Fatigue and Impact Behavior of Friction Stir Processed Dual-Phase (DP600) Steel Sheets

Abstract

This study investigates the impact of friction stir processing (FSP) on the deformation behavior of 1.1 mm-thick DP600 steel sheets under both static and dynamic loading scenarios, with a focus on the automotive applications of the material. During the process, the large plastic shear strains imposed by FSP resulted in a maximum temperature of 915 °C, leading to a morphological transformation of the martensite phase from well-dispersed fine particles into lath martensite and grain refinement of the ferrite phase. DP600 steel showed an almost two-fold increase in static strength parameters such as the hardness value, yield strength, and ultimate tensile strength. As-received and processed DP600 steel exhibited a plastic deformation behavior governed by strain hardening. However, uniform elongation and elongation to failure after FSP took lower values compared to those of the as-received counterpart. Following the improvement in the static strength of the steel, the fatigue strength of the steel increased from 360 MPa to 440 MPa after the FSP. The finite-life fatigue fracture surfaces of the as-received samples were characterized by the formation of fine bulges due to the variation in the crack propagation path in the vicinity of the martensite particles/clusters. After FSP, the transformation of the martensite particles into coarser lath martensite also transformed the fracture surface into a step-like morphology. The microstructural evolution after FSP caused a decrease in the absorbed impact energy and maximum striker reaction force from 239 J and 37.6 kN down to 183 J and 33.6 kN, respectively. However, the energy absorption capacity of the processed steel up to failure was higher than the absorbed energy value of the as-received steel at the same impact displacement. The simultaneous decrease in both impact energy and reaction force is attributed to the higher cracking tendency of the processed microstructure due to the lower volume fraction of the ferrite phase. The experimental results reported in this study mainly show that FSP is an easy-to-apply and functional solution to significantly improve the static and cyclic strength of DP600 steel. However, it is clear that the reduced total impact energy absorption capacity after FSP may be taken into account in design strategies.

1. Introduction

Friction stir processing (FSP) is a severe plastic deformation (SPD) technique developed to locally improve the microstructure and mechanical properties of metallic materials by applying large plastic shear strains in warm to hot forming temperature ranges [1,2]. In the application of this technique, a rotating tool with a pin and shoulder is forced to penetrate the workpiece until the heat generated by frictional forces at the shoulder-workpiece interface raises the temperature of the processing zone to a level of plasticity that allows the rotating pin to move through a path. As a result of this application, complex metal flow occurs in the areas under the influence of shoulder pressure. During this flow, a piece of metal is first sheared off around the rotating tool as it is transferred from the penetration depth of the pin to the surface of the workpiece and then forged back into the cavity created by the translated pin. As a result, the workpiece is subjected to large plastic strains without significant dimensional change. Such a high level of mechanical and thermal work generally results in a reorganization of the microstructure towards finer grain sizes through the effect of dynamic recrystallization. Dynamic recrystallization-induced grain refinement adds value to metals by providing superior mechanical performance after Hall–Petch type strength enhancement without a significant loss of ductility. Moreover, this added value can be tuned by selecting proper combinations of the process parameters like pin and shoulder diameter and geometry, tool rotation speed, tool transfer speed, and cooling conditions.

The flexibility and effectiveness offered by the FSP have been the subject of many engineering applications where the harmony of mechanical properties is required, such as the lightweighting of engineering structures. In the last decade, the automotive industry has placed the design and construction of lightweight vehicle bodies as its top priority. In this approach, materials to be used in vehicle BIW (body-in-white) parts are mostly replaced by new steel products widely known as advanced high-strength steels (AHSS). This steel family was mainly developed to meet the needs of applications where both high strength and adequate formability are demanded [3,4]. Dual-phase (DP) steels are one of the well-known members of the AHSS that meet these requirements with their unique microstructure consisting of fine martensite particles uniformly dispersed in a ductile ferritic matrix [5,6]. The mechanical behavior of dual-phase steels is dependent on the martensite volume fraction [5]. Increasing the volume fraction of martensite increases the strength of the steel with a loss in ductility [5]. Commercially, DP steels are manufactured to have an ultimate tensile strength in the range of 450 MPa–1200 MPa [5]. On the other hand, designing automotive body components is a difficult task because the final geometry must meet requirements such as an excellent mechanical performance, compatibility with sheet metal forming, and lightweightness in terms of environmental sustainability [5]. It is known that to fulfill this task, only sheets with a high strength and sufficient formability must be used. In contrast, the above-mentioned requirements can vary locally in a body component. For example, some local areas or geometric features may require higher strength due to their operational functions, such as load-bearing, while another area on the same part may require higher ductility to be formed into the desired geometry of the style. Meeting such a variety of mechanical behavior requirements can be achieved by designing with a multi-material/multi-part approach. With this approach, areas with high formability requirements can be manufactured as separate sub-parts using a high-formability sheet. Likewise, areas that require high rigidity with simple geometry can also be made from high-strength materials without formability problems. These sub-parts are then assembled into the final part of the geometry. As expected, this approach requires additional dies and sub-assembly processes to complete part manufacturing and this consequently increases the time and cost of the manufacturing cycle, resulting in a reduction in productivity. Thanks to its ability to tailor mechanical properties to meet the high-strength requirements of local regions, FSP can be viewed as an alternative, practical, and flexible solution to construction with a single-material approach.

Available publications discussing the processing of DP steels through friction stir technologies are mostly concentrated on similar or dissimilar joining via friction stir spot welding (FSSW) [7,8,9,10] and friction stir welding (FSW) [6,11,12,13,14,15]. Miles et al. revealed that friction stir welded DP590 sheets with varying thicknesses had a 20% better formability under tension than their laser-welded counterparts [13]. Lee et al. stated that the strength deficiency in the heat-affected zone of DP590 steel following FSW significantly reduced the formability [15]. Taniguchi et al. subjected DP980 steel to friction stir spot processing to reveal transient microstructural changes and local mechanical properties at different peak temperatures [16]. Gotawala et al. developed a coupled 3D thermomechanical and phase transformation model predicting the temperature history, strains, and phase transformation during the process [17]. Aktarer et al. examined the evolution of the processed 1.5 mm-thick DP steel microstructure and room temperature uniaxial tensile properties [18], texture development [19], Erichsen test performance [20], and hole-expanding behavior [21]. They found that the FSP of DP600 steel resulted in an increase in strength accompanied by a decrease in ductility, Erichsen index, and edge flangeability. Saray et al. showed the FSP process could be applied to thin DP steels with nearly similar mechanical properties and formability behavior compared to results obtained from previous studies [22].

Considering the applications of the AHSS, important measures of success also include fatigue performance and deformation behavior under impact loading conditions because BIW parts of the vehicles may undergo deformations either repeatedly in their service life or shock loading in the case of a crash. Our preliminary findings on the fatigue behavior of the DP600 steel have shown that FSP could be used to improve the fatigue behavior of DP steels [23]. The topic should be focused on revealing the interactions between cyclic fracture behavior and FSP-induced microstructural and mechanical evolution. Moreover, to be able to obtain a more realistic point of view through the engineering application potential of the process, the response of processed DP steel to the impact loading needs to be revealed. However, neither the impact energy absorption behavior nor the cyclic fracture and fatigue behavior of processed DP steel have been comprehensively studied. From this point of view, current work reports the effect of FSP on the deformation behavior of DP600 steel under dynamic loading conditions, including cyclic loading and the absorption of impact energy.

2. Materials and Methods

In this study, 1.1 mm-thick DP600 steel sheets were used with the chemical composition given in

Table 1. Mass percentages of alloying elements of as-received (AR) steel.

Chemical Composition (mass %)	C	Mn	P	Ti	Al	Nb	S	Si	Cr
DP600	0.14	0.5	0.1	0.22	0.030	0.09	0.015	0.5	0.77

. The specimens were processed using a tungsten carbide (WC) tool with a pin diameter, pin length, and shoulder diameter of 14 mm, 5 mm, and 0.8 mm on a homemade FSW/P machine set to run in position control of 0.1 mm shoulder penetration. The tool rotation axis was tilted by 3°, and the rotation speed and linear transition speed were set to 1000 rpm and 1.6 mm/s, respectively, and controlled to remain constant throughout the process. Temperature variations during the process were monitored using a Testo 890-2 infrared video camera. During the process, samples were placed on the FSW/P equipment so that the rolling direction of the steel (RD) was parallel to the FSP direction.

The microstructure of the as-received and as-processed steel was investigated with a scanning electron microscope (SEM) on the samples sectioned perpendicular to the FSP direction using wire electrical discharge machining (EDM). The samples were etched in 2% Nital for 10 s following standard metallographic preparation procedures. The electron backscatter diffraction (EBSD) method was used to conduct a more thorough microstructural analysis. For EBSD analysis, metallographic samples were prepared by the electropolishing technique. Hardness measurements were performed with an indentation weight of 0.5 g and a dwell period of 10 s to reveal FSP-induced local variations through the horizontal path. Dog-bone-shaped samples were used in tensile tests to evaluate the mechanical properties of the steel in both AR and processed conditions. Tensile test samples were cut parallel and perpendicular to the FSP direction using the wire EDM technique. The dimensions of the samples were determined using the ASTM E8/E8M subsize specimen; however, because of the nature of the process, the sample gauge length was shortened two-fold, with samples measuring 12.5 mm × 6 mm × 1.1 mm. The Shimadzu AG-Xplus 250 kN electromechanical tension test unit with a video extensometer was used to conduct the experiments at a 0.001 s⁻¹ strain rate. Mechanical properties were calculated as the average values of three companion samples.

Stress-based fatigue tests of AR and processed samples were performed on a Shimadzu EHF-E 100 kN servohydraulic system at a frequency of 10 Hz. Samples were loaded with a sinusoidal signal between a minimum stress of zero and maximum stress. Fatigue test specimens were extracted parallel to the process route using the wire EDM technique (Figure 1a), according to ASTM E466 in dimensions of 6 mm × 3 mm × 1.1 mm. (Figure 1b). At least three companion samples were tested at each stress level to demonstrate the repeatability of the results. The fatigue specimens that could not fracture at the fatigue life of 2 × 10⁶ cycles were defined as running out [24]. Following the fatigue tests, the fracture surfaces were examined using SEM to identify regions of crack initiation and/or propagation paths in the AR and processed samples.

Impact tests were performed on AR and processed samples using an Instron CEAST 9350 drop-weight impact tester with a 222 kN strain gauge tup and an optical velocity detector. As shown in Figure 1c, the processed samples were positioned so that the striker tip coincided with the center line of the tool path. AR and processed DP steels were tested with a drop weight of 18.5 kg and an impact speed of 5 m/s, corresponding to a strike energy of 231.25 joules. This energy value was determined by the lowest energy value that would cause cracking in the as-received specimen. Contact force (F) and displacement (X) data were collected, and the corresponding deformation energy was calculated as the area under each F-X curve.

3. Results and Discussion

Friction stir processing was applied to DP600 sheets without causing any macro damage, cracks, or deformation discontinuities using selected process parameters. The steady-state stir region temperature reached 915 ± 20 °C during the FSP.

3.1. Microstructural Evolution

The microstructure of AR DP600 steel consisted of martensite and ferrite phases. The morphology of the ferrite grains was predominantly equiaxed, with an average size of 10 ± 5 µm (Figure 2). Close to the grain boundaries of the ferrite phase, very fine grains (particles) of the martensite phase were present as a continuous network (Figure 2).

The influence of FSP on the microstructural properties of DP600 steel is shown in the OM micrograph in Figure 3a; SEM micrographs, inverse pole figure (IPF) maps, and phase/grain boundary maps of SZ and TMAZ are represented in Figure 3b–d and Figure 3e–g, respectively.

The as-received microstructure was strongly affected by FSP. Also, variations in the thermomechanical process conditions, such as plastic strain and temperature, resulted in the development of specific deformation regions, which are distinguished by their microstructural features. These deformation regions are referred to as the stir zone (SZ), thermo-mechanically affected zone (TMAZ), and heat-affected zone (HAZ) in accordance with previous research studies as shown in Figure 3a [11,18]. It is important to note that the HAZ was narrow compared to the SZ and TMAZ.

Friction stir processing strongly affected the as-received microstructure and transformed it into a lath martensite-dominated state (Figure 3b). The volume fraction of the phases in the lath martensite morphology was determined to be about 94% (Figure 3c,d). Furthermore, the microstructure had a primary ferrite phase with a grain size of around 4 μm, as seen in Figure 3d. Such a transformation may have occurred due to the increase in processing temperature over the A_c3 level, which was determined to be 860 °C [7,25]. This may cause the transformation of the as-received microstructure into austenite while severe shear strains are applied. These conditions can lead to the dynamic recrystallization of the austenite phase around the rotating tool [18]. During the cooling stage of the process, the austenite to lath martensite transformation may be accelerated due to the refined austenite grain size.

Figure 3e demonstrates that the volume fraction of lath martensite significantly decreased in the TMAZ. In the TMAZ, the undissolved ferrite stayed the same, and the carbon-rich martensite phase transformed into austenite at the maximum processing temperature in the A_c1–A_c3 range [14,16]. At this temperature range, the ferrite phase may not transform into austenite, leading to the ferrite-based microstructure remaining unchanged (Figure 3f,g). From Figure 3g, refined ferrite grains are also evident, showing that shear deformation is still effective and grain size is not uniform through the region.

3.2. Mechanical Properties

3.2.1. Hardness

The hardness evolution of DP600 steel after FSP is shown in Figure 4. This figure indicates that the as-received hardness of the steel is about 178 HV 0.5. However, the hardness values of the steel very slightly decreased to about 175 HV. This decrease in the hardness values is mainly attributed to the formation of the HAZ due to the annealing effect of the heat generated by the friction between tool–material interfaces. Consequently, the mechanical properties of the HAZ are mainly affected by the process conditions, thermal conditioning of the environment, and lastly chemical composition. In the current study, the hardness loss in the HAZ was negligible. However, hardness values sharply increased in the thermo-mechanically affected zone and saturated at the stir zone at a value of 292 ± 12 HV 0.5. Variation in the hardness values indicated that a strengthened region of 12 mm width can be achieved with the selected process parameters. Based on the observations of the microstructural evolution of the steel, an increase in the hardness of the stir zone may be related to the formation of the lath martensite. During the process, the stir zone is heated to temperature levels higher than those of the A_c3 temperature of the DP600 steel, while subjecting it to severe shear deformation. At this stage of the process, the higher density of the dislocations and grain boundaries in the microstructure of the shear-deformed austenite may accelerate the transformation rate of the austenite to martensitic transformation. As the tool moves away, the temperature of the highly deformed and fine-grained austenite phase region drops rapidly, leading to the transformation of the austenite into lath martensite. This may cause a transition of the as-received microstructure into an SZ microstructure, consisting of lath martensite with a high volume fraction (Figure 3d) leading to an increase in the hardness of the steel at the stir zone.

The hardness gradient in the thermo-mechanically affected region may be related to both the temperature gradient and the deformation-induced strengthening of the ferrite phase. The microstructure of the TMAZ in the regions near to the SZ reflects the general characteristics of the SZ microstructure. Similarly, the microstructure at the TMAZ-AR transition region resembles the as-received microstructure. This is an expected result due to the lower temperature of regions away from the stir zone. As a result of the decrease in temperature below A_c3, there is also a decrease in the volume fraction of the austenite transformation and consequently the lath martensite volume fraction and hardness values. Similarly, regions far away from the stir zone do not undergo austenite to martensite transformation if the temperature remains under A_c1. However, a more-or-less hardness enhancement may be achieved as a result of the dynamic recrystallization of the ferrite grains.

3.2.2. Tensile Test

The engineering stress–engineering strain curves obtained from the testing of AR and processed DP steel samples in the transverse direction (TD) and rolling/processing directions (RD/PD) are shown in Figure 5. The tensile strength (σ_UTS), yield strength (σ_y), uniform elongation (ε_u), and elongation at break (ε_f) derived from these curves are summarized in

Table 2. Tensile test results of AR and processed DP steel samples in the process direction (PD), transverse direction (TD), and rolling direction (RD).

Condition	Tensile Direction	σy (MPa)	σuts (MPa)	Ɛu (%)	Ɛf (%)	K (MPa)	n
AR	(RD)	301 ± 6	621 ± 13	21.3 ± 0.5	34.7 ± 2.1	1135	0.24
	(TD)	335 ± 1	640 ± 10	18.0 ± 0.8	31.0 ± 1.2	1212	0.21
Processed	(PD)	811 ± 48	1054 ± 56	5.8 ± 0.2	13.0 ± 2.0	1714	0.14
	(TD)	682 ± 35	925 ± 23	5.1 ± 0.4	5.4 ± 0.9	1915	0.15

.

As a well-studied structural material, AR DP600 steel reflects strain-hardening-dominated deformation characteristics in both the RD and TD directions. As a well-known behavior of rolled steels, AR DP600 reflects a higher capability of strain hardening with a strain-hardening exponent of 0.23 in the RD compared to that of 0.21 in TD (Table 2) due to the higher free dislocation path in the RD [26]. Consequently, uniform elongation and elongation to failure in the RD took higher values of 21.3% and 34.7%, respectively, than those of the TD (Table 2). The yield strength and ultimate tensile strength of the steel were determined to be 301 MPa and 601 MPa in the RD, respectively. These values were slightly lower than those of the TD (Table 2). The strength, ductility, and hardening behavior of the steel were strongly affected by one-pass FSP. The strain-hardening coefficient of the processed DP steel took values of about 0.15 in the PD and RD directions. A decrease in the strain-hardening coefficient also caused a decrease in uniform elongation and elongation at break values down to 5.8% and 13.0% in the PD and 5.1% and 5.4% in the TD, respectively. The strength of the steel was considerably increased after FSP. The yield strength and ultimate tensile strength reached 811 MPa and 1054 MPa in the PD and 682 MPa and 925 MPa in the TD, respectively. Considering the increase in the volume fraction of the martensite phase in the processed microstructure, an increase in the strength accompanied by a decrease in ductility is an expected result. The mechanical properties obtained from the tensile testing of the processed DP steel also indicate that the measured strength and ductility in the TD are lower than those measured in the PD. These values may indicate that the crack propagation resistance of the steel is higher in the PD. This may be due to complex material flow around the rotating pin, causing material to transfer from the retreating side (RS) to the advancing side (AS) of the process. During the lateral movement of the tool, deformation gradients cause the retreating side of the processed region to be weaker than the advancing side of the process.

3.3. Fatigue Behavior

3.3.1. S-N Curve

Stress-based fatigue curves of AR and processed DP steel samples in the finite life region are shown in Figure 6. The relationship between the stress amplitude and the cycles to failure in the finite life region can be represented with Basquin’s Equation (1).

$$\sigma =c_1{N}^{C_2}$$

(1)

where C₁ is the coefficient, C₂ is the exponent, $\sigma$ is the stress amplitude, and N is the number of cycles to failure. The C₁ and C₂ parameters for AR and processed DP steel were calculated and represented in Figure 6.

As-received DP steel has shown a linear S-N curve behavior at the finite life region and with C₁ and C₂ parameters of 1738 MPa and −0.103, respectively. The stress level to reach 2 × 10⁶ cycles was determined to be around 360 MPa. After FSP, the linear behavior of the S-N curve remained unchanged. The parameters of Basquin’s law C₁ and C₂ took values of 4308.1 MPa and −0.156, respectively. The stress level to reach 2 × 10⁶ cycles was determined to be 440 MPa. Based on these values, it is understood that the FSP process elevated the S-N curve to higher stress levels with a decrease in the exponent (C₂) of Basquin’s law. The knee points of the S-N curves of as-received and processed DP steel seem to be close to each other around 1.5 × 10⁶ cycles (indicated with “x” marks in Figure 6). Extrapolating the finite life region S-N curve characteristics towards higher cycles yields a superior fatigue load-bearing capability of the processed DP steel, which is evident until a stress level of about 280 MPa and around 8 × 10⁶ cycles where the two curves are intersecting as shown in Figure 6 (indicated with diamond).

3.3.2. Fracture Surfaces

SEM images of the fracture surface of the AR DP600 steel tested at stress levels of 500 MPa, 450 MPa, and 400 MPa are shown in Figure 7, Figure 8 and Figure 9, respectively. From these figures, fracture surface morphologies indicating fracture modes like crack initiation, crack propagation, and sudden rapture regions can be observed. From Figure 7, the fracture surface of the sample tested at 500 MPa shows a narrow region of fatigue crack propagation. In this region, the fatigue crack propagated without a considerable section reduction. Fatigue cracks were initiated at the sample surface and propagated through the thickness of the sample as shown in Figure 7a. In the crack propagation region, the fracture surface can be characterized by the formation of bulges causing the formation of a rough surface morphology. These features were indicated with arrows in Figure 7c. Bulges observed on the fracture surface of the fatigue crack propagation region seem to be formed when a crack reaches a harder martensite particle or a cluster propagates through the interface of the martensite–ferrite phases (Figure 7b,c) [22]. In the sudden rapture region of the sample on the other hand, the fracture surface morphology can mostly be characterized by the formation of dimples indicating that the fracture occurs by micro-voids formed by the accumulation of dislocations and rapidly coalescing (Figure 7d,e). The formation of the dimples may indicate that a fracture of the as-received sample occurred with a ductile fracture in the sudden rapture region after a plastic deformation accumulation (Figure 7e). This may also be understood from the considerable reduction in the sudden rapture region (Figure 7e). Reducing the maximum fatigue stress to 450 MPa and 400 MPa had no effect on fatigue crack initiation or growth, or consequently the fracture behavior of the DP600 steel (Figure 8 and Figure 9). In these stress levels, the fatigue fracture of the samples occurred by cracks initiated at the surface. The crack propagation region of the samples tested at 450 MPa and 400 MPa showed a rough surface morphology with the formation of the bulges as indicated in Figure 8b,c and Figure 9b,c with arrows. Comparing the fracture surface of the sample tested at 450 MPa (Figure 8d,e) with that of the sample tested at 400 MPa (Figure 9d–g) indicates an enlargement in the crack propagation region with the decreasing stress level. However, the morphological properties of the crack propagation surfaces reflect similar features. Similarly, in the sudden rapture region of samples tested at 450 MPa (Figure 8d,e) and 400 MPa (Figure 9f,g), the cross-section area reduction and dimple formation due to the coalescence of a high number of micro-voids dominated the fracture surface morphology.

SEM images of the fracture surfaces of the processed DP steel after fracture testing at stress levels of 800 MPa, 600 MPa, and 500 MPa are given in Figure 10, Figure 11 and Figure 12, respectively. The applied FSP considerably affected the crack initiation and crack propagation behavior of the DP600 steel. In general, the fatigue fracture of the processed DP steel is dominantly initiated at the processed surface (Figure 10a, Figure 11a and Figure 12a). It is an expected result considering the higher roughness and macro tool scars formed by the tool shoulder contact causing both residual stress and geometrical discontinuities causing stress concentrations. The fatigue crack propagation region of the sample tested at 800 MPa reflected a fracture surface dominated by periodic narrow steps through the propagation path (Figure 10b,c). Decreasing the maximum stress to 600 MPa and 500 MPa did not change this morphology. As indicated with arrows in Figure 11b,c and Figure 12b,c, the step-like crack propagation surface feature is evident, and both size and continuity are alike. However, the fatigue test stress levels affected the fatigue crack propagation paths. The fatigue crack propagation path formed after testing at high-stress levels like 800 MPa mostly occurred in a limited area/short distance before a sudden break. Decreasing the maximum stress applied during the fatigue test enlarged/prolonged the fatigue crack propagation region. It is obvious from Figure 11b–d that the fatigue crack propagation region of the samples reflected uniform characteristics and covered nearly 75% of the cross-section of the samples. The sudden fracture region of the processed samples generally reflected ductile fracture characteristics like dimples (Figure 10d,e, Figure 11f,g, and Figure 12f,g). This feature is similar to those observed on the fracture surfaces of the as-received material (Figure 7d,e, Figure 8d,e, and Figure 9f,g). However, the dimples formed on the surface of the processed samples were somehow shallower compared to those of the AR steel (Figure 7d,e, Figure 8d,e, and Figure 9f,g). This may be related to the decreased volume fraction of the ferrite phase contributing to the ductile characteristics of the stir zone.

The fatigue results obtained from the as-received and processed samples mainly indicated a considerable improvement in the fatigue strength of the sample in the finite life region with an upper limit of 2 × 10⁶ cycles. This improvement may be related to the enhanced strength of the steel by FSP. However, it is also obvious that the stress levels selected for the testing of the as-received DP600 steel and the processed counterpart have differences. In the case of testing the as-received steel, the selected stress levels are in the plastic deformation region. On the other hand, stress levels causing a fatigue failure of processed steel are in the elastic deformation region. This difference is a natural result when the endurance limit is selected to be the one (2 × 10⁶ cycles) that is widely accepted in seam weld testing due to the existence of metallurgical limitations like the heat-affected zone and geometrical discontinuities causing stress concentrations in the case of fusion welding seams. Similar limitations also exist in friction stir technologies, like microstructural transition zones and tool contact-based geometrical discontinuities or residual stresses. Consequently, limiting the endurance limit to 2 × 10⁶ cycles may be acceptable [27,28].

From another technological point of view, testing of the as-received DP600 steel in the plastic deformation range may also be a requirement because DP600 steel is developed to be stamped into parts with complicated shapes. Hence, the industrial application of the steel simply requires it to be used after a considerable plastic pre-strain [29]. The case of the application of plastic strain to DP steels results in a microstructural defect arising from the formation of tears at the ferrite–martensite interface due to the distinct flow behaviors of the micro-constituents. In our previous study, for example, the deformation instability of the DP600 steel was found to be primarily related to the deformation localization and tearing of the ferrite phase around the martensite particles/clusters [22]. In the case of fatigue behavior, tearing after plastic deformation may serve as initiated cracks and lead to the acceleration of fatigue failure. In the fracture surfaces of the as-received DP600 steels, bulge formation can be related to tears formed at the martensite–ferrite interface acting as easy crack prorogation sites (Figure 7, Figure 8 and Figure 9). Similar phenomena are also observed on the fracture surface of the processed steel with the formation of the steps due to the existence of lath martensite. However, due to the increased volume fraction of the martensite phase, stresses even lower than the yield strength of the processed DP steel may initiate and propagate fatigue cracks and lead to the formation of the step-like fatigue propagation surface morphology (Figure 10, Figure 11 and Figure 12). An increase in the volume fraction of the lath martensite phase and the stress concentration around it may also be responsible for a decrease in the exponent of Basquin’s law (C₂).

3.4. Impact Behavior

Contact force (F)-displacement curves (X) obtained from the drop-weight impact test of AR and processed DP steel are shown in Figure 13a. The contact force acting on the AR DP600 steel sample continuously increased until a striker displacement was reached at about 10.7 mm. At this striker displacement, the contact force acting on the tip of the striker took the peak value of 37.6 kN. In the case of impact testing, the processed DP steel also showed a continuously increasing curve characteristic and reached a peak load of 33.6 kN after a total striker displacement of about 6.8 mm. Considering the trend of the F-X diagrams obtained from the drop-weight impact tests shows that the contact force of the processed samples took higher values compared to that of the AR steel at a selected striker displacement lower than 6.8 mm. However, the allowable striker displacement for processed DP steel (6.8 mm) did not cause a fracture in the as-received steel, and the contact forces continued to increase through about an extra 4 mm of displacement. As a result, the plastic deformation continued to accumulate higher strains under the effect of the strain hardening of the AR microstructure which resulted in higher striker forces. This behavior resulted in AR DP600 steel exhibiting an impact energy absorption capability with a total energy of 239 J. In the processed state, however, this value was determined to be around 183 J. Macro views of the drop-weight impact-tested AR and processed samples are represented in Figure 14. AR samples have been plastically deformed into a dome shape during the test and fractured by cracks propagated circumferentially (Figure 14c). After FSP, the crack propagation behavior changed significantly. From Figure 14d, it can be understood that cracks initiate at the stir zone then propagate in the transverse direction and reach the TMAZ. In the TMAZ, the crack propagation direction shifts significantly to the longitudinal direction (Figure 14d). The impact crack propagation direction for the as-received steel is an expected result. In general, it is known that plastic instability of the thin plates deformed under a semi-spherical punch occurs at the free regions near to the punch-thin plate contact end [11]. The impact crack propagation path of the processed sample may be related to the mechanical behavior of the stir zone (Figure 5 and Table 2). In the testing of the processed DP steel, it is necessary to consider the displacement under the semispherical striker. For a selected level of striker displacement, the strain must be equal at each point on a circle whose center is aligned with the striker axis. Also, it is understood from Table 2 that the uniform elongation of the processed DP steel was determined to be 5.8% and 5.1%, respectively. Hence, to initiate a crack, punch displacement needs to generate a strain equal to the uniform elongation of PD and TD. To reach these values, the stress acting on the PD must necessarily be equal to the ultimate tensile strength because the ultimate tensile strength of the PD is higher than that of the TD. This means that higher strength in PD acts as a striker contact force source during the test. After the stress acting on the PD causes plastic instability by necking, cracks can easily propagate in both the TD and PD, leading to the formation of the fractured sample geometry shown in Figure 14.

4. Conclusions

This work used experiments to examine the effects of FSP on the mechanical characteristics, microstructure, and deformation behavior of DP600 steel under dynamic loading. The primary conclusions and findings of the research may be summed up as follows:

• With position-controlled FSP, thin (1.1 mm) DP600 steel sheets were successfully processed without macro-damage, cracking, or deformation discontinuities. During steady-state processing, the processing temperature reached 915 ± 20 °C;
• Through the intense thermo-mechanical treatment of FSP, the microstructural modification raised the hardness of the DP600 steel from 178 HV 0.5 to approximately 292 HV 0.5;
• It was found that the deformation behavior of DP600 steel is dominated by work hardening. This behavior remained unchanged after FSP. FSP also significantly improved the yield strength from 301 MPa to 811 MPa and the tensile strength from 621 MPa to 1054 MPa. However, the increase in strength was accompanied by a decrease in uniform elongation from 21.3% to 5.8% and elongation at break from 34.7% to 13.0% in the process direction;
• By applying FSP to DP600 steel, the AR fatigue limit was increased from 360 MPa to 440 MPa. FSP was found to be ineffective for the morphological characteristics of fracture surfaces;
• The finite-life fatigue fracture surfaces of the as-received samples were characterized by the formation of fine bulges due to the variation in the crack propagation path in the vicinity of the martensite particles/clusters. After FSP, the transformation of the martensite particles into coarser lath martensite also transformed the fracture surface into a step-like morphology;
• DP600 steel exhibited a high peak contact force of approximately 37.6 kN with a displacement of approximately 10.7 mm in the drop-weight impact test. FSP reduced these values to about 33.6 kN and 6.8 mm, respectively. The energy absorption capability of the DP steel decreased from 239 J to 183 J;
• FSP can be considered a practical tool for automotive lightweighting and part-specific feature development applications where DP600 steel is used due to its ability to provide simultaneous improvements in static and fatigue strength combined with an adequate impact energy absorption performance.