A review on resistance spot welding of magnesium ...

A review on resistance spot welding of magnesium alloys

S. M. Manladan, F. Yusof, S. Ramesh & M. Fadzil

The International Journal of Advanced Manufacturing Technology ISSN 0268-3768 Int J Adv Manuf Technol DOI 10.1007/s00170-015-8258-9

1 23

Your article is protected by copyright and all rights are held exclusively by SpringerVerlag London. This e-offprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.

1 23

Author's personal copy Int J Adv Manuf Technol DOI 10.1007/s00170-015-8258-9

ORIGINAL ARTICLE

A review on resistance spot welding of magnesium alloys S. M. Manladan 1 & F. Yusof 1,2 & S. Ramesh 1,2 & M. Fadzil 2

Received: 17 August 2015 / Accepted: 17 December 2015 # Springer-Verlag London 2016

Abstract This paper presents a review on resistance spot welding of magnesium alloys, with emphasis on the relationship between microstructure, properties, and performance, under quasi-static and dynamic loading conditions. It also compares the resistance spot welding of magnesium-to-aluminum alloys and the various techniques used to suppress the formation of brittle intermetallic compounds. Resistance spot welding of magnesium-to-steel, weld bonding, the effects of process parameters on joint quality, and the main metallurgical defects in resistance spot welding of magnesium alloys are also deliberated. Studies have shown that the pre-existence of coarse second phase particles in the base metal, the addition of particles, such as titanium powder, and welding under the influence of electromagnetic stirring effect can promote columnar-to-equiaxed transition, microstructure refinement, and improvement in mechanical properties of magnesium alloys resistance spot welds. For magnesium-to-aluminum alloys spot welds, the use of interlayers, such as pure nickel, gold-coated nickel foil, and zinc-coated steel, was found to suppress the formation of brittle intermetallic compounds and thus significantly improve the joint strength. Keywords Resistance spot welding . Magnesium alloys . Aluminum alloys . Microstructure . Failure mode . Welding parameters . Intermetallic compounds . Weld bonding * F. Yusof [email protected]

1

Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

2

Center for Advanced Manufacturing and Materials Processing (AMMP), Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

1 Introduction Environmental pollution and fuel consumption are major concerns for the transportation industry. Automobile manufacturers are constantly seeking ways to reduce fuel consumption and greenhouse gas emissions. One of the most cost-effective methods to improve fuel efficiency is by reducing the weight of the vehicle. It has been estimated that for every 10 % reduction in weight, fuel economy could increase by 7 %. Thus, significant investments have been made in the research and development of lightweight materials for use in the fabrication of automotive bodies and components [1–3]. As the lightest structural materials, with superior specific strength, magnesium alloys have great potentials for weight savings applications across a wide range of industries, including power tools and 3C (computer, communication, and consumer products) sectors [4, 5]. Magnesium and its alloys are described as green engineering materials and one of the most promising material categories of the twenty-first century [6]. As a lightweight material, magnesium has a low density, approximately one fourth that of steel and two thirds that of aluminum [4, 5]. It has high specific strength, high elastic modulus, strong ability to withstand shock loads, and hot formability. In addition, it possesses excellent electromagnetic interference shielding, high heat dissipation capability, good castability, damping capacity, and recyclability [7–10]. The specific strength of magnesium is significantly higher than that of aluminum and iron in the ratio of 14.1 and 67.7 %, respectively [11]. It is expected that magnesium alloys will be able to replace steel and aluminum alloys as the primary structural material in the automotive and aerospace industries [12–14]. According to “Magnesium Vision 2020,” the average magnesium content in an automobile could increase to as much as 350 lb by the year

Author's personal copy Int J Adv Manuf Technol

2020 [15]. Table 1 summarizes some of the major applications of magnesium alloys in an automobile [16]. Resistance spot welding (RSW) is the most widely used joining process for sheet metals assemblies. For decades, it has been utilized extensively to produce high quality joints in the automobile and aerospace applications [17–22]. RSW also has applications in optical, electronic, and medical packaging industries [23]. The process is efficient, inexpensive, highly productive, reliable, easy to operate and automate, and therefore an ideal joining process for mass production [19, 24–26]. There are approximately 5000 spot welds in a single automobile [21, 25, 27, 28]. The quality, structural performance, lifespan, safety design, strength, stiffness, and integrity of a vehicle depend not only on the mechanical properties of the sheets but also on the quality of spot welds. Furthermore, the vehicle crashworthiness, which is defined as the capability of a car structure to provide adequate protection to its passengers against injuries in the event of a crash, largely depends on the integrity and the mechanical performance of spot welds [28, 29]. Although RSW process is able to produce high-quality and durable welds in steels, problems such as high welding current, electrode degradation, expulsion (ejection of molten metal from the nugget), and porosity are still major concerns in RSW of magnesium alloys [30]. Therefore, to successfully incorporate magnesium alloys in to the automotive industry and to achieve the “Magnesium Vision 2020,” efforts should be made to better understand and improve the structure, properties, and performance of magnesium alloy spot welds, under static and dynamic loading conditions. The aim of this paper is to provide an account of the state of understanding of RSW of magnesium alloys in order to provide a basis for follow-on research. It deliberates on important aspects that govern the quality, performance, and failure characteristics of magnesium alloys resistance spot welds.

Table 1

Global magnesium applications in automobile [16]

System Interior

Body

Chassis

North America Europe Asia

Instrument panel

Yes

Knee bolster retainer

Yes

Yes

Yes

Seat frame Seat riser

Yes Yes

Yes Yes

Yes Yes

Seat pan Console bracket

Yes Yes

Yes

Airbag housing

Yes

Center console cover Steering wheel

Yes Yes

Yes Yes

Yes

Keylock housing Steering column parts

Yes Yes

Yes

Yes

Radio housing

Yes

Yes

Glove box door Window motor housing

Yes Yes

Yes

Door inner panel Liftgate inner panel

Yes

Yes Yes

Roof frame

Yes

Yes

Sunroof panel Mirror bracket Fuel filler lid Door handle

Yes Yes Yes

Yes Yes Yes Yes

Yes

Spare tire carrier Wheel (racing) ABS mounting bracket Brake pedal bracket Brake/accelerator bracket

Yes Yes Yes Yes Yes

Yes

Yes

Brake/ clutch bracket Brake pedal arm Powertrain Engine block Valve cover/cam cover 4WD transfer case Transmission case Clutch housing & piston Intake manifold Engine oil pan Alternator/AC bracket

2 Fundamentals of RSW In a typical RSW operation, two or more similar or dissimilar overlapping metal sheets are placed between two watercooled electrodes. Pressure is then applied on the electrodes to clamp the workpieces together, producing an intimate contact between them. Electrical current is then supplied to the workpieces via the electrodes for a controlled period of time. Due to resistance of the sheets to the flow of a localized electrical current, heat is generated and a molten nugget is produced at the faying interface. The current is then switched off and the nugget will begin to solidify, while at the same time maintaining the electrode pressure. The cooling is achieved by heat conduction via the two water-cooled electrodes, and also radially outwards through the sheets [18, 31–33].

Component

Transmission stator Oil filter adapter Electric motor housing

Yes

Yes Yes Yes Yes Yes Yes

Yes Yes

Yes

Yes

Yes

Yes Yes

Yes

Yes Yes Yes Yes

The heat generation is based on Joule’s law, which can be expressed as follows [34]: Q ¼ I 2 Rt

ð1Þ

where Q is heat input in joules, I is the current in amperes, R is the resistance in ohms, and t is the time in seconds. Therefore, the amount of heat generated depends on three factors: the current, the resistance, and the duration of the welding current [28].


As shown in Fig. 1, two types of resistances exist in RSW processes, namely, bulk resistance (R2 and R4) and contact resistance, which is found at the electrode-sheet interfaces (R1 and R5) and at the faying interface (R3) [35]. Furthermore, the resistance of the upper and lower electrodes, RU and RL, respectively, also contribute to the total resistance, which is the sum of all the resistances (RU + R1 + R2 + R3 + R4 + R5 + RL) [36]. To facilitate nugget formation, R3 needs to be the highest [36]. The bulk resistance is sensitive to temperature and independent of pressure while the contact resistance is highly sensitive to pressure distribution, temperature, surface condition, and material characteristics. Generally, increasing the electrode force increases the actual metal-to-metal contact, thus decreasing the contact resistance [32, 35, 37, 38]. The presence of dirt, oil, coatings, and other foreign substances could also affect the contact resistance [35].

3 Surface preparation for RSW of magnesium alloys To prevent corrosion, magnesium alloys are normally protected using oil coating, acid pickled surface, or chromate conversion coating [38, 39]. This could lead to surface contamination, electrodes fouling, flashing, blowholes, and porosity in the welds [38]. For good quality welds, the surface of magnesium alloys should be cleaned before welding. The cleaning would reduce variations in contact resistance and decrease the heating between the electrodes and magnesium alloys, producing better quality joints [38]. Cleaning the surface of magnesium alloys with 2.5 % (w/v) chromic acid (2.5 g CrO3 + 100 ml H2O) was found to be effective in this regard [40, 41]. For example, Zhou et al. [41] observed that the surface of as-received AZ31B magnesium alloy consisted of MgO, Mg(OH)2, and MgCO3. The surface exhibited variations in contact resistance, with an average contact resistance of 78 mΩ. Cleaning the surface with 2.5 % (w/v) chromic acid produced more uniform contact resistance and reduced the average contact resistance to 3 mΩ [41]. During RSW, due to the high contact resistance of the as-received samples, rapid heat generation resulted in expulsion and poor quality joint. For the chromic acid-cleaned samples, no expulsion was

observed even at higher welding current. Moreover, these samples produced much less damage on the electrode tip faces [40, 41]. Consequently, cleaning of magnesium alloy surfaces with 2.5 % (w/v) chromic acid has become a common practice prior to RSW process [42–45].

4 Nugget formation in RSW of magnesium alloys The nugget formation and growth during RSW of magnesium alloys can be divided into three stages: incubation, growth, and stabilization [34]. In the incubation stage, which is relatively short, usually less than 1 cycle, the nugget begins to form due to the melting of the metal. In the growth stage, which occurs in the following 2–4 cycles, the nugget grows rapidly but the growth rate decreases with time. This is due to the reduction in current density and heating rate caused by the increase in the contact area between the electrode and workpiece. Finally, the nugget growth achieves stabilization after approximately 4 cycles. The duration of the incubation stage for magnesium alloys was found to be similar to that of aluminum and much smaller than that of steel [34, 46].

5 Microstructure The microstructure is controlled by a combination of the prevailing thermal condition at the solid/liquid interface and the rate of growth of crystals, which is directly related to the thermal gradient in the weld [47]. Due to low volumetric heat capacity, good thermal conductivity, and low melting point of magnesium alloys, the cooling rate of the weld is so high that the weld solidifies under non-equilibrium conditions [14]. Behravesh et al. [42] characterized the microstructure of AZ31-H24 magnesium alloy resistance spot welds and four different zones were identified, as shown in Fig. 2, i.e., the base metal (BM), heat-affected zone (HAZ), partially melted zone (PMZ), and fusion zone (FZ). The FZ in resistance spot welded magnesium alloys usually consists of two different zones, i.e., columnar dendritic zone (CDZ) and equiaxed dendritic zone (EDZ). The CDZ is found adjacent to the fusion line, with crystals nucleating and growing epitaxially from the

Fig. 1 Illustration of the electrical resistances in a sheet stack-up during RSW [35]

Electrode R1 R3

Workpiece 1 R2

R4 R5 Electrode

Workpiece 2


Fig. 2 Different zones in AZ31B resistance spot welds: a low magnification and b high magnification [42].

unmelted BM whereas the EDZ is located at the center of the nugget [22, 43, 44, 48–50]. The columnar-to-equiaxed transition (CET) occurs when the movement of the columnar front is blocked by enough equiaxed grains formed in the liquid ahead of the columnar front [32]. Compared to the columnar dendritic structure, the equiaxed grains are finer, have more isotropic structure, less segregation of alloying elements, and better mechanical properties. The columnar dendritic structure affects the mechanical properties of the weld and is therefore undesirable. Thus, it is crucial to promote the formation of equiaxed grains for improved mechanical properties [43, 49, 51]. Liu et al. [51] reported that the size of pre-existing second phase particles in the base metal affects CET transition. It was shown that AZ31B Mg alloy (SA), containing both submicron size and coarse Al8Mn5 particles, had short, fine, and narrow CDZ, and more developed equiaxed grains in the FZ. On the other hand, AZ31B magnesium alloy (SB), consisting only of submicron size Al8Mn5 particles, had well-developed columnar dendrite region, long primary arms, and coarse grain size. The addition of 10 μm long Mn particles to SA effectively suppressed the CDZ and promoted the formation of equiaxed grains [49]. It was also shown in another study that increasing the welding current would decrease CDZ width for both SA and SB. The CDZ nearly vanished when the welding current was higher than a certain critical value, which was about 24 and 28 kA for SA and SB, respectively. It was also shown that the addition of titanium powder, with particles size less than 20 μm, to the designated FZ, during RSW of AZ31-H24 magnesium alloy, significantly suppressed the CDZ, as shown in Fig. 3. The titanium particles served as inoculants, enhanced the nucleation of α-Mg grains and the formation of equiaxed dendritic structure. In addition to suppressing the CDZ, the grains in the EDZ were effectively refined by the addition of titanium. It was found that the average diameter of the flowerlike grains in the EDZ with and without the addition of titanium was approximately 20 and 65 μm, respectively. This led to significant improvement in mechanical properties [43]. Generally, for AZ series magnesium alloys spot welds, it was shown that grain size refinement and CET improved with

increase in aluminum content. Niknejad et al. [22] investigated the microstructural evolution in RSW AZ31, AZ61, and AZ80 magnesium alloys. It was observed that the higher aluminum content in AZ61 and especially AZ80 enhanced CET and grain size refinement. The average length of the columnar dendrite zone was found to be 320, 170, and 80 μm for AZ31, AZ61, and AZ80 magnesium alloys, respectively. The size of the dendrites also decreased from AZ31 to AZ61 and AZ80. The diameter of the flowerlike dendritic grains was found to be 31, 20, and 16 μm for AZ31, AZ61, and AZ80 welds, respectively. Yao et al. [50] have shown that RSW of AZ31B magnesium alloys under the influence of electromagnetic stirring also influences the microstructure of the FZ. In this study, two permanent magnets of opposite polarities, which were co-axially mounted on the electrode arms of the RSW machine, were used as the source of electromagnetic force. The results showed that RSW with electromagnetic stirring effect (EMS-RSW) promoted early CET and produced finer grains in HAZ, CDZ, and EDZ compared to conventional RSW process, as shown in Fig. 4. The high speed movement of the molten metal driven by the circumferential external magnetic force facilitated the formation of equiaxed grains by breaking the growing dendrites during the primary crystallization process. In addition, the EMS reduced the temperature gradient and constitutional supercooling degree, improved the balance crystallization temperature, so as to promote uniform diffusion and refined the microstructure. The HAZ of magnesium spot welds is characterized by recrystallization and grain growth [14, 22, 42]. For instance, a grain size gradient (10–6 μm), decreasing towards the BM, was observed in the HAZ of AZ31B-H24 magnesium alloy spot welds. This was because in the HAZ, the regions which are closer to the BM experienced lower annealing temperature and time than regions which are closer to the PMZ. Moreover, significantly higher twin band density was found in the HAZ than in the BM [42]. Babu et al. [14] reported that grain boundary melting occurred in the HAZ of AZ31 immediately adjacent to the nugget, with grain boundaries becoming coarser as compared to the unaffected base metal.

Author's personal copy Int J Adv Manuf Technol Fig. 3 Microstructure of the FZ AZ31 alloy welded a without and b with an addition of Ti [43]

Al12Mg17 intermetallics were observed in the grain boundaries of PMZ of AZ31B-H24 magnesium alloy. The peak temperature attained in the PMZ, which is located around the nugget, is between the solidus and liquidus temperatures of the BM. As a result, grain boundary liquation might have occurred due to the lower melting point and higher aluminum content of the grain boundaries, thus promoting the formation of Al12Mg17 intermetallics [42]. β-Mg17(Al,Zn)12 phases were also observed in grain boundaries in the HAZ of AZ31, AZ61, and AZ80 magnesium alloys. The quantity of these phases was higher in AZ80 and AZ61 than in AZ31, due to higher aluminum content. Different mechanisms were proposed for the formation of these β phases, depending on the alloy type. For AZ31 alloy, Fig. 4 Microstructure of AZ31B magnesium alloy weld produced by a conventional RSW; b RSW with electromagnetic stirring (16 kA, 200 ms) [50]

even though minute traces of the β phase were found in microstructure of the BM, formation of the β phase would suggest that liquation has occurred in grain boundaries of the HAZ. For AZ61 and AZ80 alloys, the β phase pre-existed in grain boundaries of the BM and they reacted with the surrounding α matrix to form a liquid eutectic layer at the grain boundaries due to rapid heating at the HAZ [22]. The existence of these particles was detrimental to the strength of the welds, especially in AZ61 and AZ80 alloys, which failed in the HAZ along the FZ due to preferential micro-cracking at the interfaces of the β-phases and Mg matrix during tensile shear testing. Post-weld solutionizing heat treatment significantly reduced the quantity of these particles, and thus improved the strength of the joints [22, 44].


6 Mechanical properties 6.1 Hardness

and 58 HV for the BM, HAZ, CDZ, and EDZ, respectively, of AZ31 magnesium alloy. Under the influence of EMS, the hardness of each zone has increased due to grain size refinement. The hardness ratio of fusion zone to pullout failure location (usually is HAZ) was 1.28 for EMS-RSW and 1.03 for traditional RSW, implying that the EMS-RSW joint is more likely to experience pullout failure [50]. On the contrary, for continuous cast and rolled AZ31 magnesium alloy resistance spot weld, Babu et al. [14] observed a significant hardness reduction in the weld nugget and HAZ compared to the BM, as shown in Fig. 6 . The reduction in hardness in the weld nugget and HAZ was due to the formation of dendritic microstructure and coarse grains, respectively. The high hardness of the BM was due to fine grain size and cold working [14].

Generally, minimal variations in the hardness value across the BM, HAZ, and FZ has been reported in resistance spot welded magnesium alloys [42, 44]. For AZ31B-H24 magnesium alloy, the hardness in the weld area was found to be almost the same with that of the BM. This was due to the occurrence of two contrasting phenomena which counteracted each other, resulting in uniform distribution of hardness. The increase in the grain/dendrite size from the BM to the FZ tends to decrease the hardness. On the other hand, the hardness is increased due to the presence of intermetallics in the PMZ and FZ, and twin bands in the HAZ. Even under cyclic loading, AZ31B-H24 magnesium alloy did not show appreciable hardness variation across the BM, HAZ, and FZ, suggesting that both the BM and weld region did not undergo cyclic hardening [42]. Similarly, a relatively uniform hardness distribution was observed across the BM, HAZ, and FZ of resistance spot welded AZ80 magnesium alloy. After postweld heat treatment, there was a reduction in hardness across these zones, which was attributed to the partial dissolution of β-Mg17Al12 phase and grain growth [44]. Liu et al. [51] observed minimal hardness variation across the BM, HAZ, and FZ of resistance spot welded AZ31 (SA) and AZ31 (SB) magnesium alloys, with the BM having the highest microhardness value of 70 HV, as shown in Fig. 5. This was probably due to the welding process itself which has resulted in the reduction of pre-existing deformed structures such as solution strengthening, dislocation density, and defects in BM. The CDZ exhibited an average microhardness value of about 69 HV in AZ31 (SA) and 60 HV in AZ31 (SB), whereas the average microhardness value of the EDZ was 67 HV in AZ31 (SA) and 61 HV in AZ31 (SB). [51]. Similarly, Yao et al. [50] reported an average microhardness value of approximately 64, 55, 63,

The mechanical performance of spot welds is normally evaluated under quasi-static and dynamic loading conditions. Tensile-shear (TS), cross-tension (CT), and coach peel (CP) tests are examples of tests conducted under quasi-static loading conditions. Impact and fatigue tests are some of the tests conducted under dynamic loading conditions [29]. Tensileshear test is widely used to determine the strength of resistance spot welds due to the simplicity in sample preparations [14]. In this test, the load bearing capacity (peak load) and failure energy are the two most important parameters used to describe the performance of the joint [29]. Three types of failure modes commonly occur during tensile-shear test of spot welds, i.e., interfacial, partial interfacial, and pull out failure modes [29]. In interfacial failure mode, cracking occurs through the nugget centerline, separating the sheets apart. It is accompanied by little plastic deformation. In partial interfacial failure mode, a fraction of the weld nugget is removed. The crack first propagates in the weld nugget, then redirects perpendicularly to the centerline towards one of the sheets [29, 50]. Pull out failure

Fig. 5 Hardness profile across welds of two AZ31 alloys [51]

Fig. 6 Hardness profile across resistance spot welds of AZ31 Mg alloy in the as-welded condition [14]

6.2 Tensile shear load and failure mode


mode occurs by complete or partial withdrawal of the nugget from one sheet. In this mode, crack does not propagate through the nugget. Pull out failure mode is usually accompanied by increased plastic deformation, thus leading to higher energy absorption and peak load, making it the most desirable among the three failure modes [14, 29]. These failure modes are schematically illustrated in Fig. 7. The amount of force required to cause failure is equal to the product of the strength of the material and the failed area of cross section. Based on this, the following equations were derived to predict failure loads, FPO and FIF, for pullout and interfacial failure modes, respectively [52]: F PO ¼ k PO ⋅σU T ⋅d⋅t

ð2Þ

F IF ¼ k IF ⋅σUT ⋅d 2

ð3Þ

where k PO ð∼2:2Þ and k IF ð∼0:6Þ are constants ; σUT is the ultimate tensile shear strength of the base material, d is nugget diameter, and t is the sheet thickness. It can be noted from Eqs. (2) and (3) that the failure load for the pull out/through thickness failure depends strongly on the nugget diameter and sheet thickness, while failure load for interfacial failure mode is primarily a function of nugget diameter. Interfacial failure mode is common in magnesium alloys spot welds [14, 42], partly because the hardness, and therefore strength, in the FZ is comparable to or less than the BM [42]. For example, Behravesh et al. [42] have conducted tensileshear test on resistance spot welded AZ31B-H24 alloys and found that the samples failed predominantly in interfacial failure mode, with an average ultimate tensile shear load (UTSL) of 6.67 kN. Similarly, Babu et al. [14] carried out tensile shear tests on AZ31 magnesium alloy resistance spot welds at constant nugget diameter of 3.5√ t (t = sheet thickness, 3 mm) for all samples. All the samples failed in the interfacial mode with an average peak load of 4.7 kN. Xiao et al. [43] evaluated the effect of titanium addition on the mechanical properties of AZ31 magnesium alloy resistance spot welds. They found that the addition of titanium increased both the UTSL and displacement of the welds. For example, at a welding current of 26 kA, the addition of titanium increased the UTSL of the joint by 38 %, from 4.076 to 5.169 kN, while the displacement increased by 28 % from 1.09 to 1.40. For AZ80 magnesium alloy, an average UTSL of 4.69 kN was obtained with nugget pull out failure mode. Post

weld solution heat treatment improved the strength to 6.63 kN and changed the failure mode to through thickness [44]. A comparison of the mechanical performance of AZ31, AZ61, and AZ80 magnesium alloys resistance spot welds under the same conditions, during tensile-shear test, showed that AZ31 failed in interfacial failure mode while AZ61 and AZ80 alloys failed in nugget pull-out failure mode. After post weld solution heat treatment, the failure mode in AZ31 welds remained unchanged while those of AZ61 and AZ80 transited from pull out mode to through-thickness. In addition, the heat treatment increased the average strength of joints of AZ31, AZ61, and AZ80 by 3.1, 11.7, and 37.2 %, respectively, as shown in Fig. 8 [22]. Recently, Yao et al. [50] studied the effect of electromagnetic stirring on the properties of AZ31B magnesium alloy resistance spot welds. The results showed that samples produced by EMS-RSW had larger nugget diameter, higher tensile shear force and energy absorption capacity, and thus higher probability of pullout failure mode than those produced by conventional RSW, at all welding currents.

6.3 Fatigue behavior Spot welds act as sites for stress concentration and are therefore susceptible to fatigue failure [42]. Fatigue is the most critical failure mode of spot-welded and weld-bonded joints in automobiles [53]. A profound understanding of the fatigue behavior of spot welded joints is required to ensure the integrity, durability, and safety of welded structures [48, 54]. However, studies on the fatigue behavior of magnesium alloys spot welds are currently very limited. Fatigue tests have been carried out to investigate the behavior of AZ31B-H24 magnesium alloys resistance spot welds in the tensile–shear configuration. The results showed that three different failure modes occurred, i.e., coupon, interfacial, and partially interfacial failure modes, with coupon failure being the most dominant [42, 55]. The coupon failure occurred in the intermediate and high cycle fatigue (HCF) regimes, at lower loads. Since the crack did not propagate through the nugget in this failure mode, the fatigue life would be independent of the nugget strength. Instead, it would depend on the level of cyclic loading and coupon dimensions. Interfacial failure mode was observed when very high cyclic load was applied. Since crack propagated through the nugget

Fig. 7 Schematic illustration of a interfacial; b partial interfacial; and c pullout failure modes [50]

Author's personal copy Int J Adv Manuf Technol Table 2 AZ31B-H24 magnesium alloy spot-welded specimens coding and nugget diameter [55] Specimen set

Configuration

Average nugget diameter (mm)

A C

TSa TS

8.2 (0.7)b 9.5 (0.1)

E F G

TS TS-Wc CTd

10.4 (0.2) 10.4 (0.2) 10.4 (0.2)

a

Fig. 8 Failure peak load and specimen elongation (at the peak load) of magnesium alloy resistance spot welds in as-welded and heat treated conditions [22]

in this failure mode, the fatigue strength depended largely on the nugget size and strength. In contrast, partially interfacial failure mode rarely occurred and was only observed between very low and low cycle regimes (for fatigue life between 3 × 103 and 104 cycles) [42, 55]. The nugget size was found to have strong influence on fatigue resistance in the low cycle regime. However, this influence decreased gradually over fatigue life and eventually the fatigue resistance became almost independent of nugget size above 105 cycles [42]. Figure 9 depicts the load–life curves for AZ31B-H24 magnesium alloy resistance spot welded specimens of different configurations and nugget diameters (Table 2). From the load–life curves of specimens set A, C, and E, it was seen that enlarging the nugget size has insignificant effect on fatigue strength (in terms of load range). Also, by comparing the curves for sets A–E and set F, it was seen that increasing the coupon width and decreasing the mean load have led to improvement of fatigue strength for LCF, although the effect gradually decreases for HCF. Furthermore, comparing the curves for sets G, E, and F shows that the fatigue strength of CT specimen is significantly lower than that of TS specimens of the same nugget size.

Fig. 9 Load-life experimental data for A31B-H24 resistance spotwelded specimens [55]

Standard size tensile–shear test specimen

b

Values in parentheses are standard deviations

c

Wide tensile–shear test specimen

d

Standard size cross-tension test specimen

The endurance limit for specimen set A, C, E, F, and G are 0.34, 0.44, 0.48, 0.72, and 0.16 kN, respectively [55]. It was also reported that the load level has an effect on the location of crack initiation. A high cyclic load would result in crack initiation close to the nugget, and as the cyclic load decreases, the crack initiation point will move further away from the nugget. In addition, under cyclic loading, two cracks had initiated at opposite sides of the nugget. These are the primary crack, which propagated until failure occurred, and the secondary crack, which propagated to a certain extent but did not result in failure [42]. The cyclic behavior of AZ31B spot-welds was evaluated using different specimen configurations and compared with steel and aluminum spot-welds. It was found that the fatigue strength of magnesium spot-welds was similar to aluminum but considerably less than that of steel spot welds, for the same 1 d=t2 ratio [55]. Xio et al. [48] have studied the fatigue behavior of two different magnesium alloys, AZ31B (SA) and AZ31B (SB), having similar composition but containing second-phase particles of different sizes, in the as-received material and consequently different fusion zone microstructure. When tested under identical conditions of higher cycle load range, both samples failed in interfacial failure mode. However, AZ31B magnesium alloy (SA) with more refined microstructure (finer dendrite structure) had longer fatigue life than AZ31B magnesium alloy (SB) welds, and thus better fatigue resistance. Figure 10 shows the fracture surfaces of interfacial failure of SA and SB welds tested under the same cyclic load range of 3.12 kN. As indicated by the arrows in Fig. 10, typical fatigue striations are observed on both fracture surfaces, suggesting that the crack propagation was transgranular. Due to the finer dendritic structure in the fusion zone in SA welds compared to SB welds, the fatigue striations spacing for SA welds was smaller than that of SB welds, hence exhibiting slower crack propagation rate and longer fatigue life [48]. However, when the cyclic load was below 0.5 kN, both SA and SB had similar fatigue life.

Author's personal copy Int J Adv Manuf Technol Fig. 10 A comparison of fatigue crack propagation zones at higher cyclic load ranges in: a SA and b SB welds [48]

7 RSW with cover plates A very high electrical current is needed during RSW of magnesium alloys due to their high thermal and electrical conductivities. The flow of high electrical current promotes electrode tip wear, blowholes, expulsion, and the need for largercapacity machines [30, 33]. To successfully weld magnesium alloys with low welding current, Qui et al. [33] proposed the technique of RSW with cover plates. In this technique, magnesium alloy sheets are placed between two 1 mm thick cover plates made of cold rolled steel. Cold rolled steel is used since it has a lower electrical conductivity than magnesium alloy to ensure higher heat generation in the cover plate and subsequent conduction to the magnesium alloy. Other major considerations are low cost and the ability of the cover plate to separate from the magnesium alloy after welding [33, 37, 56]. The technique proved advantageous and was able to produce joints in AZ31B magnesium alloy with higher tensile shear strength and larger nugget diameter than those produced by conventional RSW. Moreover, these joints were welded with low welding currents, which are comparable to that for RSW of steel sheets [30, 33, 56]. For example, Shi et al. [56] obtained a nugget diameter and tensile shear load of 9.5 mm and 4.7 kN, respectively, at a welding current of 12 kA by RSW with cover plates whereas a nugget diameter of 3 mm and tensile shear load of less than 1 kN were obtained by traditional RSW under the same welding conditions. Although pores were observed in the nugget of the joints produced by this technique, their formation was effectively suppressed by increasing the electrode force and extending down-sloping time (time of welding current reducing to zero) [56].

8 RSW of magnesium alloys to aluminum alloys Aluminum alloys possess desirable properties such as low density, high specific strength, good corrosion resistance and appearance, good workability, and intrinsic recyclability [57]. Table 3 compares the properties of magnesium,

aluminum, and iron. Both magnesium and aluminum alloys possess lower density and higher specific strength than steels. Therefore, in automobile applications, using magnesium and aluminum alloys components would significantly reduce the weight of vehicles. However, welding magnesium alloys to aluminum alloys is very challenging due to the formation of intermetallic compounds (IMCs). Based on the Al-Mg phase diagram, IMCs such as Mg17Al12 and Mg2Al3 may form in the weld region [57]. Hayat [28] welded AZ31 magnesium alloy and 1350 aluminum alloy using RSW and observed that IMCs were formed at the nugget area. Increasing the welding current resulted in the melting of more magnesium, thus increasing the width of the IMCs. Al3Mg2, with a width of 17 μm was formed at a welding current of 22 kA, while Al12Mg17 with a width of 32 μm was formed at a welding current of 33 kA. The average hardness of the weld nugget was 190 ± 10 HV whereas the hardness of the magnesium and aluminum sides were 73 ± 5 and 40 ± 5 HV, respectively. The high hardness value of the nugget confirmed the presence of hard and brittle IMC. Moreover, the fracture surface of the weld exhibits brittle morphology, with distinctive characteristics of brittle intermetallics, as shown in Fig. 11 [28]. Similarly, Luo and Li [59] have studied the formation of nugget during RSW of AZ31B magnesium alloy and 2024 aluminum alloy. It was observed that during welding, a certain degree of plastic deformation occurred at the interface, and as the heat input increased, the magnesium alloy got softened further and the plastically deformed region expanded. The aluminum alloy protruded into the magnesium alloy, distorting the weld line and forming a wavelike appearance. These protruded regions acted as favorable sites for mixing of aluminum and magnesium at the interface and eventually nugget formation. Energy spectrum analysis revealed that the phase composition of the nugget is mainly Al12Mg17 IMCs, having a hardness significantly higher than those of aluminum and magnesium BMs, as shown in Fig. 12 [59]. Due to the high hardness and low plasticity of these IMCs, it would be easy to induce microcracking at the transition interface [28, 59].

Author's personal copy Int J Adv Manuf Technol Table 3 Comparison between properties of magnesium, aluminum, and iron [58]

Properties

Magnesium

Aluminum

Iron

Ionization energy (eV) Specific heat (J kg−1 k−1) Specific heat of fusion(J/kg) Melting point (°C)

7.6 1360 3.7 × 105 650

6 1080 4 × 105 660

7.8 795 2.7 × 105 1536

Boiling point (°C)

1090

2520

2860

Viscosity (kg m−1 s−1) Surface tension (Nm−1) Thermal conductivity (W m−1 k−1) Thermal diffusivity (m2 s−1) Coefficient of thermal expansion (1/k) Density (kg/m3) Elastic modulus (N/m3) Elastic resistivity (μΩm)

0.00125 0.559 78 3.73 × 10−5 25 × 10−6 1590 4.47 × 1010 0.274

0.0013 0.914 94.03 3.65 × 10−5 24 × 10−6 2385 7.06 × 1010 0.2425

Vapor pressure (Pa)

360

10−6

0.0055 1.872 38 6.80 × 10−6 10 × 10−6 7015 21 × 1010 1.386 2.3

8.1 Suppression of the formation of intermetallic compounds in RSW of magnesium alloys to aluminum alloys As mentioned earlier, direct welding of magnesium to aluminum tends to form IMCs, compromising the properties of the joint. The key to improving joint strength in magnesium/aluminum dissimilar joint is to control the IMCs or to reduce their negative effects [57]. Interlayers could be used as barrier materials to suppress the interaction of magnesium and aluminum and consequently restrict the formation of IMCs [57, 60, 61]. Ni has a melting point of 1455 °C which is significantly higher than that of Mg (650 °C) and Al (660.42 °C). Due to

Fig. 11 The fracture surface morphology on the Mg side of direct Al/Mg resistance spot weld [28]

this large difference in melting point, using Ni as an interlayer would prevent mixing of Al and Mg to form IMCs. Moreover, Mg-Ni and Al-Ni IMCs are less brittle than Al-Mg IMCs. Therefore, Ni based foil is a good candidate for use an as interlayer in RSW of Mg/Al dissimilar materials [61]. Penner et al. [61] investigated the possibility of using Nibased interlayers for the RSW of Al and Mg. Pure Ni foil and gold-coated Ni foil were used in the study to join AZ31B magnesium alloy and 5754 aluminum alloy. The results showed that the use of bare Ni interlayer did not produce a metallurgical bond between the Mg and Al alloys. However, the use of gold-coated Ni interlayer has completely suppressed the formation of Al-Mg IMC, significantly improving


Fig. 12 Microhardness profile across the joint interface of resistance spot welded AZ31B magnesium alloy 2024 aluminum alloy [59]

metallurgical bonding between the Mg and Al alloy. The goldcoated Ni interlayer remained intact during the welding due to its high melting point as compared to aluminum and magnesium. A strong joint with an average peak load of 4.69 kN was produced, which was as high as 90 % of the strength obtained for optimized similar AZ31B welds [61]. Unsuccessful attempt of Penner et al. [61] to join magnesium to aluminum with a Ni interlayer was attributed to the use of low heat input [62]. So, in a recent study, which also used pure Ni interlayer to join AZ31B magnesium alloy and 5754 aluminum alloy, Sun et al. [62] used higher welding current and also flattened the electrode tip faces, with smaller tip face on the Al side, so as to have better heat balance. They obtained strong joints, with a peak load value of about 5.1 kN at welding current of 32–36 kA. The Ni interlayer remained unmelted and thus the formation of brittle Al–Mg IMCs was prevented, while continuous submicron Al–Ni and Mg–Ni intermetallic layers were formed at the Al/Ni and Mg/Ni interfaces, respectively [62]. In another study [63], pure zinc foil and zinc-coated HSLA steel were used as interlayers in RSW of AZ31B Mg alloy to Al Alloy 5754. The use of pure zinc foil produced joints with poor strength due to formation of intermetallic compound with microhardness values of 224–304 HV in the nugget zone. However, zinc-coated steel, which remained in the solid state, prevented the mixing of Al and Mg, thus suppressing the formation of IMCs. A joint with acceptable strength, as per AWS D17.2 Standard, was produced (approximately 74 % of the strength the obtained for optimized similar AZ31B welds) [63]. As reported by Penner et al. [63], the use of pure zinc interlayer did not improve the strength of Mg/Al resistance spot welded joint. More recently, Zhang et al. [60] proposed the thermo-compensated RSW process, which was used to join AZ31B magnesium alloy to AA5052-H12 Al alloy using zinc interlayer. In this work, a tape made of stainless was

inserted between the upper electrode and the Al sheet. Due to the high electrical resistivity and low thermal conductivity of the stainless steel, joule heating was produced and thus it served as an additional heat source to Al alloy and as a barrier to prevent heat loss from the nugget on the Al side. As a result, increased amounts of elemental Al dissolved into the liquid nugget during welding. Thus the nugget had large amount of Al and Zn-based solid solution and subsequently improved the mechanical properties of the welded joint. The tensile– shear load of the joint produced by this process was 2199 N, which is much higher than that produced by conventional RSW with same interlayer (727 N) [60]. Comparing the results in Table 4, it can be seen that the use of zinc interlayer in most cases did not improve the strength of the joint. Although the use of thermo-compensated RSW process significantly improved the strength of the joint produced by using zinc interlayer, the strength is still low. The use of pure Ni interlayer at high welding current and gold-coated Ni interlayer produced joint with the highest strengths. However, the use of high cost material is commercially impractical and alternative interlayers with lower costs and greater availability are needed [63]. Other types of interlayers should also be tested and the feasibility of using the thermo-compensated RSW with interlayers such as zinc-coated steel and Ni should also be explored.

9 RSW of magnesium alloys to steel Steel is a primary structural material in the automotive industry. With the increased use of magnesium alloys, dissimilar joining of magnesium alloys to steel is inevitable. Studies conducted on RSW of magnesium to steel are currently very limited. The welding of magnesium to steel is extremely challenging due to the various differences in physical and metallurgical properties between magnesium alloys and steels (Table 3). There is a large difference between the melting points of iron and magnesium, with almost no intersolubility between them [64–66]. Moreover, steel has about three times the bulk resistance of magnesium and about half its thermal conductivity [67]. Thus, during the RSW of magnesium to steel, more heat would be generated on the steel side than on the magnesium side. This would cause the steel to melt and the magnesium to evaporate, forming pores in the weld nugget. Liu et al. [67] proposed the use of flat electrode and domedshaped electrode against the steel and magnesium sides, respectively. This would reduce the current density, increase the cooling rate on the steel side, and thus, balancing the heating of the materials. Using this technique, AZ31B magnesium alloy was successfully welded to zinc-coated DP600 steel and a joint with strength of 5.0 kN (which is about 95 % of the strength of an optimized Mg/Mg joint) was obtained. In another study, AZ31B Mg alloy and hot-dip galvanized

Author's personal copy Int J Adv Manuf Technol Table 4

A comparison of the strengths of Al/Mg dissimilar resistance spot welds produced using different interlayers and without interlayer

Process

Materials

Interlayer

Joint strength (kN)

Reference

Conventional RSW

AZ31B magnesium alloy and 5754 aluminum alloy

Pure nickel

[61]

Conventional RSW

AZ31B magnesium alloy and 5754 aluminum alloy

Pure nickel

Metallurgical bond not produced 5.1

[62]

Conventional RSW Conventional RSW

AZ31B magnesium alloy and 5754 aluminum alloy AZ31B Mg alloy to Al Alloy 5754

Gold-coated nickel foil Zinc-coated HSLA steel

4.69 3.86

[61] [63]

Conventional RSW Thermo-compensated RSW

AZ31B magnesium alloy to AA5052-H12 Al alloy AZ31B magnesium alloy to AA5052-H12 Al alloy

Zinc interlayer Zinc interlayer

0.727 2.199

[60] [60]

Conventional RSW

AZ31B magnesium alloy to AA5052-H12 Al alloy

No interlayer

0.833

[60]

HSLA steel were joined using similar technique of asymmetric electrodes [64]. Both studies showed that magnesium alloy was joined to zinc-coated steel by three different mechanisms, i.e., soldering with zinc-based filler material (zinc coating), solid state bonding of magnesium to steel, and weld brazing in the center of the weld [64, 67]. It was observed that no continuous IMC layer was formed at the interface of solid state, and weld brazing region and that the zinc penetrated into the magnesium BM along the grain boundaries in the soldered region. These resulted in the formation of a joint with strength similar to that of Mg/Mg joint. The feasibility of using interlayer during RSW of magnesium alloy to steel was also investigated [68, 69]. Jiang et al. [69] investigated the effect of adding Cu-Zn interlayer on the mechanical properties of resistance spot welded joints between AZ31B magnesium alloy and Q235 mild steel. The average tensile strength of the joint obtained at optimum welding parameters without interlayer was 30 MPa. The addition of Cu-Zn interlayer had an influence on the strength of the joint. An average tensile strength of 44 MPa was obtained when 0.05 mm thick Cu-Zn interlayer was used. The strength increased to 62 MPa as the interlayer thickness was increased to 0.1 mm. However, the strength decreased to 27 MPa when the interlayer thickness was in the range of 0.2 to 0.3 mm. This decrease in strength was associated with the increased interlayer thickness and the high thermal conductance of Cu-Zn interlayer, which would result in high heat input, volatilization of magnesium alloy, and formation of pores. It was found that the metallurgical bonding between the magnesium alloy and steel, with the addition of Cu-Zn interlayer, was due to the formation of CuMgZn intermetallic compounds and solid solutions of Cu in Fe [69]. In another study, Ni interlayer was used in the joining of AZ31B magnesium alloy and Q235 mild steel by RSW, obtaining an average tensile shear strength 75 MPa. The bonding was enhanced by the formation of Mg2Ni intermetallic compound and a solid solution of Ni in Fe at the interface of the center of the nugget zone [68]. A study of the fatigue behavior of magnesium/steel resistance spot welds has shown that crack initiated at the notch root of both steel and Mg sides. However, the cracks

propagated along different directions, as shown in Fig. 13. The crack at the Mg side propagated through the Mg base metal, leading to fracture. However, the crack at the steel side propagated along the magnesium/steel interface and moved only a distance of 400–500 μm, which is extremely small compared to the nugget size (9.4 mm). This shows that the crack propagation rate of the magnesium/steel interface was much lower than that of the Mg base metal. It was also shown

Fig. 13 Mg/steel spot weld after fatigue test at a maximum load of 2.0 kN: a Mg end and b steel end [64]


the FZ. In both the RSW Mg/steel and WB Mg/steel joints, the FZ formed on the magnesium side and its microstructure consisted of equiaxed dendritic and columnar dendritic structure. However, the steel side of RSW Mg/steel underwent microstructural changes, forming a mixture of lath martensite, bainite, pearlite, and retained austenite and consequently significant increase in micro hardness was observed, whereas the microstructure of the WB Mg/steel did not change due to relatively slower cooling rate. Shrinkage cracks were observed in RSW Mg/steel joint in the FZ adjacent to the steel. Such cracks were not observed in the WB Mg/steel due to relatively lower cooling rate which resulted from the use of adhesive. However, a thick layer of IMC consisting of MgZn2 and Mg7Zn3 formed in WB Mg/steel joints while a thinner IMC layer consisting of MgZn2 formed in RSW Mg/steel. Generally, the use of adhesive would improve the mechanical properties of the joint significantly. The maximum tensile shear load and energy absorption of the WB Mg/Mg and WB Mg/steel are higher than for RSW Mg/steel joints, as shown in Fig. 15. Moreover, the adhesive reduced the stress concentration around the nugget and as such, for a given number of cycles, the fatigue strength of WB Mg/Mg and WB Mg/ steel was three times higher than RSW Mg/steel joints. Xu et al. [45] have studied the influence of bonding area size on the microstructures, tensile and fatigue strengths of weld bonded 2.0 mm thick AZ31B-H24 Mg alloy similar joints, using Terokal® 5087-02P (cured at a temperature of 180 °C for 30 min) as the adhesive. Two types of joints with different size of bonding areas were produced: WB-1, which had a bonding overlap area of 35 mm × 35 mm, and WB-0.5, which had a bonding area of 17.5 mm × 35 mm. It was observed that the FZ of both WB-1 and WB-0.5 consisted of typical equiaxed dendritic structures with Mg17Al12 particles at the interdendritic and intergranular regions. Also, for both types of joint, the HAZ was characterized by equiaxed recrystallized grains. However, less solidification cracking or shrinkage was observed in WB-0.5 than in WB-1 joints. This was attributed to slower cooling rate caused by the reduced bonding overlap area. It was also found that, due its larger bonding area, the WB-1 joint was stronger than WB-0.5 joint. The tensile shear load of WB-1 joint and WB-0.5 joint

that Mg/Mg similar and Mg/steel dissimilar resistance spot welded joint had equivalent fatigue resistance; they both exhibited similar crack propagation behavior, and both failed through thickness in the magnesium side. However, the causes of crack initiation were different. For Mg/steel dissimilar joint, the crack initiation was due to the penetration of zinc into the magnesium base metal as a result of liquid metal-induced embrittlement. For Mg/Mg joint, the stress concentration, grain growth, and the existence of Al-rich phases in the grain boundaries of HAZ were responsible for crack initiation [64].

10 Weld-bonding Weld-bonding is an innovative and advanced hybrid joining technology which combines the advantages of RSW and adhesive bonding. It is now widely used in automobile, railway carriages, and aircraft manufacturing industries [70, 71]. Weld bonding produces more desirable joints than either spot welding or adhesive bonding. In addition to reducing the number of welds required in a vehicle, it offers advantages such as reduced manufacturing costs, improved stiffness and loadbearing capacity, enhanced stress distribution, fatigue resistance and crashworthiness, better corrosion resistance, and elimination of the need for sealants [71–75]. In weld bonding, two methods are mainly used, i.e., resistance spot weld bonding (RSWB) and laser weld bonding (LWB) [71]. In RSWB, structural adhesives are applied on the surface of the sheets, followed by RSW, and then curing at a certain temperature for a suitable period of time [73, 74]. The stages of weld-bonding are illustrated in Fig. 14. Xu et al. [66] have studied the microstructure and mechanical behavior of weld-bonded magnesium-to-magnesium joints (WB Mg/Mg) and magnesium-to-steel dissimilar joints (WB Mg/steel) in comparison with resistance spot welded magnesium-to-steel dissimilar joints (RSW Mg/steel). The magnesium alloy, steel, and adhesives used were AZ31B Mg alloy, hot-dip galvanized HSLA, and Terokal® 508702P (cured at a temperature of 180 °C for 30 min), respectively. It was reported that in the WB Mg/Mg joints, equiaxed dendritic and divorced eutectic structures were observed in F

Electrode

Adhesive

Weld nugget Heating F

(a)

(b)

Fig. 14 Stages of weld-bonding: a applying adhesives; b RSW; c curing in an oven

(c)


Fig. 15 Tensile load versus displacement for the WB Mg/Mg similar joint, WB Mg/steel dissimilar joint, and RSW Mg/steel dissimilar joint [66]

was about 18 and 9 kN, respectively. A study of the fatigue behavior of the joints showed that both WB-1 and WB-0.5 joints had equivalent fatigue life at low cycle fatigue regime (higher cyclic stress level). However, at high cycle fatigue regime (lower cyclic stress levels), WB-0.5 joints exhibited longer fatigue life than WB-1 joints due to less solidification or shrinkage cracks in the nugget of WB-0.5 joints. Moreover, at high cyclic load levels, it was observed that cohesive failure occurred along the adhesive layer in combination with partial nugget pull out, whereas at lower cyclic load levels, the failure occurred in the BM.

11 Effects of welding parameters on joint quality The quality and performance of resistance spot welds are influenced by heat input, which is dependent on the welding parameters, primarily welding current, welding time, and electrode force [29, 76–79]. It is important to select the correct welding parameters to produce a reliable and sound joint with the desired geometric features and mechanical properties [35]

properties of 1.2 mm thick AZ31B magnesium alloy, carried out at a constant electrode force of 2.5 kN and 8 cycles welding time, have shown that an increase in welding current from 15 to 23 kA resulted in an increase in nugget diameter (4.2–6.5 mm), with a corresponding increase in joint strength (1.46–3.0 kN) [79, 84]. However, as shown in Fig. 16, welding currents higher than 25 kA resulted in the reduction of joint strength due to excessive heat input, expulsion, and severe electrode indentation [79]. Liu et al. [51] studied the effect of welding current on the microstructure of AZ31 (SA) and AZ31 (SB) magnesium alloys (their composition described in Section 5) and reported that the CDZ became finer and shorter with increasing welding current. At a welding current of 22 k A, the width of the CDZ at the notch of the weld was around 146 μm for AZ31 (SA) and around 390 μm for AZ31 (SB). This width decreased linearly with the increase in welding current and almost disappeared when the welding current surpassed a critical value, which was about 24 kA for AZ31 (SA) and 28 kA for AZ31 (SB). It was also reported that for both alloys the nugget diameter, tension-shear fracture load, and fracture toughness increased with the increase in welding current. However, the fracture toughness and the peak load of AZ31 (SA) were higher than that of AZ31 (SB) for the same nugget size. Similarly, it has been shown that the nugget diameter and joint strength of magnesium/aluminum dissimilar resistance spot welds, with and without interlayers, also increase with the increase in welding current [28, 60–63]. However, there is a critical welding current value beyond which the joint strength would decrease due to formation of extensive IMCs, microstructure deterioration, or expulsion [28, 60, 62]. In a study by Hayat [28], it was reported that the tensile shear load of dissimilar resistance spot welds between 1.7 mm AZ31 magnesium alloy and 1.5 mm 1350 aluminum alloy increased with increasing peak weld current and reached a maximum of 2.75 kN at 29 kA. Above 29 kA, the tensile shear

11.1 Effect of welding current and time Welding current is the most influential parameter in RSW. Generally, a low current results in low heat input, an underdeveloped nugget size, and poor penetration. Increasing the current leads to an increase in heat generation, which in turn leads to an increase in nugget size and tensile shear load [28, 40–43, 51, 56, 79–81]. For magnesium alloys, high welding current, typically 2.5 to 3 times the current for welding steel, and short welding time are required due to their high electrical and thermal conductivities [30, 33, 37, 82, 83]. However, if the welding current is too high, severe expulsion and indentation would occur, leading to reduction in joint strength [40, 41, 79]. Studies on the effect of welding current on the mechanical

Fig. 16 Effects of welding current on joint tensile shear load and nugget diameter in AZ31 Mg alloy [79]


load was decreased due to excessive melting and expulsion on the magnesium side. The thickness of the IMC layer had also increased from 17 μm at 22 kA to 32 μm at 33 kA. A similar observation was made when welding 1.5 mm AZ31B magnesium alloy/AA5052-H12 aluminum by thermo-compensated RSW with a Zn interlayer [60]. As shown in Fig. 17, while the nugget diameter increased over the range of selected current values, the tensile shear load increased to a maximum of 2199 N at 6.5 kA and then decreased significantly with subsequent increase in the current value, due to the formation of bulk IMCs in the nugget [60]. Penner et al. [61, 63] investigated the effect of welding current on nugget diameter and joint strength of AZ31BH24 magnesium alloy and 5754-O aluminum alloy using zinc-coated steel interlayer [63] and Ni-based interlayers [61]. In both studies, it was found that the nugget diameter, on both magnesium and aluminum sides, increased with increase in welding current. However, the nugget on the Mg side was always larger than that on the Al side. This was attributed to differences in electrical resistivity and thermal conductivity. AZ31B magnesium has a greater electrical resistivity value than 5754 aluminum alloy; thus, greater heat generation and larger melted zone would occur in the magnesium alloy. Also, aluminum alloy has higher thermal conductivity and therefore greater heat loss would occur in the aluminum sheet. Hence the smaller nugget size at the Al side of the weld. It was also reported that increase in welding current has increased the tensile shear load [61, 63]. For example, with gold-coated Ni interlayer, at welding current of 16 kA, no joints were formed due to insufficient heat input, while a joint with average peak load of 4.69 kN was obtained at 24 kA [61]. Welding time has a similar, but less profound effect than welding current on heat input. It was reported that a linear relationship exists between tensile shear load and the nugget diameter of resistance spot welds and that increasing either the welding current or time enlarges the diameter [33, 42, 85, 86]. In one study, Lang et al. [79] reported that the nugget diameter

of 1.2 mm AZ31 magnesium alloy resistance spot welds increased from 3.4 to 7.3 mm over welding time of 1 to 16 cycles, at a current of 23 kA. The tensile shear load also increased with the increase in welding time. A low tensile shear load of 1.23 kN was obtained at 1 cycle. For the subsequent 2 to 6 cycles, a significant increase in tensile shear load was recorded. However, from 8 to 16 cycles, only small increments in tensile shear load (2930–3050 N) were obtained, despite the increase in nugget diameter over the longer welding time period. This was attributed to coarsening of the weld nugget microstructure as the welding time increases. Therefore, although it may seem favorable to select welding time longer than 6 cycles for RSW of magnesium alloys, a prolonged welding time is not necessary [79]. A similar observation was reported for dissimilar resistance spot welds between 1.5 mm AZ31B magnesium and 1.2 mm zinccoated DP600 steel, produced at 4 kN electrode force and 20 kAwelding current. It was found that the strength increased with welding time and became nearly constant after eight cycles [67]. Zhou et al. [41] optimized the welding parameters for RSW of 1.5 mm AZ31B magnesium alloy and have proposed a process window, which defined the range of welding current and welding time that will result in acceptable welds. As shown in Fig. 18, the bottom-left boundary of the process window is defined by the minimum strength requirement of 3.315 kN, as per AWS D17.2 specifications, while the topright boundary is defined by severe expulsion and indentation. For example, according to the process window, if the welding current is lower than 22 kA, the heat generation would be insufficient to form a joint that will satisfy the minimum strength requirement. On the other hand, if the welding current is higher than 28 kA, the heat input would be too high, leading to severe expulsion, unacceptable indentation, and reduced weld strength. Hence, the welding current and time should be limited within the boundaries of the process window.

Fig. 17 Effect of welding current on tensile shear performance of Al/Mg joint produced by thermo-compensated RSW with Zn interlayer [60]

Fig. 18 Process window for RSW of AZ31B Mg alloys at a constant electrode force of 4 kN [41]


11.2 Effect of electrode force Electrode force influences the properties of resistance spot welds, primarily because of its effects on contact resistance and contact area. Generally, increasing the electrode force increases the actual metal-to-metal contact, thus decreasing the contact resistance [32, 37, 38]. Although increasing the electrode force can suppress shrinkage voids and expulsion, high electrode force could lead to severe indentation, sheet separation, and distortion. On the other hand, insufficient electrode force would result in high contact resistance, excessive heat generation, expulsion, surface marking, and poor joint quality [29, 37, 82]. It was reported that lowering the electrode force from 4.5 to 1.5 kN had increased the nugget diameter of AZ31 magnesium alloy from 3.2 to 6.9 mm and improved the joint strength, although an electrode force 1.5 kN easily caused expulsion [79]. Figure 19 illustrates the effect of electrode force on the properties of AZ31 magnesium alloy joint produced by RSW with cover plates [37], showing that the tensile shear load of the joint decreased with increasing electrode force. This effect was attributed to the reduction in nugget diameter caused by superior sheet separation and decrease in the energy density of the welding region [37]. Thus, it is unfavorable to select high electrode force for RSW of magnesium alloys [79].

12 Main metallurgical defects in RSW of magnesium alloys 12.1 Expulsion Expulsion is the ejection of molten metal from the nugget area during RSW. It usually occurs due to the application of high

Fig. 19 Effects of electrode force on the nugget diameter tensile shear load of the joints [37]

welding current for a short period of time. It has negative effects on both the appearance and performance of joints and is therefore undesirable [87]. Expulsion can be categorized into two types, namely, surface expulsion and interfacial expulsion [35]. Surface expulsion occurs at the electrode-sheet interface and is mainly caused by localized surface heating caused by non-uniform distribution of contact resistance between the electrode and sheet [82]. It is detrimental to surface quality and electrode life [35]. Interfacial expulsion occurs at the faying interface when the pressure from the liquid nugget exceeds the electrode force [37, 82]. It compromises the strength of the joint because it involves loss of liquid from the nugget [35]. The most significant factor that induces expulsion is the welding current, followed by electrode force [88]. The risk of expulsion is higher in magnesium and aluminum alloys as compared to steels due to the high welding current required to weld them [35]. Figure 20 shows surface expulsion in 1.2 mm AZ31 magnesium alloy, which occurred at welding current of 25 kA, 8 cycles welding time, and 2.5 kN electrode force. In addition to surface expulsion, severe indentation was also observed in the joint [79]. Luo et al. [82] have studied the characteristics of AZ31 and AZ91 magnesium alloys during RSW. It was reported that the two alloys exhibited different behaviors. Although interfacial expulsion occurred in AZ31B magnesium alloy, liquation cracking and surface expulsion were the main problems. The electrode life in welding AZ31B magnesium alloy was greatly affected due to surface expulsion. Thus electrode dressing was generally required after about 10 welds have been made. The expulsion has also created voids in the weld joints. The expulsion in AZ91D magnesium alloy was quite different than in AZ31B magnesium alloy and aluminum alloys. It was suggested that in AZ91D magnesium alloy, a liquid network of molten grain boundaries is formed in the partially melting zone, which served as a pathway for metal ejection. Large amounts of cracks and voids were formed in the nugget due

Fig. 20 Expulsion in AZ31 magnesium alloy [79]


to expulsion. It has been reported that the expulsion and void formation in AZ31B magnesium alloy can be suppressed by increasing the electrode force [37, 56]. However, this is not the case for AZ91D, increasing the electrode force did not effectively suppress expulsion in the material. 12.2 Porosity The presence of porosity in the fusion zone of magnesium alloy resistance spot welds has been reported in the literature [12, 14, 37, 42, 89]. Pore formation in RSW of magnesium alloys may be influenced by factors such as surface contamination, hydrogen rejection during solidification [14], preexisting pores in base material, shrinkage strain, and expulsion [12, 89]. Tu et al. [89] and Qiu et al. [12] studied the influence of these factors on pore formation during RSW of AZ31B magnesium alloys. The shrinkage strain and expulsion were identified as key factors in pore formation and that hydrogen rejection resulted in the formation of small-sized pores (5– 30 μm). During RSW, the surrounding solid metal constrains the expansion of molten metal, subjecting it to shrinkage strain or high hydropressure that occasionally result in the splash of molten metal. These caused insufficient magnesium in the molten weld cavity and subsequently produced pores in the center of the nugget [89]. In addition, during the welding process, hydrogen can be absorbed into the molten metal from the atmosphere, due to the hydroscopic nature of the surface of magnesium [90, 91]. Since the solubility of hydrogen in magnesium is significantly higher in the liquid form than in solid form, hydrogen is rejected during solidification in the form of bubbles leading to pore formation in the welded joints [12, 82]. Porosity could affect the strength of welds under monotonic and high cyclic loading conditions [37, 42, 89–91]. For example, as shown in Fig. 21, Qiu et al. [37] observed several pores on the shear fracture surface of AZ31 magnesium alloy, suggesting that cracks propagated across the pores during tensile-shear test. As a result, the tensile shear strength of the joint was compromised. However, increasing the electrode force decreased the pore fraction, with the presence of pores almost eliminated once the electrode force exceeded 5 kN [37]. 12.3 Liquation cracking Luo et al. [82] reported that liquation cracking occurred frequently in RSW of AZ31B magnesium alloys. Liquation cracking was found to be intergranular and always originated in HAZ close to the nugget. The cracks occurred in the form of whiskers and could compromise the integrity of the joint. It was found that liquation cracking always occurred when the nugget size is above a critical value. For example, for a 2-mmthick AZ31B magnesium alloy, almost no cracks were

Fig. 21 Fracture surface of AZ31 magnesium alloy showing porosity [37]

observed when the nugget diameter was less than 5 mm. However, liquation cracking occurred in all welds with a nugget diameter of 5 mm and above. The extent of liquation cracking was also found to depend strongly on heat input. Moderate current produced only a few numbers of narrow cracks, whereas a high current would generate large numbers of wide cracks. Figure 22 depicts the narrow and wide cracks in AZ31B magnesium alloy welds. It was also found that the presence of expulsion would promote liquidation cracking. Similar to expulsion, increasing the electrode force was found to effectively suppress liquation cracking.

13 Electrode degradation Short electrode life is one of the greatest problems during RSW of magnesium alloys [82, 83, 92]. It was reported that the electrode life in welding AZ31B magnesium alloy is greatly affected by surface expulsion, requiring frequent electrode dressing [82]. Repeated contacts between the workpiece and electrode at high temperature usually lead to surface contamination, alloy pickup, and pitting, leading to poor quality joints. The condition worsens with increasing number of welds [31]. Lang et al. [83], in their study of the electrode degradation mechanism during RSW of hot-extruded AZ31 magnesium alloy, have found that the degradation mechanism involved four basic stages. The first stage, known as magnesium alloy pickup, began at the first weld and involved the transfer of tiny drops of molten magnesium alloy from the magnesium sheet surface to the electrode tip face. In the second stage, the molten magnesium alloy reacts with the electrode material, forming a complex alloy layer of Cu2 Mg and CuMg2 on the electrode tip face. In the third stage, particles are removed from the electrode tip face due to brittle fracture of the

Author's personal copy Int J Adv Manuf Technol Fig. 22 Liquation cracking in AZ31B welds: a narrow crack; b wide cracks [82]

complex alloy formed in second stage. Pits are formed on the electrode tip face. In the final stage, smaller pits combined to form a network of large cavities on the tip face, as shown in Fig. 23. It was observed that periodic cleaning of the electrode tip face would limit the accumulation of magnesium element on the tip and hence significantly improve the electrode life [83]. The cost of dressing or replacing electrodes contributes significantly to the cost of spot welding [93]. Improvement in electrode life would, therefore, enhance weld quality, reliability, and cost effectiveness [94]. 13.1 Future trend Compared to RSW of steel and aluminum alloys, the RSW of magnesium alloys is still in its infancy. Most of the research conducted to date is on AZ series magnesium alloys, especially AZ31. Welding behavior of other magnesium alloys such as AM, AZM, and ZK series should also be studied. The weldability of similar and dissimilar combinations of these alloys should be investigated. Currently, steel is the primary structural material in the automotive industry. The number of research conducted on dissimilar joining of magnesium alloys to steels is still very small. Furthermore, advanced high strength steels (AHSS) such as twinning-induced plasticity steels and transformation-induced plasticity steels are being developed and are being taken up by the industry. Thus, the weldability of magnesium alloys to AHSS should also be studied further. Fig. 23 Damaged electrode tip faces a top electrode; b bottom electrode [83]

The use of interlayers during RSW of magnesium to aluminum alloys appears to be promising in suppressing the formation of intermetallic compounds. So far, the use of pure Ni interlayer and gold-coated Ni interlayer produced joint with the highest strengths. Alternative interlayers with lower cost and higher availability should be tested. Other techniques such as the thermo-compensated RSW should also be investigated further. Fatigue is a common cause of structural failure. The fatigue behavior of spot welds of magnesium alloys and magnesium alloys to other materials should be studied further to guarantee the safety and crashworthiness of automotive structures. The weld bonding technology which is found to improve the mechanical properties of magnesium alloys should be studied extensively. The feasibility of using weld bonding to join magnesium to aluminum alloys and its influence on the formation of intermetallic compounds should also be studied. Corrosion resistance is crucial to the life of automobiles; thus, it would be important to evaluate the corrosion behavior of magnesium alloys spot welds in detail.

14 Summary Successful incorporation of magnesium alloys into the automotive structures would lead to significant weight savings and improved fuel efficiency due to their low density and high specific strength. The resistance spot-welding of magnesium


alloys is usually characterized by the formation of columnar dendritic structure and equiaxed dendritic structure in the fusion zone. The columnar dendritic structure is undesirable due to its unfavorable effects on mechanical properties. Therefore, there is a need to promote CET for enhanced mechanical properties. It was shown that the pre-existence of coarse second phases in the base metal, the addition of particles such as titanium powder, and resistance spot welding under the influence of electromagnetic stirring effect could lead to enhanced CET and, hence, better mechanical properties. The fatigue behavior of magnesium alloys was also reviewed. It was shown that the fatigue resistance of magnesium alloys is comparable to that of aluminum and much less than that of steel. The welding of magnesium to steel is particularly challenging due to significant differences in properties between them. Usage of domed-shaped electrode on the magnesium side and flat electrode on the steel side to balance the heating has been successful in joining magnesium to steel pair, producing welds with strength comparable to that of Mg/Mg joints. Weld-bonding, which combines resistance welding and adhesive bonding, was found to significantly improve the mechanical properties of magnesium alloys spot welds both under static and dynamic loading conditions. Direct resistance spot welding of magnesium to aluminum produced joints with poor strengths due to the formation of brittle IMCs. The use of interlayers such as pure nickel, zinccoated steel, and gold-coated nickel was found to suppress the formation of IMCs and thus significantly improve the joints strength. Welding parameters, such as electrode force, welding time, and most importantly, welding current, have great influence on joint quality in RSW of magnesium alloys. High welding current, typically 2.5 to 3 times the current for welding steel, and short welding time are required due to the high electrical and thermal conductivities of magnesium alloys. For optimized joint strength, the welding parameters should be within certain critical range, to avoid poor joint strength, severe expulsion, and unacceptable indentations.

2.

3. 4.

5.

6.

7.

8.

9.

10.

11. 12.

13.

14.

15.

16. 17.

Acknowledgments The authors would like to acknowledge the University of Malaya for providing the necessary facilities and resources for this research. This research is supported by High Impact Research Grant (UM/MOHEUM.C/625/1/HIR/MOHE/H16001-D000001) from the Ministry of Higher Education Malaysia and partially funded by University Malaya Research Officer Grant Scheme (ROGS) (BR001-2014)

18.

19.

References 1.

D’Errico F, Plaza GG, Giger F, Kim SK (2013) Final assessment of preindustrial solid-state route for high-performance Mg-system alloys production: concluding the EU green metallurgy project. JOM 65(10):1293–1302

20.

D’Errico F, Ranza L (2015) Comparative environmental benefits of lightweight design in the automotive sector: the case study of recycled magnesium against CFRP and steel. Paper presented at the Magnesium Technology 2015 Orłowicz A, Mróz M, Tupaj M, Trytek A (2015) Materials used in the automotive industry. Arch Foundry Eng 15(2):75–78 Luo AA (2013) 8 - Applications: aerospace, automotive and other structural applications of magnesium. In: Pekguleryuz MO, Kainer KU, Kaya AA (eds) Fundamentals of magnesium alloy metallurgy. Woodhead Publishing, Cambridge, pp. 266–316 Palanivel S, Nelaturu P, Glass B, Mishra R (2015) Friction stir additive manufacturing for high structural performance through microstructural control in an Mg based WE43 alloy. Mater Des 65: 934–952 Liu L (2010) 1 - Introduction to the welding and joining of magnesium. In: Liu L (ed) Welding and joining of magnesium alloys. Woodhead Publishing, Cambridge, pp. 3–8 Zhang LJ, Zhang XJ, Ning J, Zhang JX (2015) Modulated fiber laser welding of high reflective AZ31. Int J Adv Manuf Tech 76(14):721–733 Patel VK, Bhole SD, Chen DL (2013) Formation of zinc interlayer texture during dissimilar ultrasonic spot welding of magnesium and high strength low alloy steel. Mater Des 45:236–240 Zhao Y, Lu ZP, Yan K, Huang LZ (2015) Microstructural characterizations and mechanical properties in underwater friction stir welding of aluminum and magnesium dissimilar alloys. Mater Des 65:675–681 Zhang N, Wang WX, Cao XQ, Wu JQ (2015) The effect of annealing on the interface microstructure and mechanical characteristics of AZ31B/AA6061 composite plates fabricated by explosive welding. Mater Des 65:1100–1109 Kulekci MK (2008) Magnesium and its alloys applications in automotive industry. Int J Adv Manuf Technol 39(9-10):851–865 Qiu R, Wang N, Shi H, Zhang K, Satonaka S (2014) Nonparametric effects on pore formation during resistance spot welding of magnesium alloy. Sci Technol Weld Joining 19(3):231–234 Li L, Tan C, Chen Y, Guo W, Mei C (2013) CO 2 laser welding– brazing characteristics of dissimilar metals AZ31B Mg alloy to Zn coated dual phase steel with Mg based filler. J Mater Process Technol 213(3):361–375 Babu NK, Brauser S, Rethmeier M, Cross C (2012) Characterization of microstructure and deformation behaviour of resistance spot welded AZ31 magnesium alloy. Mater Sci Eng, A 549:149–156 Cole G Magnesium vision 2020—a North American automotive strategic vision for magnesium. In: IMA proceedings, 2007. International Magnesium Association, p. 13 Luo AA (2013) Magnesium casting technology for structural applications. J Magnes Alloys 1(1):2–22 Alizadeh-Sh M, Marashi SPH, Pouranvari M (2014) Resistance spot welding of AISI 430 ferritic stainless steel: phase transformations and mechanical properties. Mater Des 56:258–263 Charde N, Yusof F, Rajkumar R (2014) Material characterizations of mild steels, stainless steels, and both steel mixed joints under resistance spot welding (2-mm sheets). Int J Adv Manuf Tech 75(1-4):373–384 Eshraghi M, Tschopp MA, Zaeem MA, Felicelli SD (2014) Effect of resistance spot welding parameters on weld pool properties in a DP600 dual-phase steel: a parametric study using thermomechanically-coupled finite element analysis. Mater Des 56:387–397 Baca N, Ngo TT, Conner RD, Garrett SJ (2013) Small scale resistance spot welding of Cu47Ti34Zr11Ni8 (Vitreloy 101) bulk metallic glass. J Mater Process Technol 213(11):2042–2048

Author's personal copy Int J Adv Manuf Technol 21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35. 36.

37.

38. 39.

40.

Li YB, Wei ZY, Li YT, Shen Q, Lin ZQ (2013) Effects of cone angle of truncated electrode on heat and mass transfer in resistance spot welding. Int J Heat Mass Tran 65:400–408 Niknejad S, Liu L, Lee MY, Esmaeili S, Zhou NY (2014) Resistance spot welding of AZ series magnesium alloys: effects of aluminum content on microstructure and mechanical properties. Mat Sci Eng a-Struct 618:323–334 Wei PS, Wu TH Nugget shape control in resistance spot welding. In: Microsystems, Packaging, Assembly and Circuits Technology Conference (IMPACT), 2013 8th International, 2013. IEEE, pp 283-285 Cukovic JP, Klopcic B, Petrun M, Polajzer B, Dolinar D (2014) Optimization of resistance spot welding transformer windings using analytical successive approximation and differential evolution. Ieee T Magn 50(4) Florea RS, Bammann DJ, Yeldell A, Solanki KN, Hammi Y (2013) Welding parameters influence on fatigue life and microstructure in resistance spot welding of 6061-T6 aluminum alloy. Mater Des 45: 456–465 Hassanifard S, Feyzi M (2015) Analytical solution of temperature distribution in resistance spot welding. J Mech Sci Technol 29(2): 777–784 Hamidinejad SM, Kolahan F, Kokabi AH (2012) The modeling and process analysis of resistance spot welding on galvanized steel sheets used in car body manufacturing. Mater Des 34:759–767 Hayat F (2011) The effects of the welding current on heat input, nugget geometry, and the mechanical and fractural properties of resistance spot welding on Mg/Al dissimilar materials. Mater Des 32(4):2476–2484 Pouranvari M, Marashi S (2013) Critical review of automotive steels spot welding: process, structure and properties. Sci Technol Weld Joining 18(5):361–403 Satonaka S, Iwamoto C, Murakami GI, Matsumoto Y (2012) Resistance spot welding of magnesium alloy sheets with cover plates. Weld World 56(7-8):44–50 Wei PS, Wu TH, Chen LJ (2013) Joint quality affected by electrode contact condition during resistance spot welding, components, packaging and manufacturing technology. IEEE Trans 3(12): 2164–2173 Liu L, Feng J, Zhou Y (2010) 18 - Resistance spot welding of magnesium alloys. In: Liu L (ed) Welding and joining of magnesium alloys. Woodhead Publishing, Cambridge, pp. 351–367e Qiu RF, Satonaka S, Iwamoto C (2009) Mechanical properties and microstructures of magnesium alloy AZ31B joint fabricated by resistance spot welding with cover plates. Sci Technol Weld Joining 14(8):691–697 Wang Y, Mo Z, Feng J, Zhang Z (2007) Effect of welding time on microstructure and tensile shear load in resistance spot welded joints of AZ31 Mg alloy. Sci Technol Weld Join 12(8):671–676 Zhang H, Senkara J (2011) Resistance welding: fundamentals and applications. CRC Press, Boca Raton Williams N, Parker J (2004) Review of resistance spot welding of steel sheets part 1 modelling and control of weld nugget formation. Int Mater Rev 49(2):45–75 Qiu RF, Shi HX, Yu H, Zhang KK, Tu YM, Satonaka S (2010) Effects of electrode force on the characteristic of magnesium alloy joint welded by resistance spot welding with cover plates. Mater Manuf Process 25(11):1304–1308 RWMA (2003) Resistance welding manual. Resistance Welder Manufacturer’s Association, Philadelphia Liu L (2010) 3—Preparation for welding of magnesium alloys. In: Liu L (ed) Welding and joining of magnesium alloys. Woodhead Publishing, Cambridge, pp. 16–22 Liu L, Zhou S, Tian Y, Feng J, Jung J, Zhou Y (2009) Effects of surface conditions on resistance spot welding of Mg alloy AZ31. Sci Technol Weld Join 14(4):356–361

41.

Zhou S, Liu L, Jung JP, Lee M-Y, Zhou YN (2010) Effects of welding parameters and surface pretreatments on resistance spot welding of AZ31B Mg alloy. Met Mater Int 16(6):967–974 42. Behravesh SB, Jahed H, Lambert S (2011) Characterization of magnesium spot welds under tensile and cyclic loadings. Mater Des 32(10):4890–4900 43. Xiao L, Liu L, Esmaeili S, Zhou Y (2012) Microstructure refinement after the addition of titanium particles in AZ31 magnesium alloy resistance spot welds. Metal Mater Trans A 43(2):598–609 44. Niknejad ST, Liu L, Nguyen T, Lee M-Y, Esmaeili S, Zhou NY (2013) Effects of heat treatment on grain-boundary β-Mg17Al12 and fracture properties of resistance spot-welded AZ80 Mg alloy. Metal Mater Trans A 44(8):3747–3756 45. Xu W, Liu L, Zhou Y, Mori H, Chen D (2013) Tensile and fatigue properties of weld-bonded and adhesive-bonded magnesium alloy joints. Mater Sci Eng, A 563:125–132 46. Feng JC, Wang YR, Zhang ZD (2006) Nugget growth characteristic for AZ(31)B magnesium alloy during resistance spot welding. Sci Technol Weld Joining 11(2):154–162 47. Wang Y, Feng J, Zhang Z (2006) Microstructure characteristics of resistance spot welds of AZ31 Mg alloy. Sci Technol Weld Join 11(5):555–560 48. Xiao L, Liu L, Chen DL, Esmaeili S, Zhou Y (2011) Resistance spot weld fatigue behavior and dislocation substructures in two different heats of AZ31 magnesium alloy. Mater Sci Eng, A 529: 81–87 49. Xiao L, Liu L, Zhou Y, Esmaeili S (2010) Resistance-spot-welded AZ31 magnesium alloys: part I. Dependence of fusion zone microstructures on second-phase particles. Metal Mater Trans A 41(6): 1511–1522 50. Yao Q, Luo Z, Li Y, Yan F, Duan R (2014) Effect of electromagnetic stirring on the microstructures and mechanical properties of magnesium alloy resistance spot weld. Mater Des 63:200–207 51. Liu L, Xiao L, Feng JC, Tian YH, Zhou SQ, Zhou Y (2010) Resistance spot welded AZ31 magnesium alloys, part II: effects of welding current on microstructure and mechanical properties. Metall Mater Trans A 41a(10):2642–2650 52. Radakovic D, Tumuluru M (2008) Predicting resistance spot weld failure modes in shear tension tests of advanced high-strength automotive steels. Weld J 87(4):96 53. Pereira A, Ferreira J, Antunes F, Bártolo P (2014) Assessment of the fatigue life of aluminium spot-welded and weld-bonded joints. J Adhes Sci Technol 28(14-15):1432–1450 54. Patel VK, Bhole SD, Chen DL (2014) Fatigue life estimation of ultrasonic spot welded Mg alloy joints. Mater Des 62:124–132 55. Behravesh SB, Jahed H, Lambert S (2014) Fatigue characterization and modeling of AZ31B magnesium alloy spot-welds. Int J Fatigue 64:1–13 56. Shi HX, Qiu RF, Zhu JH, Zhang KK, Yu H, Ding GJ (2010) Effects of welding parameters on the characteristics of magnesium alloy joint welded by resistance spot welding with cover plates. Mater Des 31(10):4853–4857 57. Liu L, Ren D, Liu F (2014) A review of dissimilar welding techniques for magnesium alloys to aluminum alloys. Materials 7(5): 3735–3757 58. Cao X, Jahazi M, Immarigeon J, Wallace W (2006) A review of laser welding techniques for magnesium alloys. J Mater Process Technol 171(2):188–204 59. Luo Y, Li JL (2014) Analysis of nugget formation during resistance spot welding on dissimilar metal sheets of aluminum and magnesium alloys. Metall Mater Trans A 45a(11):5107–5113 60. Zhang Y, Luo Z, Li Y, Liu Z, Huang Z (2015) Microstructure characterization and tensile properties of Mg/Al dissimilar joints manufactured by thermo-compensated resistance spot welding with Zn interlayer. Mater Des 75:166–173

Author's personal copy Int J Adv Manuf Technol 61.

Penner P, Liu L, Gerlich A, Zhou Y (2013) Feasibility study of resistance spot welding of dissimilar Al/Mg combinations with Ni based interlayers. Sci Technol Weld Joining 18(7):541–550 62. Sun M, Niknejad S, Zhang G, Lee M, Wu L, Zhou Y (2015) Microstructure and mechanical properties of resistance spot welded AZ31/AA5754 using a nickel interlayer. Mater Des 87:905–913 63. Penner P, Liu L, Gerlich A, Zhou Y, Penner P, Liu L, Gerlich A, Zhou Y (2014) Dissimilar resistance spot welding of aluminum to magnesium with Zn-coated steel interlayers. Weld J 93(6):225s– 231s 64. Liu L, Xiao L, Chen D, Feng J, Kim S, Zhou Y (2013) Microstructure and fatigue properties of Mg-to-steel dissimilar resistance spot welds. Mater Des 45:336–342 65. Ren DX, Liu LM (2014) Interface microstructure and mechanical properties of arc spot welding Mg-steel dissimilar joint with Cu interlayer. Mater Des 59:369–376 66. Xu W, Chen D, Liu L, Mori H, Zhou Y (2012) Microstructure and mechanical properties of weld-bonded and resistance spot welded magnesium-to-steel dissimilar joints. Mater Sci Eng, A 537:11–24 67. Liu L, Xiao L, Feng JC, Tian YH, Zhou SQ, Zhou Y (2010) The mechanisms of resistance spot welding of magnesium to steel. Metall Mater Trans A 41a(10):2651–2661 68. Zhang G, Tian C, Liu L, Xu R (2014) Interfacial structure of the joints between magnesium alloy and mild steel with nickel as interlayer. Paper presented at the International Conference on Logistics Engineering, Management and Computer Science (LEMCS 2014) 69. Jiang X, Yu H, Lin Q, Xu R (2015) Effects of Cu-Zn alloy additions on microstructure and strength of welded joints between magnesium alloy and mild steel. Paper presented at the International Conference on Mechatronics, Electronic, Industrial and Control Engineering (MEIC 2015) 70. Tao W, Ma Y, Chen Y, Li L, Wang M (2014) The influence of adhesive viscosity and elastic modulus on laser spot weld bonding process. Int J Adhes Adhes 51:111–116 71. Liu L, Ren D, Li Y (2011) Static mechanics analyses of different laser weld bonding structures in joining AZ61 Mg alloy. Int J Adhes Adhes 31(7):660–665 72. Shen J, Zhang Y, Lai X, Wang P (2012) Adhesive placement in weld-bonding multiple stacks of steel sheets. Weld J 91-2: 59–66 73. Zhang Y, Sun H, Wang P-C, Chen G (2014) Improvement of process robustness in weld bonding of galvanized DP780 steel. Weld J 93(12):472S–481S 74. Sam S, Shome M (2010) Static and fatigue performance of weld bonded dual phase steel sheets. Sci Technol Weld Join 15(3):242– 247 75. Khan MF, Sharma G, Dwivedi D (2015) Weld-bonding of 6061 aluminium alloy. Int J Adv Manuf Technol 78(5-8):863–873 76. Emre HE, Kaçar R (2015) Development of weld lobe for resistance spot-welded TRIP800 steel and evaluation of fracture mode of its weldment. The International Journal of Advanced Manufacturing Technology:1-11 77. Jagadeesha T, Jothi TS (2015) Studies on the influence of process parameters on the AISI 316L resistance spot-welded specimens. The International Journal of Advanced Manufacturing Technology:1-16

78.

Ashtiani HRR, Zarandooz R (2015) Microstructural and mechanical properties of resistance spot weld of Inconel 625 supper alloy. The International Journal of Advanced Manufacturing Technology: 1-13 79. Lang B, Sun D, Li G, Qin X (2008) Effects of welding parameters on microstructure and mechanical properties of resistance spot welded magnesium alloy joints. Sci Technol Weld Join 13(8): 698–704 80. Ghazanfari H, Naderi M (2013) Influence of welding parameters on microstructure and mechanical performance of resistance spot welded high strength steels. Acta Metall Sin-Engl 26(5):635–640 81. Uslu P, Demir B, Hayat F (2011) Effect of the weld current on tensile shear properties of the RSW junctions of the AZ31 Mg alloy sheet. Paper presented at the 6th International Advanced Technologies Symposium (IATS’11), Elazığ, Turkey 82. Luo H, Hao C, Zhang J, Gan Z, Chen H, Zhang H (2011) Characteristics of resistance welding magnesium alloys AZ31 and AZ91. Weld J 90:249–257 83. Lang B, Sun DQ, Li GZ, Zhu BQ (2009) Electrode degradation in resistance spot welding of magnesium alloy. Isij Int 49(11):1744– 1748 84. Sun D, Lang B, Sun D, Li J (2007) Microstructures and mechanical properties of resistance spot welded magnesium alloy joints. Mater Sci Eng, A 460:494–498 85. Aslanlar S, Ogur A, Ozsarac U, Ilhan E (2008) Welding time effect on mechanical properties of automotive sheets in electrical resistance spot welding. Mater Des 29(7):1427–1431 86. Kaya Y, Kahraman N (2012) The effects of electrode force, welding current and welding time on the resistance spot weldability of pure titanium. Int J Adv Manuf Technol 60(1-4):127–134 87. Ghazanfari H, Naderi M (2014) Expulsion characterization in resistance spot welding by means of a hardness mapping technique. Int J Miner Metall Mater 21(9):894–897 88. Yi L, Rui W, Xiaojian X, Yang Z (2015) Expulsion analysis of resistance spot welding on zinc-coated steel by detection of structure-borne acoustic emission signals. The International Journal of Advanced Manufacturing Technology:1-8 89. Tu YM, Qiu RF, Shi HX, Yu H, Zhang KK (2011) Analyses of influencing factors of pore formation during resistance spot welding of magnesium alloy. Mater Process Technol Pts 1-4:291–294 90. Harooni M, Ma J, Carlson B, Kovacevic R (2015) Two-pass laser welding of AZ31B magnesium alloy. J Mater Process Technol 216: 114–122 91. Harooni M, Carlson B, Strohmeier BR, Kovacevic R (2014) Pore formation mechanism and its mitigation in laser welding of AZ31B magnesium alloy in lap joint configuration. Mater Des 58:265–276 92. Choi D-S, Kim D-C, Kang M-J (2013) Resistance spot welding characteristics of Mg alloy using process tape. J Weld Join 31(3): 49–53 93. Mathers G (2002) The welding of aluminium and its alloys. Woodhead Publishing, Cambridge 94. Zou J, Zhao Q, Chen Z (2009) Surface modified long-life electrode for resistance spot welding of Zn-coated steel. J Mater Process Technol 209(8):4141–4146