Abstract
An attempt was made for the first time to fabricate tailor-welded blanks (TWBs) through the feasibility of selective alloying in the solid state of vertical compensation friction stir welding (VCFSW) technique. The results revealed that the production of TWBs by VCFSW technique had higher homogeneity without any defects when compared to the production of TWBs by conventional FSW process at constant tool rotational speed 2000 rpm, traverse speed 20 mm min−1 and an inclination angle 2.5°. The maximum mechanical properties of the fabricated TWBs were obtained when the width of compensation strip 1.5 mm was used. The fracture surface of TWB at the width of compensation strip 1.5 mm showed ductile fracture mode.
Keywords
A comparative study was conducted between friction stir welding (FSW) without additional material and vertical compensation friction stir welding (VCFSW) with additional material. The use of additional materials by inserting strip of compensation material between two edges of the base metal (BM) assisted in eliminating the cavities and produced joints defects-free; which led to good quality of the weldments. The production of tailor-welded blanks (TWBs) by VCFSW technique had higher homogeneity without any defects. The quality of the welded joints by VCFSW depended on the ability of compensation material to fill in and mix with the BM through the welding process.
Introduction
TWBs are considered one of the important manufacturing methods used in many industrial applications such as automotive, rail transportation, shipbuilding and aerospace industries. This technology has various advantages, i.e. cost-effective, and it reduces weight and the number of forming dies. TWBs which are known as semi-finished products that can be produced by several geometrical alternatives such as different materials, strengths, thicknesses, surface conditions and grades [1]. Some errors of the manufacturing or workpiece variations between large structures to be welded by FSW process of TWBs appear as disturbances such as material tolerances, gaps and mismatches in welding zone at the interface between two workpieces in huge structural industries such as automotive, rail transportation, shipbuilding and aerospace industries, which are very difficult to be neglected because they reduce the quality of the welded joint.
Accordingly, we can focus on the welding defects and/or the volume defects such as groove, cavity, tunnel and specifically gap. Therefore, several researchers have investigated the influence of volume defects of friction stir-welded joints on microstructures characterisation and mechanical properties. Leonard and Lockyer [2] investigated the causes of various common flaws from FSW process, and they suggested possible corrective action. They found that the gap between any two plates to be welded constituted a major cause of the void defects. Similar gap defects were investigated by Wanjara et al. [3], Yang et al. [4], Shultz et al. [5], Widener et al. [6] and Threadgill et al. [7] they concluded that if the width of the gap increases; the maximum joint efficiency decreases. Finally, Seidel et al. [8] concluded that the gap caused multi problems such as (i) insufficient plasticised materials flow, (ii) loss of materials, (iii) a lack of root penetration and finally (iv) resulted in the reduction of the joint thickness. Consequently, the welding defects and/or volume defects such as groove, cavity, tunnel and gap significantly affect the quality of the welded joints in FSW. Therefore, some researchers have proposed and put forward new methodologies to eliminate the welding defects of FSW and got higher quality of the joint [9–12]. Many researchers have studied microstructures characterisation and mechanical features of TWBs fabricated by FSW process [13–18]. On the other hand, a few investigations have focused on VCFSW technique which was concentrated on temperature, microstructure and mechanical properties [19–21].
Till now, the influence of compensation strip width by VCFSW technique on the quality of the TWBs has still been rare. Hence, the present study has been performed on TWBs to evaluate the microstructural characterisation and mechanical properties. Finally, the main goal of the present study was divided into two main objectives; the first objective was the elimination of the welding defects, while the second objective was the enhancement of the joint efficiency of the TWBs through the feasibility of selective alloying in the solid state of VCFSW by adding the width of compensation strip of AA7075 as reinforcement material at the interface between two sides of substrates surfaces of AA2024 BM.
Experimental procedure
In this work, TWBs were produced by both FSW process (without additional material) and VCFSW technique (with additional material as compensation material at the interface between two sides of substrates surfaces of the BM), respectively, as shown in Figure 1(a,b). In the FSW process, wrought aluminium alloy AA2024-T6 with nominal thickness of 3.5 mm was chosen as BM that can be seen in Figure 1(a). However, in the VCFSW technique, high strength aluminium alloy AA7075-T6 was selected as compensation material which was added to fill the gap at the interface between the two sides of substrates surfaces of AA2024 BM as shown in Figure 1(b). The chemical composition and mechanical properties as well as thermal properties of both aluminium alloys were listed in Tables 1–3, respectively. The sheets of the BM were cut to have a length of 150 mm and a width of 100 mm (parallel to the rolling direction). On the other hand, the compensation materials were cut into 150 mm length. Five different values of the width of compensation strips were used with dimensions 1, 1.5, 2, 2.5 and 3 mm. The welding tool in this study was taper pin thread profile which was designed by the composing of a concentric circles flute shoulder with diameter of 15 mm and right screwed pin with length of 3.2 mm as shown in Figure 2. The width of compensation strip must be smaller than the diameter of the pin tip. Thus, the diameters of pin bottom and tip were 6 and 4 mm, respectively. The tool material in this work was made from high carbon steel K110 which was used significantly for FSW purposes. The tool was formed and then subjected to the standard heat treatment cycle of high speed steels to induce hardness of up to 62 HRC. The clamping system was designed especially in the present study as shown in Figure 3; where (A) top jig, (B) base, (C) top bolt and nut, (D) side bolt and nut, (E) clamping, (F) clamping holder and (G) screw bar. The design of the clamping system has two advantages; it reduces the distortion of both the thin BM and compensation material strip and the preparation time during welding process. The welding direction was parallel to the rolling direction of the BM. During the welding process, the inclination angle was 2.5° and the tool rotates clockwise. The tool rotational speed and traverse speed were constant at 2000 rpm and 20 mm min−1, respectively. The depth of plunging shoulder was 0.2 mm. The dwelling time of the plunging was 15 s, while the dwelling time during welding was 4 s.
(a) TWB without additional material by FSW process. (b) TWB with additional material by VCFSW technique. TWBs: (a) without additional material by FSW process and (b) with additional material by VCFSW technique. FSW tool. Clamping system. Chemical composition of AA2024 and AA7075 (mass %). Mechanical properties of AA2024 and AA7075; were measured experimentally. Thermal and processing properties of AA2024 and AA7075.
After the welding process, the fabricated joints were cut perpendicularly to the welding direction by an electrical discharge cutting machine to perform of both microstructural and mechanical characterisation. The microstructure analysis was conducted after the polishing and the etching of specimens with Keller's reagent on an optical microscope (OM, Olympus-GX71) and scanning electron microscope (SEM, SU3500 manufactured by Hitachi Company) equipped with an energy-dispersive X-ray spectroscopy (EDS) analysis system. Grain size was measured in the stirred zone and at the BM by technique of the image-analysing (J MicroVision program). The tensile, micro-hardness and bending tests were carried out to compare the strength, hardness and ductility of both friction stir-welded joints without additional material and vertical compensation friction stir-welded joints with additional material of TWBs. Tensile test at room temperature was performed at constant speed of 1 mm min−1 by universal testing machine. After tensile test, fracture positions of the specimens were observed using a stereoscopic microscope (ZSA403) and fracture surfaces were analysed by scanning electron microscopy. After that, Vickers micro-hardness test was measured with the load of 200 g at dwelling time 10 s and the interval between the two adjacent points was 2 mm. Finally, bending test at room temperature was performed at the experimental speed of 2 mm min−1 by a universal testing machine.
Results and discussion
Surface formation and macrostructural analyses
Figure 4(a–f) on the left side exhibits the formation of burrs (small flashes) at the welded surfaces in the stirred zones of all TWBs by both FSW process and VCFSW technique. Semicircular crown appearance was formed at the welded surface in the stirred zone of both friction stir welded and vertical compensation friction stir-welded joints can be seen in Figure 4(a–f) on the left side. On the other hand, Figure 4(a–f) on the right side introduces the classification of the different macrostructure zones in the cross section of both friction stir welded and vertical compensation friction stir-welded joints such as BM, thermo-mechanical-affected zone (TMAZ), heat-affected zone (HAZ) and nugget zone (NZ). From this macro-level assessment, all welded joints were defect-free except for the friction stir-welded joint. In this joint, the welded zone (WZ) especially NZ has contained groove-like cavity. This defect was termed as ‘surface tunnel’ or ‘void’. The tunnel or a cavity appeared at the advancing side (AS) of the stirred zone. When using the same process parameters of FSW (tool rotational speed of 2000 rpm, traverse speed of 20 mm min−1 and the inclination angle of 2.5°) through the welding process, we have inferred that the production of welded joints by VCFSW technique had higher homogeneity without any defects when compared to the production of welded joints by the conventional FSW process as shown in Figure 4(a–f). Therefore, the use of additional materials by inserting strip of compensation material between two edges of the BM assisted in eliminating the cavities and produced joints defects-free; which led to good quality of the weldments.
The surface formation (on the left side) and the macrostructure cross section (on the right side) of VCFSW with additional material with compensation strip width at: (a) 1 mm, (b) 1.5 mm, (c) 2 mm, (d) 2.5 mm and (e) 3 mm, where AS: advancing side, HAZ: heat-affected zone, NZ: nugget zone, RS: retreating side and TMAZ: thermo-mechanical-affected zone.
An important point has been noted was that the quality of the welded joints by FSW is depending on behaviour of material flow. Therefore, Elangovan et al. [22] classified the behaviour of material flow during FSW process into three main categories: (i) the insufficient material flow, (ii) the excessive material flow and (iii) the balanced material flow. Kim et al. [23] discussed the causes of the insufficient and the excessive material flow; in the excessive and the insufficient materials flow states the defects were easily created due to both of high rotational speed and high welding speed. In the present study, we have chosen both of high rotational speed of 2000 rpm and low welding speed of 20 mm min−1 in order to obtain balanced material flow. As a result, the balanced material flow was detected in welded joints by VCFSW technique while the insufficient material flow was detected in welded joint by FSW process. The insufficient material flow produced tunnel defect in the NZ at the cross section of friction stir-welded joint that can be seen in Figure 4(a).
Microstructural analyses
The microstructural characterisations of the stirred zones were illustrated in Figure 5. The microstructure of the stir zone plays a vital role in the quality welded joints. During the stirring action of the rotational tool, dynamic recrystallisation happened and fine grains can be observed in stirred zone as shown in Figure 5. The distribution of the reinforcement particles (additional material) in the stirred zone can be influenced by the processing parameters of FSW such as the tool rotational speed and welding speed. Microstructural observations showed the presence of the typical WZs, such as the NZ, the TMAZ and the HAZ. According to Figure 5, interfaces between the TMAZ and the stirred zone can be clearly observed. Overall, uniform distribution and very fine grains always lead to better strength and higher hardness. On the other hand, non-uniform grains of the stirred zone result in poor weld strength. According to Figure 5(a), we have inferred that non-uniform grains of the stirred zone were observed in friction stir-welded joint without additional material. However, uniform and refiner grains were observed in vertical compensation friction stir-welded joints with additional material that can be seen in Figure 5(b–f).
Microstructures of different regions of welded joints by optical microscope: (a) friction stir-welded joint without additional material and vertical compensation friction stir-welded joints with compensation strip width at (b) 1 mm, (c) 1.5 mm, (d) 2 mm, (e) 2.5 mm and (f) 3 mm, respectively. All regions were taken at the AS.
The grains size plays an important role at incorporation level and enhancement of the resulting properties (i.e. microstructural and mechanical) of welded joints. Figure 6 shows the variations of the grain size both of FSW process without additional material and VCFSW technique with additional material at different compensation strip widths. According to Figure 6, we have noted that an average grain size of the FSW specimens exhibited 6.7 µm, while an average grain size of the VCFSW at 1, 1.5, 2, 2.5 and 3 mm showed 6.1, 5.8, 5.9, 5.85 and 6.3 µm, respectively. We have inferred that the grain sizes were very close to each other. Therefore, the influence of grain size became less-effective. The reason behind the reduction of the grain size in NZ (although the grain sizes were very close to each other) was due to some agglomerations of AA7075 (Al–Zn) particles that were dispersed and distributed homogenously in the stirred zone. Thus, in case of VCFSW technique the more uniformly dispersed Al–Zn particles prevented and stood effectively as an obstacle in the way of the grain growth and grain boundaries.
Grain size in the NZ as a function of compensation strip width values (error bars the maximum and minimum grain size measured in each specimen).
According to spot EDS analysis, we have inferred that there are three kinds of particles or precipitates as shown in Figure 7. The first kind of precipitates were recognised as MgZn2 which were found in vertical compensation friction stir-welded joint at 2 mm as shown in Figure 7(d). These precipitates were characterised as susceptible to embrittlement because of micro segregation of MgZn2 particles which have led to catastrophic failure of joint. However, the second kind of precipitates were recognised as Al2Cu which were found in friction stir welded and vertical compensation friction stir-welded joints, as shown in Figure 7(a–f). These precipitates were characterised as strengthening precipitates. On the other hand, the third kind of precipitates were recognised as copper particles which were found in friction stir welded and vertical compensation friction stir-welded joints, as shown in Figure 7(a–f). These precipitates were characterised as un-dissolved precipitates.
Microstructures of stirred zone for welded joints by SEM: (a) friction stir-welded joint without additional material and vertical compensation friction stir-welded joints at (b) 1 mm, (c) 1.5 mm, (d) 2 mm, (e) 2.5 mm and (f) 3 mm, respectively.
Figure 8 shows FE-SEM micrograph and presents MAP analysis as well as EDS spectrum that were used for determining the distribution of the elements and chemical compositions in the structure of the vertical compensation friction stir-welded joint with the width of compensation strip at 1.5 mm. According to the scanning of the surface, the results showed that the distribution of the elements was homogeneous in the structure. It was observed that the aluminium covered whole the surface. The copper and magnesium showed the medium presence, whereas the zinc was found the least appearance. Consequently, we have inferred that the quality of welded joints by VCFSW depended on the ability of compensation material to fill in and mix with the BM through the welding process.
FE-SEM image, MAP and EDS analysis of the vertical compensation friction stir-welded joint with compensation strip width 1.5 mm.
Micro-hardness of the joints
In order to investigate the effect of the width of compensation strip by VCFSW technique of TWBs on hardness property, Vickers micro-hardness test was carried out in the mid-thickness. As a matter of fact, the grains size and particles formation plays a vital role to determine the mechanical properties of the welding zone. According to the Hall–Petch formula, the hardness property of the material was inversely proportional to the grain size [24]. Hence, the finer grains were formed by severe plastic deformation which has led to higher hardness values in the stirred zone than BM as shown in Figure 9. According to Figure 9 (in case of friction stir-welded joint), it was noticed that the hardness distribution of welded joint presented a typical W shape. Minimum hardness values were detected in FSW sample at the NZ. Nevertheless, lower hardness values of vertical compensation friction stir-welded samples were detected in the TMAZ and HAZ. Yet, the greatest hardness in the softened regions was obtained in vertical compensation friction stir-welded joints at the stirred zone with an approximate value of ∼124 HV. Consequently, the width of softened region (NZ) has been related to the VCFSW technique. Therefore, the micro-hardness of the vertical compensation friction stir-welded joint depended on the AA7075 contents rather than the AA2024 BM in the stir zone.
Micro-hardness profiles across the WZ in AA2024-T6 as a function of compensation strip width for VCFSW process.
Tensile properties of the joints
The results of the tensile test of TWBs with FSW and VCFSW were presented in Figure 10, Tables 4 and 5. Figure 10 shows the stress–strain tensile curves of both friction stir welded and vertical compensation friction stir-welded joints. The failures in both friction stir welded and vertical compensation friction stir-welded specimens were shear fracture accompanied by no obvious reduction in area, i.e. limited necking. Overall, the vertical compensation friction stir-welded samples exhibited higher strength and elongation than those in friction stir-welded samples. In friction stir-welded samples specifically, the ultimate tensile strength reached 221.98 MPa. The welding efficiency of friction stir welded without additional material was calculated 44.48% by law (UTSweld/UTSBM). On the other hand, in vertical compensation friction stir welded with additional material samples, the tensile strength and welding efficiency of vertical compensation friction stir welded with additional compensation strip width at 1 mm was 260.1 MPa and 52.52%, respectively. A higher tensile strength of 307.19 MPa was attained by vertical compensation friction stir-welded joint at 1.5 mm. while a lower tensile strength of 246.65 MPa was attained by vertical compensation friction stir-welded joint at 2 mm. The lower welding efficiency of vertical compensation friction stir-welded joint at 2 mm was 49.82%, while the upper welding efficiency of vertical compensation friction stir-welded joint at 1.5 mm was 62.06%. Consequently, the welding efficiency increased from 44.48% to 62.06% in VCFSW technique. When comparing the tensile test results of the vertical compensation friction stir-welded joints with those of the friction stir-welded joints, it was evident that the ultimate tensile strengths and elongations of the vertical compensation friction stir-welded joints with additional material were better than of the friction stir-welded joints without additional material as shown in Table 4.
Tensile curves of the friction stir welded and vertical compensation friction stir-welded samples; a representative stress–strain curve of the friction stir welded and vertical compensation friction stir-welded specimens is also reported for comparison. Tensile properties of TWBs produced by FSW process and VCFSW technique. Fracture locations of tailor-welded joints produced by FSW process and VCFSW technique.
From the assessment of the cross sections of the fractured samples, we have inferred that the fracture in friction stir-welded joint without additional material occurred in the centre of the NZ at the region that exhibited the least hardness as shown in Table 5. On the other hand, the fracture of the vertical compensation friction stir-welded joint with additional compensation strip width at 1.5 mm occurred in the HAZ (adjacent to the retreating side) at the region that exhibited the least hardness. Specifically, the failure behaviour of vertical compensation friction stir welded at 1.5 mm took place at almost 45° without obvious reduction in area. However, in the rest cases of vertical compensation friction stir-welded samples with additional compensation strip widths at 1, 2, 2.5 and 3 mm the fracture behaviour occurred in the TMAZ (adjacent AS) at the region that displayed the least hardness.
The fracture morphologies of both friction stir welded and vertical compensation friction stir-welded joints were described in Figure 11(a–f). We have inferred that the fracture morphologies of friction stir welded without additional material and vertical compensation friction stir welded with additional compensation strip width at 1, 2, 2.5 and 3 mm showed shear type fracture which was characterised by shallow dimples as shown in Figure 11(a, b, d, e and f). Nonetheless, the fracture morphology of the vertical compensation friction stir-welded joint at 1.5 mm was determined mainly by a ductile mechanism that was characterised by large and deep dimples as shown in Figure 11(c).
Fracture morphologies of friction stir-welded joint without additional material at (a) without compensation strip width and vertical compensation friction stir-welded joints with additional material at compensation strip widths: (b) 1 mm, (c) 1.5 mm, (d) 2 mm, (e) 2.5 mm and (f) 3 mm.
Bending properties of joints
In order to test the integrity of the TWBs joints, bending test was carried out according to International Organization for Standardization ISO 5173. In point of fact, if lack of root penetration, hole or tunnel defect emerges in the welded joint, it is very difficult to reach the bending angle of surface and root of welded joint till 180° [2]. The results of the three points bending test were presented in Figures 12–14. Figure 12 shows the applied bending loads as functions of the compensation strip widths. Also, Figure 13 presents the front and lateral views of bending test of vertical compensation friction stir welded at 1.5 mm. However, Figure 14 illustrates the bending angles as functions of the compensation strip widths. We have inferred that the maximum bending load (4.39 KN) occurred at maximum bending angle (180°) without any cracks in case of vertical friction stir-welded joint with compensation strip width 1.5 mm as shown in Figures 12–14. Nevertheless, the rest of the bending test specimens failed at the small bending loads (varies from 1.2 KN till 1.8 KN) and bending angles (varies from 25° till 60°) as shown in Figures 12 and 14, respectively. Finally the integrity of welded joints by VCFSW technique closely linked to the mixing and bonding of the compensation material (AA7075) with the BM (AA2024).
Bending loads of friction stir welded and vertical compensation friction stir-welded specimens. Bending test specimen under vertical compensation friction stir-welded joint at 1.5 mm (a) front view of bending test specimen and (b) lateral view of bending test specimen. Bending angles of friction stir-welded and vertical compensation friction stir-welded joints.
Conclusions
In the present study, TWBs were fabricated through the feasibility of selective alloying in the solid state of VCFSW technique with additional material as compensation material. Based on the present results, the main and new findings have been summarised as follows:
The production of TWBs by VCFSW technique had higher homogeneity without any defects when compared to the production of TWBs by conventional FSW process. The balanced material flow was detected in welded joints by VCFSW technique while the insufficient material flow was detected in welded joint by FSW process. The maximum mechanical characteristics of the fabricated TWBs were obtained when the width of compensation strip 1.5 mm was used. The fracture surface of the tailor-welded blank at compensation strip width of 1.5 mm showed the typical ductile fracture. The additional material or compensation material has played an important role in eliminating the cavities and produced more sound joints. The quality of the welded joints by VCFSW depended on the ability of compensation material to fill in and mix with the BM through the welding process.
Footnotes
Acknowledgements
The authors are thankful to the ministry of defense and war production; Helwan Company for Non-Ferrous Industries for providing the base material and compensation material for this research work. As well as Helwan Engineering Industries Company for converting the milling machine to friction stir welding machine.
Disclosure statement
No potential conflict of interest was reported by the authors.