Ultrasonic spot welding of aluminum alloys: A review

Journal of Manufacturing Processes 35 (2018) 580–594

Contents lists available at ScienceDirect

Journal of Manufacturing Processes journal homepage: www.elsevier.com/locate/manpro

Ultrasonic spot welding of aluminum alloys: A review Z.L. Ni a b

a,b,⁎

T

b

, F.X. Ye

School of Materials Science and Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450045, China School of Materials Science & Engineering, Tianjin University, Tianjin 300072, China

A R T I C LE I N FO

A B S T R A C T

Keywords: Ultrasonic spot welding Aluminum alloy Mechanical properties Macrostructure and microstructure Weld interface

As a solid joining technique, ultrasonic spot welding is a promising spot welding process to fabricate the aluminum alloy joints. This paper summarizes the current state of joining aluminum alloy by ultrasonic spot welding with numerous critical issues, such as general process parameters, materials flow, interfacial temperature, interfacial shear force, stress distribution, relative motion, strengthening mechanism, macrostructure, microstructure and mechanical properties. Meanwhile future trends in the field are pointed out.

1. Introduction Aluminum (Al) alloys have been widely used in the domains of power devices modules packing, electronic technologies, automobile body structure and wind and solar power controlling, due to their advantages of high specific strength, superior processability, predominant anti-erosion, high conductivity, environmental friendly character and recoverability [1–4]. Unfortunately, fusion welding processes employed in aluminum alloys are conductive to the generation of hot cracking, high levels of welding distortion and the poor weldability [5–8]. Resistance spot welding is difficult to join aluminum alloys due to their high conductivity, low strength at elevated temperature and tendency to degrade the electrodes [9]. Friction stir spot welding has disadvantage of long weld cycle (e. g. 2–5 s) [10,11]. However, ultrasonic spot welding (USW) has advantages of pollution-free, high efficiency, short weld time (typically < 0.5 s), insensitivity to material conductivity and heterogeneity [12,13], which is a promising spot welding process to fabricate the aluminum alloy joints. There are numerous studies about aluminum alloy fabricated by USW. However, to the best of the authors’ knowledge, no review papers about USWed aluminum alloy joints exist. This paper reviews the USW process, macrostructure, microstructure, parameter optimization and mechanical properties of aluminum alloy joints, and then future trends and conclusions are given. The aim of this paper is offer a good basis for follow-on study. 2. USW process Since the 1950s USW, a solid phase joining technique, was first

⁎

introduced for thin foils joining, wire bonding and tube sealing [14]. Due to the advancements in the welding system technology [15], it is feasibly used to join thicker metal sheets, but generally the thickness of sheet is less than 3.0 mm. Because when the thickness of metal sheet is over 3 mm, the heat generation and relative motion at the weld interface is poor. Thus increasing the power of USW machine is urgent in the future. USW utilizes high shear frequency mechanical vibration to generate a friction-like shear relative motion between two faces of the sheets, resulting in plastic deformation and shearing of surface microasperities that transmit contaminants and oxides, to promote the generation of effective metal-to-metal contact areas and joining at the faying surfaces. Therefore, a solid state joint is achieved. In general, only the lapped joint can be successfully achieved by USW. Schematic of illustration of USW is shown in Fig. 1. 2.1. General process parameters Macrostructure, microstructure evolution and mechanical properties of the USWed joint are mainly determined by the employed process parameters. The process parameters during USW involve ultrasonic frequency, vibration amplitude, clamping force, power (P) and energy (U) or time (t). Ultrasonic frequency, generated by transducer that is designed to operate at a specific frequency, is 20, 30 or 40 kHz in the metal welding machine. In fact, welding frequency is determined by a basic metallurgical physics characteristic as welding power requirements, which is determined by the component dimensions and materials of base metal, and the overall design of transducers and coupling components. Vibration amplitude of the sonotrode tip is a critical parameter affecting the joint quality, which can transmit mechanical

Corresponding author at: School of Materials Science and Engineering, North China University of Water Resources and Electric Power, Zhengzhou 450045, China. E-mail address: [email protected] (Z.L. Ni).

https://doi.org/10.1016/j.jmapro.2018.09.009 Received 4 July 2018; Received in revised form 24 August 2018; Accepted 7 September 2018 1526-6125/ © 2018 Published by Elsevier Ltd on behalf of The Society of Manufacturing Engineers.

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 1. Schematic of illustration of USW.

Fig. 3. SEM of weld cross-sections for the USWed joint [12].

Fig. 2. Relationship among peak power, weld energy and weld time [16].

throughout the weld interface occurs, leading to a wave-like displacement at the weld interface combined with complex plastic deformation that has been demonstrated as the formation of wave characteristics near the weld interface. The wave characteristics contain vortices, ripples, and spiral like patterns, as shown in Fig. 3. In addition, Jahn et al. [17], Allameh et al. [18] and Haddadi et al. [19] also gave the detailed descriptions about wave characteristics for understanding the material flow. It is noted that when the base metal sheets are similar or have few differences, the wave characteristics can occur. If the base metal sheets have big differences in melt point, yield strength and hardness, the wave characteristics can not be presented, thus the material flow will be not obvious, as reported in the previous studies [8,20–27]. Material flow is extremely complex. Not only welding parameters play a key part, material types and dimensions bring about another challenge in governing the migration of the materials at the weld interface.

energy to the weld interface. It is noted that vibration amplitude is generally in the range of 10–100 μm. In some welding machine, vibration amplitude is a dependent variable that has a relationship with the welding time or energy applied to the machine. In other machines, vibration amplitude is an independent variable; impedance is being set up and regulated by the supply of power due to the added characteristics of the feedback control system. The selection of welding vibration amplitude hinges on the welding conditions as determined by materials. Clamping force is a crucial parameter during USW, which is exerted on the metal sheets by the anvil and sonotrode tip, thus the metal sheets being welded can be pressed firmly together. The selection of clamping force depends on the materials being welded. The optimum clamping force can be achieved by adjusting welding parameters, below which joints will be weak to nonexistent, and above which the phenomena of welding zone thinning and sonotrode sticking may take place. It is noted that USW machine can be set up to operate in time or energy control mode, thus energy and time are approximately interchangeable; i.e. as U = P×t. For instance, 1 kJ at 4 kW is equal to 0.25 s. Fig. 2 shows the relationship among peak power, weld energy and weld time. The area under the power curve is weld energy. It can be seen that power, energy and time are not independent. Once the power is set, as the weld process progresses, the given level of power can be reached, meanwhile weld energy and time will reach a specific value. Or, time can be set, and the weld will progress until such energy as the given level is obtained. In fact, the power-time curve can present various forms that are dependent on material categories, dimensions, surface conditions, welding parameters (amplitude and clamping force), tooling, and specific characteristic of a fixed welding machine.

2.3. Interfacial temperature Temperature of weld interface is a key factor to the weld formation. Because appropriate weld interface temperature is conductive to improve the weldability via lowering the material yield strength and controlling the intermetallic compounds [4,28]. During USW, the heat generation, leading to the increase of weld interface temperature, are achieved from plastic deformation heating, frictional heating at the weld interface and possible acoustic heating from the ultrasonic wave [29,30]. The frictional heat generation is due to the large vibration amplitude between the faying interfaces. The vibration frequency is set at 20 kHz by the spot welder, increasing amplitude leads to the increase of velocity, which can have a positive effect on the decrease of yield strength of Al alloy sheet, the local Al alloy sheets become softer, and then severe plastic deformations take place under the action of shear force and clamping pressure. In addition, as the welding process progresses, the yield strength of the Al alloy is continuously lowered due to

2.2. Materials flow Once the welding parameters are optimized, underlying material flow under the combined action of the sonotrode tip and anvil 581

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

the heat generation by the plastic deformation. The alloys bear the cyclic deformation with a frequency of 20 kHz. It takes place with a predicted strain rate of ～103 within a welding time of less than 1 s, the material is subjected to 20,000 deformation cycles with a high strain of 1000. Thus numerous dislocations and excess vacancy concentration are presented in the weld interface region that is ascribed to the severe plastic deformation and high strain rate dynamic deformation [31,32]. In addition, the ultrasonic wave can be validly transmitted in the regular metal lattice, but it has a tendency to be absorbed in the microstructure defects, such as dislocation and vacancy [32]. Preferential absorption of acoustic energy in the material defects promotes the Al alloy soften, and reduces the obvious shear stress of Al alloy. Specially, dislocations are activated by the absorbed acoustic energy and glide with the improved mobility of the atoms, the material flow becomes easier, leading to the generation of severe plastic deformation. Meanwhile severe plastic deformation can further in turn conduce to the increase of temperature ascribed to the dissipation of deformation heating. Considering the heat generation from deformation and friction, Elangovan et al. [33] simulated the temperature distribution in weld interface for the aluminum joint using ANSYS software. The result is shown in Fig. 4. It is demonstrated that the peak temperature of weld interface is 336.18 ℃, which is in the center of the weld interface, and the heat effected zone is most in the deformation area that is below the sonotrode tip. In addition, it is shown that temperature difference between weld interface and top surface of aluminum sheet is about 58 ℃ along the vertical direction. Weld interface temperature has a proportional relationship with the weld time and friction coefficient of aluminum sheet surface, an inverse relationship with the clamping force and the thickness of aluminum sheet. Ngo et al. [34] have proposed an inverse algorithm based on the Broyden–Fletcher–Goldfarb–Shanno (BFGS) method, which can quickly and accurately figure out the unknown time-dependent heat generation at the weld interface and transfer coefficient of convection heat during USW when the temperature data of some locations can be obtained by making measurements. Therefore it can be used to the optimization of weld process parameters. Few studies about the measured weld interface temperature during USW are reported, due to the rapid weld time and small size of the joint. For example, Bakavos et al. [12] measured the weld interface temperature employing 0.5 mm diameter K-type thermocouples placed as close as possible to the weld interface center. The results are shown in Fig. 5. It can be observed that with increasing weld energy, the weld

Fig. 5. Relationship with the weld interface temperature and weld energy [12].

interface temperature increases. For the optimum weld energy (1000 J), the weld interface temperature reaches 391 ℃, which can lead to a significant loss of yield strength of the base metal (6061aluminum alloy). For instance, the yield strength of 6063 aluminum alloy deceases to 14 MPa when the temperature is about 370 ℃, which can enhance the weldability [35]. In addition, the temperature of weld interface quickly increases to the maximum value in less than 1 s with a high heating rate of 1000 K/s. Then the temperature decreases rapidly, due to the low input energy and good heat conductivity of 6061 aluminum alloy. Similar results were reported by Chen et al. [30], Zhang et al. [36] and Haddadi et al. [19]. In general, the peak temperature measured at the weld interface center is 50–130 °C higher than that at the weld edge for different weld time, and it is agreement with the simulated results reported by Elangovan et al. [33,37]. For the 1 mm-thick A5052-H32 alloy sheet joint, with increasing vibration amplitude, the temperature of weld interface increases, and the maximum temperature of weld interface is 455 ℃ at a 70% vibration amplitude and weld time of 0.8 s, as stated by Shin et al. [38]. The peak temperature of the thermal field at the top aluminum 6111-T4 sheet measured by employing a high frequency infrared camera is shown in Fig. 6, as demonstrated by Haddadi et al. [19]. It can be observed that the hottest area is close to the sonontrode tip. To precisely measure the temperature inside the weld interface, the interior weld interface is required to directly expose to the infrared camera, as proposed by De Vries [39]. Schematic of weld interface temperature measured by infrared methods for the asymmetric situation is shown in Fig. 7, thus the temperature that takes place at the weld interface can be measured directly. However, edge plastic deformation during USW cannot be thoroughly eliminated, thus the measured temperature is lower than that in a symmetric weld situation. It was indicated by Watanabe et al. [40] that the temperature of the weld interface was enhanced by the addition of ethanol to the weld interface. The reasons can be summarized as follows. Ethanol with polarized terminals can physically absorb to the surface of base metal sheet, leading to a sound lubricating action at the interface, and the relative motion on the two sheets is improved, thus the temperature of the weld interface was enhanced. In another study, it was reported by Ni et al. [4] that the addition of Al2219 particle between the two sheets can improve the coefficient of the weld interface, thus the weld interface temperature can be enhanced from 335 ℃ to 402 ℃. Effect of tool edge geometry on the temperature near the edge/sheet interface was investigated by Komiyama et al. [42]. For the joint with sonotrode tip with serrated edge, the temperature near the edge/sheet interface is significantly higher than that of the joint with sonotrode tip with trapezoidal edge at the clamping fore of 588 N, because the vibration amplitude is enhanced by sonotrode tip with serrated edge. 2.4. Stress distribution

Fig. 4. Temperature distribution for the aluminum joint with clamping force of 1600 N and weld time of 0.5 s [33].

During USW, the base metal sheets bear clamping force, high 582

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 6. Thermal imaging of the top sheet temperature field for the 0.92 mm-thick aluminum 6111-T4 sheet joints with different weld time [19].

vibration frequency and non-uniform temperature field, thus stress is generated in the USWed joint. The Von-Mises stress distribution in the anvil, sonotrode tip and aluminum sheets is shown in Fig. 8, as simulated by Elangovan et al. [33]. It is demonstrated that the aluminum sheets is moving away from the anvil, due to the inappropriate clamping method for holding the aluminum sheets on the anvil, the maximum Von Mises stress is 0.5 × 107 N/m2. In addition, it is stated that, with increasing clamping force, the stress at the weld interface increases, and the maximum stress is in the center of the sonotrode tip. It is confirmed by De Vries [39] that with increasing thickness of base metal sheet, the peak contact stress at the weld interface remarkably decreases. For the metal sheet with thickness up to 1.2 mm, the compressive stress distribution is almost uniform over the deformation zone area; for the thicknesses 2 mm and above, the peak compressive stress remarkably decreases from the center of the weld to the sonotrode tip edge. In another study, the von Mises stress distribution along the longitudinal symmetry plane for a single and double spot joint is demonstrated in Fig. 9 as simulated by Carboni et al. [32] using the ABAQUS software. For the single spot joint(Fig. 9a), the stress distribution (all typical characteristics of a crack tip bear mixed mode I + II) has no difference between the two spot tips, leading to be antisymmetric with reference to the sheets interface plane. The behavior is agreement with the expected result because the joint itself is inherently antisymmetric. For the double spot joint (Fig. 9b), the stress distribution between the two spot tips is markedly different. Specially, the tip towards the other spot (the left one in the figure) bears a significantly lower stress, demonstrating an affect between the two spots and the need to investigate and optimize a multiple joint configuration. Stress distributions for the joint with sonotrode tip edge angles of 135° and 170° are shown in Fig. 10 when the sonotrode tip edge moves toward the right (t = 20 ms), as demonstrated by Sasaki et al. [43]. It can be

Fig. 8. Stress distribution for the aluminum sheet joint with clamping force of 1600 N and weld time of 0.5 s [33].

observed that the stress distributions are unsymmetrical, the stress near the right edge is much higher than that in the left for the both edge angles, and the stress concentration in Fig. 10a is higher than that in Fig. 10b. Thus severer plastic deformation in the top sheet in Fig. 10a takes place in comparison with that in Fig. 10b, leading to the generation of larger gap between the top sheet and the bottom sheet in Fig. 10a, as demonstrated by the arrows in Fig. 10a. The result is also reported by Elangovan et al. [33], as shown in Fig. 8. Therefore, the sonotrode tip edge angle is a key factor to the USWed joint quality, and

Fig. 7. Schematic of weld interface temperature measured by infrared methods for the asymmetric situation [39]. 583

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 9. Von Mises stress distribution at spot borders along longitudinal symmetry plane (a) single spot joint; (b) double spot joint [32].

Fig. 10. Stress distribution at a vibration time of 20 ms for different sonotrode tip edge angles [43]. 584

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 11. Schematic of equipment for interfacial shear force measurement [39].

Fig. 12. In-situ PDV velocity measurement equipment with a zoomed-in view demonstrating the focus of laser spots (0.09 mm in diameter) on sonotrode tip, top and bottom foils [3].

Fig. 13. Indentations on the surface of top sheet for the joint with different weld time [46].

involves three stages: the initial stage, the middle stage and the final stage. In the initial stage, the interfacial shear force rises quickly. In the middle stage, the interfacial shear force has a slow increase rate. In the final stage, the interfacial shear force has remained stable.

it is need to make an intensive study to achieve an optimized sonotrode tip edge angle.

2.5. Interfacial shear force

2.6. Relative motion

Interfacial shear force that can be over 10 times higher than the weld strength is achieved by numerical simulation, as stated by Takahashi et al. [44]. However, few experimental results are reported. For example, the interfacial shear force can be measured in the combined action of shear force sensor and charge amplifier, as demonstrated by De Vries [39]. The schematic of equipment for interfacial shear force measurement is shown in Fig. 11. Similar principle of interfacial shear force measurement is reported by Ando et al. [45]. From the experiment results, it can be concluded that, increasing the ultrasonic power don’t always result in a sound bonding process; interface friction coefficient can be dramatically high due to the formation of welded region at the weld interface; with the increase of welded region, the interfacial shear force increases. In addition, the welding process

Relative motion in USW of pure aluminium sheets was investigated using a high speed camera, as demonstrated by Sasaki et al. [46]. The ultrasonic welding process can be divided into three stages. In the first welding stage, the top aluminum sheet first vibrates under the action of sonotrode tip, together with the slippage. At this stage, the bottom aluminum sheet remains static, thus a large relative motion takes place between the top sheet and the bottom sheet. In the second welding stage, the relative motion is constrained by the formation of some welded areas at the weld interface, while relative motion takes place between the sonotrode tip and the top sheet. In the third welding stage, with the increase of weld interface temperature, plastic deformation 585

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 14. The size of indentations on the surface of top sheet for the joint with different weld time [46].

Fig. 16. Optical images of 6111 aluminum joint cross-sections with different weld energy (or weld time) [12].

pull-out. If the welding energy is too high, stage 4 occurs, the velocity of the top sheet and the bottom sheet decreases quickly. Meanwhile the phenomena of sonotrode tip sticking (sticking between the sonotrode tip and the top sheet) and weld zone thinning take place, and the welded areas in stage 3 are destroyed, both of which have poor effects on the joint strength. 3. Macrostructure and microstructure 3.1. Macrostructure

Fig. 15. Phenomena of sonotrode tip weld and weld zone thinning [31].

During USW, with increasing weld time, the deformed aluminum gradually flows into the valley areas of the sonotrode knurl pattern, thus the depth that sonotrode tip penetrates into the top aluminum sheet becomes deeper, as shown in Fig. 13. Meanwhile, the size of indentation on the surface of the top sheets increases, as indicated in Fig.14. The indentation on the surface of the bottom sheet is similar with the indentation mentioned above, as stated by Lu et al. [3]. When the weld time, clamp force or vibrate amplitude is too high, the phenomena of sonotrode tip sticking and weld zone thinning take place, and the aluminum on the surface of top sheet is squeezed out, as demonstrated in Fig.15. Similar results were also reported by Peng et al. [47], Shin et al. [38], Watanable et al. [48], Jahn et al. [17] and Zhang et al. [36]. From the weld cross section in Fig.16, it is demonstrated that with increasing weld time, the increased depth of indentations takes place in both side of the joint, the weld linear density increases, the effective thickness of sheet decreases, and the plastic deformation at the weld interface becomes severer. In addition, the formation of welded area initially occurs heterogeneously at specific regions under the sonotrode tip, and then the welded area spreads outwards to the edge of the joint, as well as inwards, until it expands across the entire weld interface.

takes place in the top aluminum sheet and the bottom aluminum sheet. The formation of welded areas occurs in the first welding stage, grows in the second welding stage, and the joint strength reaches a maximum value in the third welding stage. The relative motion between the sonotrode tip and the top aluminum sheet significantly promotes the increase of temperature at the weld interface and accelerate the growth of welded areas. Unfortunately, the longest measurement time is only 600 ms, due to focal limitations of the high speed camera. Therefore it can not detect the entire USW process. In another study, relative motion during USW of aluminum alloy is investigated employing in-situ Photonic Doppler Velocimetry (PVD) velocity measurement method, which does not have the disadvantage of measurement time limitation of the high speed camera, as reported by Lu et al. [3]. Fig. 12 shows the in-situ PDV velocity measurement equipment with a zoomed-in view demonstrating the focus of laser spots (0.09 mm in diameter) on sonotrode tip, top and bottom foils. From the velocity features of sonotrode tip, the top sheet and the bottom sheet, the welding stages can be divided into four stages: stage 1 (slip stage), stage 2 (slip–stick transition stage), stage 3 (stick stage), and stage 4 (over-welding stage). A large relative motion between the top sheet and the bottom sheet takes place at the stages of 1and 2, leading to disperse the contaminants and oxides on the surface of aluminum sheet and promote the formation of effective nascent metal contact area. Stage 3 is a critical stage for the growth and spread of the welded area, which is beneficial to enhance the joint strength and change the joint failure mode from interface to button

3.2. Microstructure The cross section of USWed joints involves three zones: weld zone, weld affected zone and compression zone, as stated by Allameh et al. 586

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 17. (a, c, d, e) Inverse pole figure maps and (b, f, g, h) corresponding band contrast and grain boundary maps from high-resolution EBSD, (a, b) original material, and the sheets with different deformation reduction (c, f) 30%, (d, g) 40% and (e, h) 50% [49].

Fig. 18. Orientation distribution functions for (a) original material and the sheets with different deformed reduction (b) 30%, (c) 40% and (d) 50% [49]. 587

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 19. Relationship of hardness of USWed aluminum joint with ultrasonic amplitude (a) and deformation reduction (b) [49].

Fig. 20. Hardness distributions from cross-sections through weld centre, at half the top sheet thickness, for two different weld energy: (a) immediately after welding (measured within 1 h) and after post-weld room temperature natural ageing for (b) 2 weeks and (c) 8 months [30].

advances in science and technology of equipment, the doubt about the relatively equiaxed smaller grain mentioned by Allameh et al. [18] can be completely resolved. Ji et al. [49], Haddadi et al. [19], Xie et al. [50], Peng et al. [47], Lu et al. [3] and Mirza et al. [51] investigated the microstructure evolution of the USWed aluminum joint by using electron backscattered diffraction. From the analysis of the inverse pole figure maps corresponding band contrast and grain boundary maps (Fig. 17), orientation distribution functions (Fig. 18) and grain size distribution, it can be concluded that the relatively equiaxed smaller grains mentioned by Allameh et al. [18] take place due to the continuous dynamic recrystallization, and the nucleation and the growth of

[18]. For the compression zone, elongated grains exist in the matrix far away from the weld zone. For the weld affected zone, grains are deformed and crushed by compressive and shear forces operating on the weld interface during USW. The weld zone shows no resolvable grains, rather a matrix with uniform contrast decorated with trajectories of material flow that appears as strings of small particles. The size of grains with crystalline structure is 500–1000 nm. It is noted that the grain is not elongated. The reasons may be summarized as below. The larger elongated grains in the base metal are crushed into relatively equiaxed smaller grains, or new grains generates from old crushed grains, depending on the weld interface temperature. With the 588

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 21. Schematic of tensile shear (a) [4], U-peel (b) and T-peel tests (c) [55]. Table 1 Tensile strength of various Al/Al joints fabricated by USW. Base metal

Thickness of sheet (mm)

Weld energy (J)

AA6061-T6 Al6111-T4 AA6111-T4 AA6111-T4 Al6111-T4 AA6022-T4 Al5754-O Al A5052-H32

0.4 0.92 0.9 0.9 0.92 1.2 2 0.3 1

250 750 800

A6061-T6

Weld time (s)

Clamping force

Vibration amplitude

Tensile strength

Failure mode

Ref.

1754 N 40 MPa 100 psi

50 μm, constant

Nugget pull-out Nugget pull-out Nugget pull-out Interface failure Interface failure Nugget pull-out Partial button pull-out

[3] [12] [17] [18] [19] [31] [36] [37] [38]

2.5 0.6

1.4 kN 910 N 1.64 kN 2.0 bar 3.1 kN

35 μm, constant 6 μm, constant 45 μm, constant 60%, constant

1.35 kN, TS 3.5 MPa, TS 3.1 kN, TS 68.5 MPa, TS 2.9 kN, TS 1871 N, TS 4.7 kN, TS 3.29 MPa, TS 2.8 kN, TS

1

1

1176 N

53 μm,constant

943 N, UP

AA1050 AA1050-H24 Al6022-T43 A6061-T6 Al5754-O A1050/A5052

3 0.5/3 1.3 1 1.5 1.5

2 0.3

53 μm, constant 31 μm, constant

2000 400

882 N 588 N 0.4 MPa 1147 N 0.414 MPa 0.2 MPa

560 N, UP 120 N, TP 94 MPa, TS 1300 N, UP 85 MPa, TS 3 kN, TS

Al AA5754-H111

0.4 3

3500

2.5 bar 3360 N

AA5754-H111

3

4500

3400 N

Al6022-T4

1.2

0.1 0.3 1.4 5500

1400 1.5

2

1.2

1170 N

53 μm, constant 80 μm, constant 57 μm, constant 60 to 43 μm, amplitude profiling 60 to 43 μm, amplitude profiling 40 μm, constant

Mixed fracture pattern of shear and pull-out Fracture occurs in the base metal Nugget pull-out Nugget pull-out Nugget pull-out Nugget pull-out Interface failure Fracture occurs in the base metal

[40] [42] [43] [47] [48] [51] [56]

21.4 Mpa, TS 8 kN, TS

[57] [59]

8 kN, TS

[61]

4125 N, TS

Interface failure

[62]

at the weld interface, rod-like T phase (Al20Cu2Mn3) in the substrate 2024-T3 Al is crushed into nano-size dispersoid (< 70 nm) at the edge of the weld interface, and then the broken dispersoid is dissolved and disappeared at the weld interface. In addition, the amount of point-like type θ phase (Al2Cu) precipitates decreases.

grain never occur. In another study, it was stated by Lu et al. [3] that continuous dynamic recrystallization is ascribed to the severe plastic deformation and high weld interface temperature, and the increase of recrystallizion regions can lead to the increase of joint strength and a change of failure mode from interfacial failure to button pull-out. In addition, fragmentation and dissolution of the Al-Cu-Mg alloys precipitates during USW process was investigated by Xie et al. [50]. It is concluded that due to shear plastic deformation and high temperature 589

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

hardness. The more absolutely the recrystallization process occurs, the softer the sheets are. In addition, the hardness in the middle sheet is the lowest, as shown in Fig. 19a, due to the weak effect of cold working by the deformation. As the deformation reduction increases, the hardness in the sheet increases, as seen in Fig. 19b, due to the strengthening of grain refinement attributed to the recrystallization. In another study, the hardness of USWed aluminium 6111-T4 sheet was reported by Chen et al. [30], as shown in Fig. 20. Compared with the hardness of base metal, a lower hardness area occurs in the centre of the weld zone for both weld energies. Moreover, for the joint with lower weld energy, the hardness distribution has a U-shaped profile and softening extends only slightly outside of the sonotrode tip area; for the joint with higher weld energy, the softened area extends farther away from the sonotrode tip area, resulting in the generation of a W-shaped profile with a fairly uniform central plateau across the sonotrode indentation. The similar W-shaped hardness profile can be observed in the friction stir welded joints [41,52]. When the joints are subjected to natural ageing with 2 weeks, the hardness for the both joints reaches a higher level than that of the base metal. It is noted that when the joints are subjected to natural ageing with 8 months, the level of hardness for the both joints increases continuously. The reasons can be summarized as follows. The loss of hardness observed under the sonotrode tip is ascribed to cluster/GPZ dissolution. During USW, deformation-induced excess vacancy concentration that is generated due to the high strain rate dynamic deformation is about 100 times higher than that in a conventionally solution-treated and quenched material, which is conductive to enhance the post-weld ageing, leading to the existence of more fine precipitates in the weld regions. Therefore the hardness increases after post-weld room temperature natural ageing for 2 weeks and 8 months.

Fig. 22. Relationship of lap shear strength with welding energy [47].

4.2. Tensile strength and failure mode Tensile strength is used as a criterion to assess the quality of the USWed joint, which can be measured under quasi static loading conditions involving tensile shear (TS), U-peel (UP) and T-peel (TP) tests [53,60]. The schematic of tensile shear, U-peel and T-peel tests is shown in Fig. 21. For the spot joints, failure mode is employed as a criterion to assess the mechanical properties and an indicator of load bearing and energy absorption capacities. In general, the failure modes involve three types, such as interfacial failure mode, partial interfacial failure mode, and button pullout failure mode [53]. In the case of interfacial failure mode, the fracture spreads through the weld interface, leading to separation of the base metal sheets. For the button pullout failure mode, the fracture first spreads along the weld interface and then redirects perpendicular to the centerline towards the thickness direction. For the pullout failure mode, the fracture may initiate in the base metal or heat effected zone [54]. Tensile strength of various Al/Al joints fabricated by USW is shown in Table 1. Fig. 23. (a) Relationship among weld strength, pressure and weld time; (b) Relationship among weld strength, amplitude and weld time; (a) Relationship among weld strength, pressure and amplitude [57].

4.2.1. Effect of welding parameters Lap shear tensile strength of USWed 6022-T43 aluminum alloy joints was investigated by Peng et al. [47]. Fig. 22 shows the relationship of lap shear tensile strength with welding energy. It is observed that with increasing welding energy, lap shear tensile strength first increases, and then decreases. With increasing welding energy, the temperature of weld interface increases, leading to metallurgical adhesion and mechanical interlocking across the weld interface. Therefore, when the welding energy is 1400 J, the lap shear tensile strength reaches the maximum value. Higher weld interface temperature leads to the material soften, and severe stress concentration takes place at the edge of the sonotrode tip indentation, which is a weak region, thus the failure mode of the joint at the welding energy of 1200 J or 1400 J is nugget pull-out, as indicated in the insert figure in Fig.22. When the welding energy is higher than 1400 J, the depth of sonotrode tip penetrating into the top 6022-T43 aluminum sheets is deeper, severer

4. Mechanical properties 4.1. Hardness Because of the significant changes of aluminum microstructure during USW, the hardness of aluminum sheet changes. Few studies about hardness of aluminum sheet during USW were reported. For example, it was demonstrated by Ji et al. [49] that with increasing ultrasonic amplitude, the hardness of 1100 aluminum sheet decreases, as shown in Fig. 19a, due to the occurrence of dynamic recrystallization in the welded sheets. Recrystallization and cold working by deformation that have a completed relationship are key factors to the level of 590

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 24. Schematic of fracture modes of USWed 1.0 mm thickness A5052-H32 alloy joint [38].

appropriate weld time depends on the vibration amplitude, as indicated in Fig. 25. For example, shorter weld time is desired for higher vibration amplitude to promote the generation of enough weld interface temperature and formation of more effective welded areas, and to control the effective thickness of the joint. For the same reason, longer weld time is desired for lower vibration amplitude. Similar results was also reported by Imai et al. [58]. Effect of weld energy on the lap shear failure load of USWed AA6111-T4 joint was investigated by Jahn et al. [17]. When the weld energy is above 500 J, weld strength that arrives to a plateau of 2.6–3.1 kN is insensitive to the weld energy. In the case of weld energy below 400 J, the fracture location is in the weld interface, whereas pull out failure mode is predominant. In another study, comparison of control algorithms for USWed aluminum alloy was reported by Baboi et al. [59]. It is found that the weld strength variance of USWed joints fabricated in the time mode is smaller than that of joints fabricated in the modes of height and energy. Fig. 25. Combined actions of weld time and vibration amplitude on the lap shear failure load and failure energy [38].

4.2.2. Effect of welding tip geometry Effects of horn and anvil surface tip patterns (in Fig. 26) on the lap shear tensile strength of USWed 1 mm thick A5052-H32 alloy joints were investigated by Shin et al [38]. When the weld time and vibration amplitude is 0.6 s and 50%, under the combined actions of horn A and anvil A, the highest lap shear tensile strength (99 MPa) was obtained. In another study, it was stated by Janh et al. [17] that effect of anvil geometry on the weld strength of USWed AA6111-T4 is only slight. For instance, the average lap shear failure load for the joints with cap CT-4 anvil is about 10% higher than that for the joints with Cap-139 anvil. Because the combination of sonotrode tip and cap CT-4 anvil can prevent the weld zone thinning and enhance the temperature, plastic deformation and wavy morphology formation at the weld interface. It is reported by Komiyama et al. [42] that when the weld time and clamping force are 2000 ms and 588 N, respectively, the weld strength (U-peel test) for the joint with sonotrode tip surface geometry of serrated edge (560 N) is 95.8% than that for the joint with sonotrode tip surface geometry of trapezoidal edge (286 N). Meanwhile the fracture types of the joint with sonotrode tip surface geometry of serrated edge (Fig. 27b) and trapezoidal edge (Fig. 27a) are interface fracture mode and pull out fracture mode, respectively. Compared with the trapezoidal edge, the serrated edge can enhance the relative amplitude and reduce the depth of sonotrode tip penetrating into the top sheet, leading to the temperature increase and severe plastic deformation at the weld interface. Effect of weld tip geometry on the mechanical properties of USWed A6061 aluminium alloy joints was reported by Watanabe et al. [48]. No knurl appears on the C-tip surface, and the knurl pattern of the K-tip is pyramidal. It can be concluded that the maximum tensile load (1300 N, T-peel test) of the joints welded by employing the C-tip was 87.9% higher than that of the joints welded by utilizing the K-tip and the fracture type of the joint with the C-tip was pull out mode. Due to the pyramidal knurl pattern of the K-tip, the surface of the joint is easily damaged, but there is only very slight damage for the joint with C-tip.

stress concentration and plastic deformation occur at the edge of the sonotrode tip indentation, thus the lap shear tensile strength decreases, and the failure mode of the joint is transverse through-thickness (TTT) crack growth. Similar results was also reported by Zhang et al. [36], Annoni et al. [31], Mirza et al. [51], Haddadi et al. [19], Abdel-aleem et al. [56] and Bakavos et al. [12]. In another study, Elangovan et al. [57] optimized the weld parameters of USWed 0.3 mm thickness aluminum sheet. Relationship of weld parameter with weld strength is shown in Fig. 23. It can be observed from Fig. 23a, with increasing clamping pressure, the weld strength decreases, because higher clamping pressure reduces the relative motion between the two sheets. Moreover the values of weld strength are more sensitive the clamping pressure changes than the weld time changes. From Fig. 23b, with increasing amplitude, the weld strength increases, because higher amplitude is beneficial to the relative motion between the sheets, and the values of weld strength are more sensitive the amplitude changes than the weld time changes. In Fig. 22c, the values of weld strength are more sensitive the clamping pressure changes than the amplitude changes. Effects of weld parameter on the mechanical properties of USWed 1.0 mm thickness A5052-H32 alloy were investigated by Shin et al. [38]. It is found that the fracture modes involve three types: interface fracture, pull out fracture and mixed fracture, as shown in Fig. 24. Under the combined actions of weld time and amplitude, when the depth of sonotrode tip penetrating into the top sheet is smaller than 0.32 mm, interface fracture is predominant; if the depth of sonotrode tip penetrating into the top sheet is at the range of 0.32 mm to 0.36 mm, mixed fracture is dominant, indicating a sound joint. When the depth of sonotrode tip penetrating into the top sheet is larger than 0.36 mm, pull out fracture is predominant. It is noted that when the mixed fracture is dominant, the weld strength can reach the maximum value. In addition, 591

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 26. Horn and anvil surface tip patterns (a) Horn A (triangular prism, no space), Horn B (triangular prism, with space), and Horn C (pyramidal tips); (b) All the tips of anvil surfaces with different sizes are pyramidal [38].

without ethanol droplet, the dimples at the fracture surface increases, and new shrivelled-looking region appears. The weld strength is significantly enhanced due to the addition of ethanol droplet, and the fracture occurs in the base metal. In another study, due to the addition of Al2219 particle on the weld interface of Al/Al joint, the temperature and effective bonded areas at the weld interface are improved, thus the peak tensile shear load of the joint with Al2219 particle interlayer is 35.7% higher than that for the joint without Al2219 particle interlayer, as reported by Ni et al. [4]. In addition, Baboi et al. [61] found that the Zn butter sheet can enhance the weld strength of USWed AA5754-H111 aluminum alloy joints, due to the reduction of sonotrode tip sticking and part marking. It was reported by Annoni et al. [62] that aluminum 6022-T4 sheets have been joined by the combined means of ultrasonic welding and structural adhesives. The weld strength can be significantly improved, and the hybrid joining technique is insensitive to the temperature in comparison with the adhesive joining one. Imai et al. [58] investigated the weldability of A1050-H24/A1050-O, A1050H24/A3003-O and A1050-H24/A5052 joints, respectively. It was found that material with low hardness is beneficial to the transmission of vibration energy, and it is easier to achieve sound A1050-H24/A1050-O joint using the soft A1050-O material. Moreover, Imai et al. [58] stated that the shape of specimen affects the weld strength, and lower weld strength is obtained for the large size or thickness of specimen.

Fig. 27. Sonotrode tip surface geometries [42].

Meanwhile the bonded areas at the weld interface for the joint with Ctip can produce easily in comparison with the joint with K-tip. Therefore the weld strength for the joint with C-tip is higher.

4.3. Fatigue behavior 4.2.3. Effect of other factors Effect of ethanol droplet on the tensile load (U-peel test) USWed 1 mm thick rolled A6061-T6 aluminum alloy joint was investigated by Watanabe et al. [40]. Due to the addition of ethanol droplet on the weld interface, the increase of relative motion of base metal sheet promotes the enhancement of weld interface temperature, more welded areas occur at the weld interface. In comparison with the fracture surface

Although numerous applications of USWed aluminum joints were designed to bear dynamic loading conditions, few studies about the fatigue properties of USWed joints were reported. For example, Peng et al. [47] investigated the weld energy on the fatigue strength of USWed 6022-T43 aluminum alloy joints. The fatigue life of the USWed 6022-T43 aluminum alloy joints with the weld energy of 1400 J and 592

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

Fig. 28. (a) S-N curves of USWed Al5754 alloy joints; (b) the maximum tensile shear stress versus the number of reversals to failure (2Nf) in the double-log scale for welded joints at a welding energy of 2000 J [51].

Fig. 29. Pmax-N fatigue curves at R = 0.1 [32].

sonotrode tip indentation that is regarded as a sound crack propagation site. It is demonstrated that fatigue failure mode is significantly ascribed to the lap joint geometry and the load configuration not the material type.

2000 J is equivalent within the experimental scatter at the lower cyclic loads, but the joints with weld energy of 1400 J possess a longer fatigue life at the higher cyclic loading levels. With decreasing the cyclic load, the fatigue fracture mode changes from nugget pull-out to transverse through-thickness (TTT) crack growth for the USWed joints with weld energy of 1400 J, but the fatigue fracture mode is TTT crack growth at all the cyclic load for the USWed joints with weld energy of 2000 J. Fatigue crack initially occurs at the edge of nugget attributed to the severe stress concentration, and propagates along fan-shaped divergent direction with some “river-flow” patterns, showing the features of fatigue striation perpendicular to the fatigue crack growth direction. Mirza et al. [51] investigated the weld energy on the fatigue strength of USWed Al5754 alloy joints. When the maximum cyclic load is 2 kN, 2.5 kN and 3 kN, the joints with the weld energy of 2000 J possess a higher fatigue life; when the maximum cyclic load is 0.5 kN and 1 kN, the joints with the weld energy of 2000 J show lower fatigue strength, as demonstrated in Fig. 28a. For the joints with the weld energy of 2000 J, the failure mode changes from the transverse through-thickness crack growth to the interfacial failure when the value of log σmax is above 1.49, as shown in Fig. 28b. For the joints with the weld energy of 1000 J, the failure mode changes from the transverse through-thickness crack growth to the interfacial failure when the value of log σmax is above 1.32. In another study, fatigue strength of USWed AA 6022-T4 alloy joints was investigated by Carboni et al. [32]. When the joints bear the same cyclic load, the fatigue life of double spot joints is longer than that of the single spot. Meanwhile, for the single spot joints, two runout specimens occur when Pmax is 280 N (10% of the ultimate load); in the case of Pmax = 440 N (again 10% of the ultimate load), no runout specimen can be found in the double spot joints, as shown in Fig.29. For the joints subjected to high cyclic load, the failure mode is interface failure; for the joints subjected to low cyclic load, the failure mode is pull-out failure due to the severe stress concentration at the edge of

5. Conclusions and future trends This review aims to summarize the current state of joining aluminum alloy by USW with numerous critical issues involving general process parameters, materials flow, interfacial temperature, stress distribution, interfacial shear force, relative motion, macrostructure and microstructure, and mechanical properties of USWed joints. Macrostructure, microstructure evolution, interfacial temperature and mechanical properties of the USWed joint are mainly determined by the employed process parameters. Once the welding parameters are optimized, underlying material flow in the combined action of the sonotrode tip and anvil throughout the weld interface occurs, leading to a wave-like displacement at the weld interface combined with complex plastic deformation. In addition, temperature of weld interface is a key factor to the weld formation. Because appropriate weld interface temperature is conductive to improve the weldability via lowering the material yield strength and controlling the intermetallic compounds. During USW, the base metal sheets bear clamping force, high vibration frequency and non-uniform temperature field, thus stress is generated in the USWed joint, which affects the mechanical strength. With increasing weld time, the deformed aluminum gradually flows into the valley areas of the sonotrode knurl pattern, thus the depth that sonotrode tip penetrates into the top aluminum sheet becomes deeper. The cross section of USWed joints involves three zones: weld zone, weld affected zone and compression zone. Mechanical properties of USWed joints have been overviewed in terms of hardness, tensile strength and fatigue behavior. The more 593

Journal of Manufacturing Processes 35 (2018) 580–594

Z.L. Ni, F.X. Ye

absolutely the recrystallization process occurs, the softer the sheets are. Meanwhile, hardness is affected by deformation-induced excess vacancy concentration generated due to the high strain rate dynamic deformation and material attribute. Stress distribution and effective thickness of joints are critical factors to the mechanical properties of USWed joints. Plastic deformation at the weld interface is critical to the weld strength, but the dynamic process can not be observed in the experiment, thus computer simulation is needed. Indentations on the surface of welded sheets during USW affect the mechanical strength and appearance quality of joints, thus more methods should be proposed to improve the weld strength and reduce the penetration depth of snonotrde tip. In addition, numerous data of tensile strength are achieved at room temperature, and few studies about fatigue strength are reported. For the application of USWed joints in the vehicle structure, the test temperature of mechanical properties should be at the range of -50℃-50℃, and more studies about fatigue strength of joint are needed.

2011;27:617–24. [27] Fujii HT, Goto Y, Sato YS, Kokawa H. Microstructure and lap shear strength of the weld interface in ultrasonic welding of Al alloy to stainless steel. Scr Mater 2016;116:135–8. [28] Ni ZL, Ye FX. Weldability and mechanical properties of ultrasonic welded aluminum to nickel joints. Mater Lett 2016;185:204–7. [29] Panteli A, Robson JD, Chen YC, Prangnell PB. The Effectiveness of surface coatings on preventing interfacial reaction during ultrasonic welding of aluminum to magnesium. Metall Mater Trans A 2013;44:5773–81. [30] Chen YC, Bakavos D, Gholinia A, Prangnell PB. HAZ development and accelerated postweld natural ageing in ultrasonic spot welding aluminium 6111-T4 automotive sheet. Acta Mater 2012;60:2816–28. [31] Annoni M, Carboni M. Ultrasonic metal welding of AA 6022-T4 lap joints: part I – technological characterisation and static mechanical behaviour. Sci Technol Weld Join 2011;16:107–15. [32] Carboni M, Annoni M. Ultrasonic metal welding of AA 6022-T4 lap joints: part II – fatigue behaviour, failure analysis and modeling. Sci Technol Weld Join 2011;16:116–25. [33] Elangovan S, Semeer S, Prakasan K. Temperature and stress distribution in ultrasonic metal welding—an FEA-based study. J Mater Process Technol 2009;209:1143–50. [34] Ngo TT, Huang JH, Wang CC. The BFGS method for estimating the interface temperature and convection coefficient in ultrasonic welding. Int Commun Heat Mass Transf 2015;69:66–75. [35] Ni ZL, Ye FX. Effect of lap configuration on the microstructure and mechanical properties of dissimilar ultrasonic metal welded copper-aluminum joints. J Mater Process Technol 2017;245:180–92. [36] Zhang CY, Chen DL, Luo AA. Joining 5754 automotive aluminum alloy 2-mm-thick sheets using ultrasonic spot welding. Weld J 2014;93:131s–8s. [37] Elangovan S, Henry SPJ, Kalakkath P. Experimental studies on optimization of process parameters and finite element analysis of temperature and stress distribution on joining of Al–Al and Al–Al2O3 using ultrasonic welding. Int. J. Adv. Manuf. Technol. 2011;55:631–40. [38] Shin HS, Leon MD. Parametric study in similar ultrasonic spot welding of A5052-H32 alloy sheets. J Mater Process Technol 2015;224:222–32. [39] De Vries E. Mechanics and mechanisms of ultrasonic metal welding (Dissertation). The Ohio State University; 2004. [40] Watanabe T, Itoh H, Yanagisawa A, Hiraishi M. Ultrasonic welding of heat-treatable aluminium alloy A6061 sheet. Weld Int 2009;23:633–9. [41] Aydın H, Bayram A, Durgun İ. The effect of post-weld heat treatment on the mechanical properties of 2024-T4 friction stir-welded joints. Mater Des 2010;31:2568–77. [42] Komiyama K, Sasaki T, Watanabe Y. Effect of tool edge geometry in ultrasonic welding. J Mater Process Technol 2016;229:714–21. [43] Sasaki T, Komiyama K, Pramudita JA. Influence of tool edge angle on the bondability of aluminum in ultrasonic bonding. J Mater Process Technol 2018;252:167–75. [44] Takahashi Y, Suzuki S, Ohyama Y, Maeda M. Numerical analysis of interfacial deformation and temperature rise during ultrasonic Al ribbon bonding. J Phys: Conf Ser 2012;379:012028. [45] Ando M, Maeda M, Takahashi Y. Evolution of interfacial shear force during ultrasonic Al ribbon bonding. Mater Trans 2013;54:911–5. [46] Sasaki T, Watanabe T, Hosokawa Y, Yanagisawa A. Analysis for relative motion in ultrasonic welding of aluminium sheet. Sci Technol Weld Join 2013;18:19–24. [47] Peng H, Chen DL, Jiang X. Microstructure and mechanical properties of an ultrasonic spot welded aluminum alloy: the effect of welding energy. Materials 2017;10:449. [48] Watanabe T, Miyajima D, Yanagisawa A. Effect of weld tip geometry on ultrasonic welding of A6061 aluminium alloy. Weld Int 2010;24:336–42. [49] Ji H, Wang J, Li M. Evolution of the bulk microstructure in 1100 aluminum builds fabricated by ultrasonic metal welding. J Mater Process Technol 2014;214:175–82. [50] Xie J, Zhu Y, Bian F, Liu C. Dynamic recovery and recrystallization mechanisms during ultrasonic spot welding of Al-Cu-Mg alloy. Mater Charact 2017;132:145–55. [51] Mirza FA, Macwan A, Bhole SD, Chen DL. Microstructure and fatigue properties of ultrasonic spot welded joints of aluminum 5754 Alloy. JOM 2016;68:1465–75. [52] Bakavos D, Chen Y, Babout L, Prangnell P. Material interactions in a novel pinless tool approach to friction stir spot welding thin aluminum sheet. Metall Mater Trans A 2011;42:1266–82. [53] Manladan SM, Yusof F, Ramesh S, Fadzil M. A review on resistance spot welding of magnesium alloys. Int J Adv Manuf Technol 2016;86:1805–25. [54] Lang B, Sun DQ, Li GZ, Qin XF. Effects of welding parameters on microstructure and mechanical properties of resistance spot welded magnesium alloy joints. Sci Technol Weld Join 2008;13:698–704. [55] Zhou B, Thouless MD, Ward SM. Determining mode-I cohesive parameters for nugget fracture in ultrasonic spot welds. Int J Fract 2005;136:309–26. [56] Abdel-Aleem H, Yamagata T, Katoh M, Nishio K, Yamaguchi T. Joining of A1050/A5052 and A1050/Cu by ultrasonic bonding and their materials evaluation. Q J Jpn Weld Soc 2003;21:493–500. [57] Elangovan S, Anand K, Prakasan K. Parametric optimization of ultrasonic metal welding using response surface methodology and genetic algorithm. Int J Adv Manuf Technol 2012;63:561–72. [58] Imai H, Matsuoka SI. Finding the optimum parameters for ultrasonic welding of aluminum alloys. JMSE Int J 2005;48:311–6. [59] Baboi M, Grewell D. Comparison of control algorithms for ultrasonic welding of aluminum. Weld J 2010;89:243s–8s. [60] Xu Z, Li Z, Ji S, Zhang L. Refill friction stir spot welding of 5083-O aluminum alloy. J Mater Sci Technol 2018;34:878–85. [61] Baboi M, Grewell D. Effect of buffer sheets on the shear strength of ultrasonic welded aluminum joints. Weld J 2009;88:86s–91s. [62] Annoni M, Moroni F, Mussi V. Performance variability of aluminium hybrid LAP-joints obtained by means of adhesives and ultrasonic welding. Int J Mater Form 2010;3:1051–4.

References [1] Huang Y, Wang J, Wan L, Meng X, Liu H, Li H. Self-riveting friction stir lap welding of aluminum alloy to steel. Mater Lett 2016;185:181–4. [2] Padhy GK, Wu CS, Gao S. Subgrain formation in ultrasonic enhanced friction stir welding of aluminium alloy. Mater Lett 2016;183:34–9. [3] Lu Y, Song H, Taber GA, Foster DR, Daehn GS, Zhang W. In-situ measurement of relative motion during ultrasonic spot welding of aluminum alloy using Photonic Doppler Velocimetry. J Mater Process Technol 2016;231:431–40. [4] Ni ZL, Ye FX. Ultrasonic spot welding of Al sheets by enhancing the temperature of weld interface. Mater Lett 2017;208:69–72. [5] Ni ZL, Ye FX. Microstructure and mechanical performances of ultrasonic spot welded Al/ Cu joints with Al 2219 alloy particle interlayer. Mater Des 2016;92:779–86. [6] Mirza FA, Macwan A, Bhole SD, Chen DL, Chen XG. Effect of welding energy on microstructure and strength of ultrasonic spot welded dissimilar joints of aluminum to steel sheets. Mater Sci Eng: A 2016;668:73–85. [7] Balasundaram R, Patel VK, Bhole SD, Chen DL. Effect of zinc interlayer on ultrasonic spot welded aluminum-to-copper joints. Mater Sci Eng: A 2014;607:277–86. [8] Haddadi F. Microstructure reaction control of dissimilar automotive aluminium to galvanized steel sheets ultrasonic spot welding. Mater Sci Eng: A 2016;678:72–84. [9] Peng J, Fukumoto S, Brown L, Zhou N. Image analysis of electrode degradation in resistance spot welding of aluminium. Sci Technol Weld Join 2004;9:331–6. [10] Bakavos D, Prangnell PB. Effect of reduced or zero pin length and anvil insulation on friction stir spot welding thin gauge 6111 automotive sheet. Sci Technol Weld Join 2013;14:443–56. [11] Su P, Gerlich A, North TH, Bendzsak GJ. Energy utilisation and generation during friction stir spot welding. Sci Technol Weld Join 2013;11:163–9. [12] Bakavos D, Prangnell PB. Mechanisms of joint and microstructure formation in high power ultrasonic spot welding 6111 aluminium automotive sheet. Mater Sci Eng A 2010;527:6320–34. [13] Ni ZL, Ye FX. Weldability and mechanical properties of ultrasonic joining of aluminum to copper alloy with an interlayer. Mater Lett 2016;182:19–22. [14] Neppiras EA. Ultrasonic welding of metals. Ultrasonics 1965;3:128–35. [15] Macwan A, Kumar A, Chen DL. Ultrasonic spot welded 6111-T4 aluminum alloy to galvanized high-strength low-alloy steel: microstructure and mechanical properties. Mater Des 2017;113:284–96. [16] Matheny MP, Graff KF. Ultrasonic welding of metals. Power Ultrasonics 2015:259–93. [17] Jahn R, Cooper R, Wilkosz D. The effect of anvil geometry and welding energy on microstructures in ultrasonic spot welds of AA6111-T4. Metall Mater Trans A 2007;38:570–83. [18] Allameh SM, Mercer C, Popoola D, Soboyejo WO. Microstructural characterization of ultrasonically welded aluminum. J Eng Mater Technol 2005;127:65–74. [19] Haddadi F, Tsivoulas D. Grain structure, texture and mechanical property evolution of automotive aluminium sheet during high power ultrasonic welding. Mater Charact 2016;118:340–51. [20] Ren D, Zhao K, Pan M, Chang Y, Gang S, Zhao D. Ultrasonic spot welding of magnesium alloy to titanium alloy. Scr Mater 2017;126:58–62. [21] Macwan A, Chen DL. Microstructure and mechanical properties of ultrasonic spot welded copper-to-magnesium alloy joints. Mater Des 2015;84:261–9. [22] Wang SQ, Patel VK, Bhole SD, Wen GD, Chen DL. Microstructure and mechanical properties of ultrasonic spot welded Al/Ti alloy joints. Mater Des 2015;78:33–41. [23] Patel VK, Bhole SD, Chen DL. Formation of zinc interlayer texture during dissimilar ultrasonic spot welding of magnesium and high strength low alloy steel. Mater Des 2013;45:236–40. [24] Santella M, Brown E, Pozuelo M, Pan TY, Yang JM. Details of Mg–Zn reactions in AZ31 to galvanised mild steel ultrasonic spot welds. Sci Technol Weld Join 2013;17:219–24. [25] Patel VK, Bhole SD, Chen DL. Characterization of ultrasonic spot welded joints of Mg-togalvanized and ungalvanized steel with a tin interlayer. J Mater Process Technol 2014;214:811–7. [26] Prangnell P, Haddadi F, Chen YC. Ultrasonic spot welding of aluminium to steel for automotive applications - microstructure and optimization. Mater Sci Technol

594