 Microstructural analysis

Morphology of cross sections of welded joint at different parameters (laser power, welding speed and heat input) are given in Figure 2. The oscillatory laser welding parameters were found to affect macrostructural. It has been observed that secondary melting areas occur due to the oscillatory movement during laser welding. The formation of secondary melting areas can be explained by the energy distribution in the oscillating laser welding performed by Horník et al. [26]. In case of low heat input, the heat was not sufficient for penetration into the weld center. The laser power distribution was more intense at the boundaries of the oscillatory motion. These secondary melting zones are more obvious at high welding speeds and low heat inputs Figure 2b, c, f and i. A good weld seam with full penetration was obtained in welded joints above 53.33 J mm⁻¹ linear heat input. At lower heat inputs, unmelted areas and secondary melting zones were formed in the weld seam. 1.2 kW laser power, 30 mm s⁻¹ welding speed, 40 J mm⁻¹ heat input could not provide sufficient heat input for joining. As a result of the oscillating movement, a non-welded area was detected in the centre of the weld Figure 2c. At 1.4 kW laser power, 30 mm s⁻¹ welding speed, 46.67 J mm⁻¹ heat input, an unmelted area at the bottom of the weld seam and macroporosity were detected in this transition region Figure 2f. In the case of low heat input, it can be seen that there are areas where the microstructure does not melt. At high welding speeds, secondary melt areas appeared on either side of the weld center. This is due to the oscillating movement and the insufficient time for penetration at high speed. In case of high heat input, full penetration is observed to be achieved. The material melted and flowed from the lower surface of the weld seam due to high heat Figure 2g and h.

Figure 2:

The cross-sections images according to the sample no.

The micro-structure images for high and low heat input cases are given in Figure 3 (Sample g) and Figure 4 (Sample i), respectively. There are three basic zones in the microstructure: base material (BM), heat affected zone (HAZ) and fusion zone (FZ). The welded joint have an austenitic structure containing dendrites [19]. Columnar dendrites were detected in the fusion zone. The dendrites formation can be seen in austenitic steel welding [27], [28]. Compared to the linear laser weld, the dendrites exhibited a relatively more irregular arrangement in the weld fusion zone. Moreover, dendrites were dispersed in the weld pool. It is thought that this is due to the oscillating movement and the solidification mode.

Figure 3:

Microstructure images of sample g.

Figure 4:

Microstructure images of sample i.

The shrinkage porosity occurs due to shrinkage of the metal in the fusion zones. During cooling after welding, shrinkage porosity may occur due to the molten area not being fed with molten metal [29]. Guo et al. [30] reported that TWIP steels are weak in terms of weld defects due to their high shrinkage rate and are prone to the formation of shrinkage porosity in the fusion zone. They emphasized that shrinkage porosity can be reduced by optimizing the welding parameters. The SEM image taken from the fusion zone of sample h. The SEM image shows micro-porosities in the internal structure with diameters ranging from 0.621 to 1.649 µm. In HAZ, grains have become larger due to the effect of heat. After the welding process, the SEM/EDS elemental mapping image taken from the fusion zone are shown in Figure 5. Fe and Mn dominance is clearly seen. The elements exhibit an almost uniform distribution [31].

Figure 5:

EDS elemental mapping of the fusion zone.

3.2 Mechanical testing

In this study, the effects of laser power, welding speed and linear heat input on the oscillating laser welding process were examined. Laser power and welding speed are parameters that directly affect linear heat input. The linear heat input was calculated according to equation (1). The tensile test was applied to determine the mechanical properties of the welded joint. Tensile test stress-strain curves obtained according to the determined process parameters are shown in Figure 6a. The effects of laser power and welding speed on the tensile strength values ​​are shown in Figure 6b. The laser power and welding speed appears to have significant effects on mechanical strength in oscillating laser welding. At 1.2 kW laser power, tensile strength decreased when the welding speed increased. The highest tensile strength of 652.324 MPa was measured at a welding speed of 20 mm s⁻¹. The lowest tensile strength was measured as 287.64 MPa at a welding speed of 30 mm s⁻¹. This may be due to partial penetration and unjoinied region in the microstructure. Although close tensile strengths were achieved at 1.4 kW laser power, there was a minimal decrease at 30 mm s⁻¹ welding speed. At 1.6 kW laser power, the tensile strength first increased and then decreased. The highest tensile strength was measured as 705.137 MPa at a welding speed of 25 mm s⁻¹. At a low welding speed of 20 mm s⁻¹, the pores occurred on the upper surface of the weld as this caused high heat input. These weld defects may have reduced the tensile strength. The effects of laser power and welding speed on the elongation values ​​are shown in Figure 6c. At 1.2 kW laser power and 20–25–30 mm s⁻¹ welding speeds, elongation values ​​of 7.749 %, 3.635 % and 1.855 % were obtained, respectively. At 1.4 kW laser power and 20–25–30 mm s⁻¹ welding speeds, elongation values ​​of 8.626 %, 7.023 % and 4.848 % were obtained, respectively. The elongation values ​​depending on welding speed showed similar trends at 1.2 kW and 1.4 kW laser powers. Elongation decreased when the welding speed increased. At 1.6 kW laser power and 20–25–30 mm s⁻¹ welding speeds, elongation values ​​of 6.139 %, 10.603 % and 2.55 % were obtained, respectively. At 1.6 kW laser power, elongation first increased and then decreased. Partial penetration observed at 30 mm s⁻¹ welding speed caused a significant decrease in elongation. Crash boxes of vehicle bodies are the most common areas of use of TWIP steels. The crash boxes require high energy absorption to prevent the impact energy from being transmitted to the driver’s cabin. In addition, since the usage fields of TWIP steels are fields where safety is at the highest level, energy absorption values ​​are of critical importance. The effects of laser power and welding speed on the energy absorption values ​​are shown in Figure 6d. At 1.2 kW laser power and 20–25–30 mm s⁻¹ welding speeds, energy absorption values ​​of 37417.69 kJ m⁻³, 11141.43 kJ m⁻³ and 2011.31 kJ m⁻³ were obtained, respectively. At 1.4 kW laser power and 20–25–30 mm s⁻¹ welding speeds, energy absorption values ​​of 40563.81 kJ m⁻³, 31957.94 kJ m⁻³ and 17118.77 kJ m⁻³ were obtained, respectively. At 1.6 kW laser power and 20–25–30 mm s⁻¹ welding speeds, energy absorption values ​​of 23884.95 kJ m⁻³, 55755.62 kJ m⁻³ and 4,871.08 kJ m⁻³ were obtained, respectively. The maximum energy absorption value was obtained at 1.6 kW laser power and 25 mm s⁻¹ welding speed. The energy absorption values ​​exhibited almost similar trends to tensile strength and elongation.

Figure 6:

Mechanical properties of oscillating laser welded samples, a) the stress-strain curves, b) tensile strength, c) elongation, d) energy absorption.

Oscillatory laser welding is more difficult to set optimum laser welding parameters in terms of weld quality because the heat input is applied to a wide area and different beam geometry. Since laser power and welding speed will affect the heat input to the material surface, it is necessary to examine the effects of heat input on the mechanics and weld seam. The change in mechanical properties of the linear heat input calculated according to the determined laser welding parameters is shown in Figure 7. In case of complete penetration, a significant improvement in mechanical properties was observed. At high heat input (56–80 J mm⁻¹), tensile strength, elongation and energy absorption values ​​first increased and then decreased. The increase in tensile strength was relatively limited. At low heat input (40–53.33 J mm⁻¹), lower tensile strength, elongation and energy absorption values ​​were obtained compared to high heat input. This difference became more evident in elongation and energy absorption values.

Figure 7:

Heat input depended mechanical properties, a) tensile strength, b) elongation, c) energy absorption.

In this study, all welded samples were subjected to fatigue testing at a constant stress amplitude of 180 MPa. Fatigue life must be high in steels subjected to repeated load [32]. Repeated load may cause sudden fractures. Fatigue test cycle results are given in Table 3. Infinite fatigue life was considered to be 10⁶ cycles. The fatigue testing was terminated in 10⁶ cycles. In welded samples, 150,457–8,197 fatigue life cycles were obtained. The fatigue strength of all welded joints is lower than the fatigue strength of non-welded joints. Maximum fatigue strength was determined at 1.4 kW laser power and 20 mm s⁻¹ welding speed. It was determined that fatigue strength improved at high heat inputs. Due to insufficient heat input at 1.2 kW laser power and 30 mm s⁻¹ welding speed, fatigue test could not be performed due to the lack of successful joining.

The hardness profiles of the welded samples are shown in Figure 8. V-shaped microhardness profiles were obtained in the weld zone. All fusion zones exhibited lower hardness values ​​than the base material. The heat input softened the fusion zones. The base material was measured to be approximately 290 ± 10 HV0.2, and the fusion zone was measured to be approximately 245 ± 15 HV0.2. The FZ exhibited approximately 45 HV0.2 lower hardness than the base material. This can be attributed to the fully austenitic structure and coarse dendritic grains. Wang et al. found similar hardness values in their studies [33]. Hardness values ​​were measured at approximately 265 ± 15 HV0.2 in the grain zones where grain coarsening occurred due to heat. In Figure 8c, sudden hardness increases were detected in the FZ. This is due to insufficient heat input and unmelted regions in the microstructure due to the lowest laser power and high welding speed. The effect of welding parameters on the change in hardness values ​​was limited.

Figure 8:

Microhardness profile of TWIP980 steel welded at, a) 1.2 kW & 20 mm s⁻¹, b) 1.2 kW & 25 mm s⁻¹, c) 1.2 kW & 30 mm s⁻¹, d) 1.4 kW & 20 mm s⁻¹, e) 1.4 kW & 25 mm s⁻¹, f) 1.4 kW & 30 mm s⁻¹, g) 1.6 kW & 20 mm s⁻¹, h) 1.6 kW & 25 mm s⁻¹, i) 1.6 kW & 30 mm·s⁻¹.

3.3 Fractography

Fracture surfaces contain important information about the fracture behavior and reliability of the joint. The fracture surface examination of the welded sample at 1.6 kW laser power, 25 mm s⁻¹ welding speed is shown in Figure 9. All welded joints were broken from their fusion zones. This is attributed to the low hardness in the FZ. The fusion zone can be a cause of low strength because the high heat input is melting the joint and affecting its chemical structure. Additionally, micro-porosities in the fracture zone may have accelerated fracture and reduced the strength of the joint. The welded joints exhibited a ductile fracture mode. Many dimples with small grain were detected on the fracture surface. The presence of small dimples and frequency density on the fracture surface support the ductile fracture mode.

Figure 9:

Fracture surface evaluation, a) SEM image fracture surface at 240 × magnifications, b) SEM image fracture surface at 4,000 × magnifications, c) SEM image with identified dimples, d) dimple size distribution of fracture surface of sample.