Resistance spot welding of Mg to electro-galvanized steel with hot-dip galvanized steel interlayer

Resistance spot welding of Mg to electro-galvanized steel with hot-dip galvanized steel interlayer

Accepted Manuscript Title: Resistance spot welding of Mg to electro-galvanized steel with hot-dip galvanized steel interlayer Author: Yueqiao Feng Yan...

2MB Sizes 0 Downloads 22 Views

Accepted Manuscript Title: Resistance spot welding of Mg to electro-galvanized steel with hot-dip galvanized steel interlayer Author: Yueqiao Feng Yang Li Zhen Luo Zhanxiang Ling Zhengming Wang PII: DOI: Reference:

S0924-0136(16)30147-9 http://dx.doi.org/doi:10.1016/j.jmatprotec.2016.05.015 PROTEC 14817

To appear in:

Journal of Materials Processing Technology

Received date: Revised date: Accepted date:

22-11-2015 13-4-2016 11-5-2016

Please cite this article as: Feng, Yueqiao, Li, Yang, Luo, Zhen, Ling, Zhanxiang, Wang, Zhengming, Resistance spot welding of Mg to electro-galvanized steel with hot-dip galvanized steel interlayer.Journal of Materials Processing Technology http://dx.doi.org/10.1016/j.jmatprotec.2016.05.015 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title page

Resistance spot welding of Mg to electro-galvanized steel with hot-dip galvanized steel interlayer

Yueqiao Fenga,b, Yang Lia,b ,Zhen Luoa,b*, Zhanxiang Linga,b , Zhengming Wanga,b

a

School of Materials Science and Engineering, Tianjin University, Tianjin, China

b

Tianjin Key Laboratory of Advanced Joining Technology, Tianjin University, Tianjin,

China

Yueqiao Feng, first author, master student, Email: [email protected] Yang Li, Post-doctor, Email: [email protected] Zhen Luo, corresponding author, Professor, Email: [email protected] Zhanxiang Ling, master student, Email: [email protected] Zhengming Wang, undergraduate, Email: [email protected]

*Corresponding author: Prof. Zhen Luo E-mail address: [email protected] Tel.: +86-22-27406602 Fax: +86-22-27406602 Postal address: 25-C-1201, School of Materials Science and Engineering Tianjin University No.92 Weijin Road Tianjin, China, 300072

1

Abstract An AZ31 magnesium alloy and electro-galvanized DP600 steel were joined via resistance spot welding (RSW) with a hot-dip galvanized Q235 steel interlayer, and the microstructure and tensile properties of the weld joint with and without the interlayer were compared. The tensile properties of the weld joint with the interlayer were superior to the joint without the interlayer. With the optimal welding parameters, the peak load increased by 30% and the energy absorption was twice that of the traditional RSW joint. During welding, the hot-dip zinc coating on the interlayer melted and was squeezed out of the welding zone under the electrode force, where it assembled out of the nugget to form a soldered region. This soldered region acted as a mechanical seal, which improved the solid-state bonding and braze-welding in the weld joint. This mechanical seal effect cannot form during RSW without an interlayer because the thin and compact characteristics of the electro-galvanized zinc coating make it difficult to squeeze out.

Abbreviations used: EDX (energy-dispersive X-ray) HAZ (heat-affected zone) HDG (hot-dip galvanized) HSLA (high-strength low-alloy) RSW (resistance spot welding) SEM (scanning electron microscope)

Keywords: Resistance spot welding; AZ31; Electro-galvanized DP600; Hot-dip galvanized Q235; Interlayer

2

Introduction Resistance spot welding (RSW) is an important joining technique in the automotive industry because it is economical, fast and possesses robust features; and typically 4000–6000 RSW processes are used in each automobile (Zhang et al., 2015). Magnesium (Mg) alloys are attractive lightweight materials because of their low density, excellent dimensional stability, high specific strength and acceptable process ability (Hamu et al., 2009). Dissimilar RSW of Mg to steel has become a popular issue currently, and especially in the automotive industry, though this technique has some challenges. The difficulties arising when welding Mg to steel are the large difference in the melting points of the two metals, the low intersolubility between Mg and Fe, and the large differences in electrical/thermal conductivity (Liu et al., 2011). Several works regarding dissimilar RSW of magnesium alloy to steel have been published. Liu et al. (2013) have investigated the microstructure and fatigue properties of AZ31B Mg alloy and hot-dip galvanized (HDG) high-strength low-alloy (HSLA) steel RSW. Their results have shown that the microstructure of the Mg/steel dissimilar spot welds was different from that of Mg/Mg welds, though the welds had an equivalent fatigue resistance owing to similar crack propagation characteristics and failure modes. Xu et al. (2012) have also studied the RSW and weld-bonding of AZ31B-H24 Mg alloy and HDG HSLA steel, while Liu et al. (2010) have found that a hot-dipped zinc coating played a significant role in the joining of magnesium alloy to steel. From these works, it can be seen that most of the current research has focused on the RSW of Mg alloy to steel with a hot-dip Zn coating. However, the RSW of Mg alloy to electro-galvanized steel, which is also widely used in the automotive industry, has been neglected. The zinc coating present on electro-galvanized steel is finer, denser, thinner, more uniform and has better corrosion resistance than the hot-dip zinc coating. Although the resistance spot weldability of a similar electro-galvanized steel is better than that of HDG steel, the zinc coating on the electro-galvanized steel does not help significantly during dissimilar welding of electro-galvanized steel to Mg alloy. In this paper, electro-galvanized DP 600 steel and AZ31Mg alloy were welded using RSW with HDG Q235 steel as an interlayer. The microstructure and tensile shear property of the weld joint were analyzed and compared with the traditional RSW process.

3

Experimental Methods The materials used in this work were 2-mm-thick AZ31B Mg alloy and 1.2-mm-thick electro-galvanized DP600 steel. The interlayer was 0.6-mm-thick HDG Q235 and the thickness of its zinc coating was about 10 µm. The chemical compositions of all of these elements are listed in Table 1.

According to standards GB 2651-88 (Tensile test standard on welded joints, China), the specimen dimension for the resistance welding process and tensile shear test is shown in Fig. 1. Before welding, the surface oxides on the AZ31 Mg alloy were cleaned with abrasive paper, and any grease present on the steel surface was removed with acetone. Welding was conducted on a medium-frequency direct current RSW machine with the capability of a 2–22 kA welding current. A water-cooled domed electrode with a sphere radius of 50 mm and a face diameter of 20 mm was used on the Mg side, while a conical electrode with 8-mm-diameter tip was used on the steel side. Following preliminary test results, 150 ms was selected as the welding time for the specimen with an interlayer. As a comparison, traditional RSW was also conducted and 200 ms was selected as the welding time. A welding current varying from 10–20 kA in 2 kA increments and an electrode force of 3600 N were used for all welding experiments. Three specimens were welded under each set of welding parameters for each type of joint. The joint with an interlayer made using a welding current of 18 kA and a welding time of 150 ms (18 kA–150 ms) and the traditional RSW joint made using a welding current of 16 kA and a welding time of 200 ms (16 kA–200 ms) were chosen to investigate the microstructure. To investigate the weld formation process of the joint with an interlayer, the welding time was chosen to start from 35 ms and to gradually increase to 150 ms at 18 kA according to preliminary test results. The weld cross-sections were prepared for metallographic analysis using standard metallographic procedures. The macrostructures were observed using an Olympus SZX12 stereomicroscope and the microstructures of the welds were observed using an Olympus GX51 metallographic microscope. Vickers hardness tests were carried out on polished and etched specimens, and the tensile shear tests were carried out on a CSS-44100 material test system at a speed of 1 mm min−1. An S4800 4

scanning electron microscope (SEM) with energy dispersive X-ray (EDX) capability was used to observe the tensile fractured surfaces as well as the microstructure at the weld interface.

Results and Discussion Macrostructure and microstructure Figure 2 shows a typical macrostructure and microstructure of the joint with an interlayer. In Fig. 2(a), the nuggets on the steel side are indicated by blue lines, the area between the white and blue lines is the heat-affected zone (HAZ) on the steel side, and the area indicated by white lines on the Mg side is the nugget. The HAZ on the Mg side is not visible in the cross-sections because no phase transformation occurs. Also, the grain size does not grow very large in the HAZ compared with the AZ31 base metal, though it does exist adjacent to the nugget and can be observed at a higher magnification. Clearly, two separated nuggets form respectively on the Mg side and the steel side for both types of joints. These ‘two nuggets’ have also been observed when welding Al to steel (Wang et al., 2015). During welding, the heat generation is higher on the steel side because the electrical conductivity of steel is lower than Al or Mg and the bulk resistance of DP600 steel is relatively high. The Mg has a lower resistivity and a higher heat conductivity than steel, and so the Mg alloy operates as a cooling electrode for the steel (Wang et al., 2007). Therefore, the nugget forms at the center of the steel sheet and cannot grow to the Mg/steel interface, leading to the formation of two separated nuggets. The HAZ on the steel side is large because of the additional heat generation of the interlayer, and the nugget size on the steel side is 6.24 mm in diameter. It should be noted that although there is no void in Fig. 2(a), overall a relatively small number of voids are still present in the joints with an interlayer, such as that seen later in Fig. 4(d) and (e). Generally in RSW joints, a dendrite structure forms at the edge of the nugget and an equiaxed structure forms in the center of the nugget. However, in the joint with an interlayer the equiaxed structure is observed at both the edge of nugget (Fig. 2(b) and (d)) and the center of nugget (Fig. 2(e)) on the Mg side. This may be because of the increased heat generation in the joint with an interlayer, which leads to a lower temperature

gradient

and

a

higher

cooling

rate

that

promotes

the

columnar-to-equiaxed transition at the edge of nugget (Xu et al., 2012). As shown in 5

Fig. 2(c), in the HAZ the grain size is coarsely affected by the heat during welding, but the equiaxed structure remains. The grain size in HAZ is 6.62 µm, and the grain size of the base metal is 5.65 µm. As shown in Fig. 2(f), near the Mg/steel interface the dendrite structure is observed and the grain growth direction is perpendicular to the interface. The cooling rate is lower near the interface compared with the center zone of the nugget owing to the presence of the interface and the lower thermal conductivity of the steel, and the dendrite structure thereby generates easily according to supercooling theory. Additionally, overall the nugget possesses a small grain size, with a grain size of only 5.43 µm in the center equiaxed structure area. On the steel side, it is obvious that the HDG Q235 interlayer and the DP600 base metal are melted during welding and a nugget is formed, as shown in Fig. 2(g). Also, the large lath martensite is observed in the nugget for both types of joints, as shown in Fig. 2(j). The formation of martensite is owing to the high cooling rate and the high carbon equivalent of DP600 steel (Wan et al., 2014). Ignasiak et al. (2012) have found that a similar microstructure is produced in the DP600 RSW process. The HAZ can be divided into a two-phase zone, a fine-grained zone and a coarse-grained zone for both types of joints, indicated in Fig. 2(g) by C, B and A, respectively. Martensite and ferrite are distributed uniformly in the two-phase zone (Fig. 2(h)); while fine martensite, ferrite and austenite are distributed unevenly in the fine-grained zone (Fig. 2(i)). The peak temperature in the coarse-grained HAZ is greater than the AC3 (fully austenitic temperature) and the austenite grains are coarser than those in the fine-grained HAZ. Similar microstructures in the HAZ have also been observed in the DP600 RSW process by Zhao et al. (2013). A mixed microstructure of pearlite and ferrite is mainly observed in the HAZ on the interlayer, but the grain size is coarsened compared with the base metal (12.67 µm for HAZ and 7.75 µm for base metal), as shown in Fig. 2(k). The microstructure of the base metal DP600 consists of fine ferrite and martensite, as shown in Fig. 2(l).

Figure 3 shows the typical macrostructure and microstructure of the joint without an interlayer. The ‘two nuggets’ are also clearly observed in Fig. 3(a), where the nugget size on the steel side is 5.65 mm in diameter. Several voids are observed at the Mg/steel interface as defects, which indicate that joining in these areas is not achieved. 6

The voids in the joint with an interlayer are inhibited to some extent compared with the joints without an interlayer, but are still not avoided. A dendrite structure is observed at the edge of the nugget, as shown in Fig. 3(b), and the equiaxed structure is observed in the center of the nugget, as shown in Fig. 3(e). During solidification, the weld microstructure is divided into five types of crystal morphology according to the temperature gradient and constituent supercooling. These crystal morphologies are planar grain, cellular grain, cellular dendrite grain, dendrite grain and equiaxed grain, as demonstrated by Kou (2003). After welding, the temperature gradient is largest at the edge of the nugget, and decreases continuously from the edge to the center of the nugget. Therefore, the dendrite and equiaxed structures form at the edge and in the center of the nugget, respectively. As shown in Fig. 3(c), in the HAZ the equiaxed grain is also coarsened, with a grain size of 7.54 µm. The dendrite structure is also observed near the interface, as shown in Fig. 3(f). Overall, the grain size in the nugget is relatively large. For instance, in the center equiaxed structure area, the grain size is 6.31 µm. On the steel side, the microstructure is similar to that in the joint with an interlayer. The large lath martensite is also observed in the nugget, as shown in Fig. 3(i). The two-phase zone, fine-grained zone and coarse-grained zone also exist in the HAZ, as shown in Fig. 3(h) and (i).

Weld formation process Figure 4 shows the cross-sections of the joint with an interlayer at different welding times. When the welding time is 35 ms, a hat-shaped nugget forms on the Mg side (Fig. 4(a)), while no melting can be observed on the steel side. As the welding time increases, the nugget on the Mg side grows in the diameter direction and reaches a relatively steady value (about 7.8 mm) at 100 ms. On the steel side, the fusion zone begins to appear at 50 ms (Fig. 4(b)) and a small nugget, which is only located on DP600, forms when the welding time reaches 75 ms (Fig. 4(c)). With a further increase in the welding time, the nugget grows in both diameter and thickness, growing into the interlayer at 100 ms. Finally, a non-melted zone (about 0.2 mm thick) still exists on the interlayer near the interface after 150 ms, as discussed above. The peak temperature during welding in the non-melted zone is below the melting point of Q235 steel, and consequently the Q235 steel is not melted during welding in this zone. 7

This may be because of the low melting point and higher thermal conductivity for AZ31. The heat loss in this area during welding is faster because it is adjacent to the magnesium alloy (Xu et al., 2012).

Hardness distribution

As shown in Fig. 5, the hardness distribution on the Mg side in the joint with an interlayer is similar to that in the traditional RSW joint. The average hardness in the nugget is 55.9 HV for the traditional RSW joint, while it is 59.6 HV for the joint with interlayer. This increased hardness is owing to the fine grain size in the nugget of the joint with an interlayer, as mentioned above. As can be seen, the hardness in the nugget on the steel side is roughly above 350.0 HV for both joints, owing to the formation of large lath martensite. The average hardness in the nugget is 355.7 HV for the joint without an interlayer, while it is 379.3 HV for the joint with an interlayer. This increase in hardness in the nugget is also owing to the fine grain size on the steel side in the joint with an interlayer (Wan et al., 2014). For instance, the average lath width of martensite in the nugget of the joint with an interlayer is 50.0 µm, while that in the joint without an interlayer is 61.5 µm. In the HAZ on the steel side, obvious softening is observed because of martensite tempering, and Zhang et al. (2014) also have found a similar situation in a dissimilar DP780/DP600 RSW joint. The hardness of the interlayer increases from the Mg/steel interface to the steel side, and is lower than the nugget because of the absence of lath martensite.

Tensile shear test As shown in Fig. 6(a), the average peak load of the traditional RSW joint increases from 2489.8 N at 10 kA to 4143.5 N at 16 kA. The peak load continuously decreases to 3456.6 N at 20 kA, where the maximum peak load occurs at 16 kA. For joints with an interlayer, however, the average peak load increases from 2052.3 N at 10 kA to 5493.0 N at 18 kA, and then decreases to 3880.5 N at 20 kA. The maximum peak load (5493.0 N) occurs at 18 kA, and is 32.6% higher than that of the traditional RSW joint. The scatter at the optimum welding condition of the joint with an interlayer is comparable to that of the joint without an interlayer because voids still exist in the joint with an interlayer, although some areas of the joint have no voids. 8

As shown in Fig. 6(b), the energy absorption, representing the ductility of a joint, increases from 0.94 J at 10 kA to 2.38 J at 16 kA, and continuously decreases to 2.14 J at 20 kA for traditional RSW joints. For joints with an interlayer, the average energy absorption increases from 0.86 J at 10 kA to 5.16 J at 18 kA, and then decreases to 3.38 J at 20 kA. Overall, the energy absorption for the joint with an interlayer is much higher than that for the RSW joint. The maximum energy absorption of the joint with an interlayer is 5.16 J, and is 2.17 times that of the traditional RSW joint (2.38 J). The weld fusion (weld nugget) size is the most important factor affecting the mechanical strength of a spot weld (Zhou et al., 2003). As shown in Fig. 7, the trend of the button size as a function of the welding current is similar for both the joints with and without an interlayer. The button size is measured after failure of the welded joint and is different than the nugget size. In the photographs shown in Fig. 7, the button is outlined by a red circle, and the average diameter value measured in the two perpendicular directions gives the button size. The nugget size is measured from the micrograph of the etched cross-section of the weld. When the welding current is low, the heat production is not sufficient for nugget growth and the button is small. Conversely, when the welding current is too high, spatter is observed during welding that leads to a decrease in the size of the button. The maximum button size (9.67 mm) for the joint with an interlayer is 10.4% greater than that of the traditional RSW joint (8.76 mm), and the button size of the joint with an interlayer is greater than that of the traditional RSW joint overall. This is one of the reasons why the joint with an interlayer has a higher peak load.

Typical tensile curves for joints with and without an interlayer are shown in Fig. 8. Obviously, the peak load and elongation of the joint with an interlayer are higher than that of the RSW joint, where the elongation of the joint with an interlayer is about 33% greater than that of the RSW joint. In addition, an interfacial failure mode is observed in the tensile fractured specimen whether or not there was an interlayer, as shown in Fig. 8.

Figure 9 shows the fracture morphology of the joint with an interlayer on the steel side and the Mg side. Three different regions (soldered region, solid-state 9

bonding region and braze welding region) are observed both on the steel side and the Mg side, and are identified as regions A, B and C, respectively. The braze welding region is located at the Mg/steel interface in the nugget, the solid-state bonding region is observed near the fusion line out of the nugget, and the soldered region is adjacent to the solid-state bonding region. A similar phenomenon has also been reported by Liu et al. (2010) while investigating the RSW of magnesium to HDG steel. During welding, the hot-dip zinc coating of the interlayer is melted and squeezed out of the welding zone under the electrode force, and is then assembled out of the nugget and cooled to form a peripheral soldered region after welding. In the soldered region (region A) on the steel side (Fig. 9(b)), 84.1% Zn, 7.06% Mg and 8.85% Fe are observed. On the Mg side (Fig. 9(f)), 39.28% Zn, 58.39% Mg and 2.33% Al are observed. The existence of the soldered region is proved using EDX, where an EDX line scan is taken on the soldered region of the cross-sectioned specimen (Fig. 10). From the EDX result, it is obvious that a reaction layer is formed in this area. The lamellar eutectoid structure consists of 43.99 Wt% Mg and 56.01 Wt% Zn, indicating a MgZn+Zn phase according to the Mg-Zn binary diagram. A similar reaction phase has also been found by Li et al. (2012) during laser welding-brazing of Mg to steel. The soldered region acts as a mechanical seal, which helps suppress the formation of metallic oxides in the solid-state bonding region during welding, and consequently improves the bonding effect. A shear dimple structure is observed in the solid-state bonding region (region B), as shown in Fig. 9(c) and (g). In region C on the steel side, two different types of areas are observed. In most areas in this region, such as Position 4, Fe, Mn, Si and some Mg and Al elements are observed, indicating that in this type of area some Mg and Al elements are diffused into the steel matrix during welding. In the other type of area in this region, such as at Position 3, Mg and a small amount of Al and Zn elements are observed, indicating that this type of area is residual AZ31 on the steel side. According to these EDX results, it can be inferred that the AZ31 is melted and braze-welded to the steel during welding in region C. On the Mg side, no Fe, Mn or Si elements are observed on the fracture surface, indicating that the steel near the interface is not melted during welding. In addition, a small number of cracks and some pores are observed on the Mg side (Fig. 9(e)). A shear dimple structure is even observed in the nugget on the Mg side (Fig. 9(h)), although the dimples are very small and shallow, indicating a relatively good joining in the nugget. 10

Figure 11 shows the fracture morphology of the traditional RSW joint on the steel and Mg sides. Only two different regions (solid-state bonding region and braze welding region) are observed, and are identified in Fig. 11(a) and (e) as regions B and C, respectively. The electro-galvanized zinc coating is finer, thinner and denser and the soldered region, produced by a relatively large amount of squeezed and assembled zinc, cannot form in the joint without an interlayer. In the solid-state bonding region (region B) on the steel side (Fig. 11(b)), the EDX analysis indicates that this region mainly consists of Mg alloy. Though 33.67% Fe is observed in the solid-state bonding region on the steel side of the joint with an interlayer, there is no Fe element observed in that of the traditional RSW joint, indicating poor diffusion welding in this region owing to the absence of a mechanical seal made by the soldered region. Shrinkage cavities are observed, also indicating a relatively poor joining. The AZ31 is also melted and braze-welded to the steel in region C according to the EDX results. Residual AZ31 on the fracture surface on the steel side is also observed and the steel near the interface remains unmelted during welding, also owing to the absence of Fe, Mn or Si elements in region C on the Mg side. Some pores and many cracks are observed close to the nugget center on the Mg side and several shrinkage cavities (Fig. 11(h)) are observed in this region, indicating the relatively poor joining property.

In summary, the hot-dip zinc coating of the interlayer was melted during welding and squeezed out of the welding zone under the electrode force and then assembled out of the nugget to form a soldered region. The mechanical seal made by this soldered region improved the solid-state bonding and braze-welding in the weld joint. The hot-dipped zinc proved to play a significant role when joining the Mg to steel again. Unfortunately, the zinc coating that existed on the electro-galvanized steel was finer, denser, thinner and more uniform than that provided by the interlayer, and therefore the soldered region and mechanical seal produced by a relatively large amount of squeezed and assembled zinc cannot form in the joint without an interlayer. The absence of a soldered region and the deficient solid-state bonding and braze-welding effect therefore led to lower tensile properties with direct RSW of Mg to electro-galvanized steel. In addition, a hot-dip of the galvanized steel was selected to produce the interlayer instead of using zinc foil because a joining between the 11

interlayer and the DP600 base metal was more assured with this interlayer technique and its cost was lower compared with pure zinc foil, which are both beneficial attributes for industrial applications.

Conclusions A 2-mm-thick AZ31 Mg alloy and a 1.2-mm-thick electro-galvanized DP600 steel plate were welded using a RSW process with a 0.6-mm-thick layer of HDG Q235 as an interlayer. The weld formation, microstructure, and tensile properties of the joints were analyzed and compared with traditional RSW joints. The following conclusions can be drawn: 1. With the help of the HDG Q235 interlayer, a high quality AZ31/DP600 joint could be obtained using RSW. The maximum peak load increased by 32.6% and the energy absorption increased by 117.0% compared with a traditional RSW joint. 2. The electro-galvanized zinc coating on the DP600 steel could not be squeezed out and the soldered region could not form during welding. When a HDG Q235 interlayer was added, however, the hot-dipped zinc coating was squeezed out of the nugget and a peripheral soldered region formed during welding. A vacuum chamber made by the soldered region improved the solid-state bonding and braze-welding in the joint. Therefore, the tensile properties improved with the help of the interlayer. 3. Two separated nuggets formed respectively on the Mg side and the steel side whether or not there was an interlayer. When the interlayer was added, however, the interlayer and the DP600 base metal were melted during welding and a nugget formed on the steel side, indicating perfect joining between the interlayer and steel base metal. In addition, the HAZ and the nugget were larger on the steel side compared with that of the traditional RSW joint because of increased heat generation. 4. A hat-shaped nugget formed on the Mg side, but there was no melt observed on the steel side at the beginning of the welding process. With an increased welding time, the nugget on the Mg side grew and reached a relatively stable size at 100 ms. On the steel side, the nugget formed in the DP600 steel initially, and then grew into the interlayer at 100 ms. With a further increase of the welding time, the nugget on the steel side grew continuously until the welding was complete. Acknowledgment This research is supported by the National Natural Science Foundation of China (Grant Nos. 51275342 and 51405334).

12

References Hamu, G.B., Eliezer, D., Wagner, L., 2009. The relation between severe plastic deformation microstructure and corrosion behavior of AZ31 magnesium alloy. J. Alloys. Compd. 468, 222-229. Ignasiak, A., Korzeniowski, M., Ambroziak, A., 2012. Investigations of microstructure of resistance spot-welded joints made of HSLA340 and DP600 steels. Arch. Metall. Mater. 57(4), 1081-1086. Kou, S., ‘Welding Metallurgy, Seconded’, 2003, John Wiley&Sons, Inc., New Jersey. Li, L.Q., Tan, C.W., Chen, Y.B., Guo, W., Hu, X.B., 2012. Influence of Zn Coating on

Interfacial

Reactions

and

Mechanical

Properties

During

Laser

Welding-Brazing of Mg to Steel. Metall. Mater. Trans. 12(43A), 4740-4754. Liu, L., Xiao, L., Feng, J., Li, L., Esmaeili, S., Zhou, Y., 2011. Bonding of immiscible Mg and Fe via a nanoscale Fe2Al5 transition layer. Scr. Mater. 65, 982-985. Liu, L., Xiao, L., Chen, D.L., Feng, J.C., Kim, S., Zhou, Y., 2013. Microstructure and fatigue properties of Mg-to-steel dissimilar resistance spot welds. Mater. Des. 45, 336-342. Liu, L., Xiao, L., Feng, J.C., Tian, Y.H. Zhou, S.Q., Zhou, Y., 2010. The Mechanisms of Resistance Spot Welding of Magnesium to Steel. Metall. Mater. Trans. 10, (41A), 2651-2661. Method of tensile test for welded joint. General administration of quality supervision, inspection and quarantine of the People’s Republic of China, GB 2651-88, A China National Standard. Beijing: China Standard Publishing House; 1990. Wan, X.D., Wang, Y.X., Zhang, P., 2014. Modelling the effect of welding current on resistance spot welding of DP600 steel. J. Mater. Process. Technol. 214, 2723-2729. Wang, J., Wang, H.P., Lu, F.G., Carlson, B.E., Sigler, D.R., 2015. Analysis of Al-steel resistance spot welding process by developing a fully coupled multi-physics simulation model. Int. J. Heat. Mass. Transfer. 89, 1061-1072. Wang, Y.R., Mo, Z.H., Feng, J.C., Zhang, Z.D., 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, 641-646. Xu, W., Chen, D.L., Liu, L., Mori, H., Zhou, Y., 2012. Microstructure and mechanical 13

properties of weld-bonded and resistance spot welded magnesium-to-steel dissimilar joints. Mater. Sci. Eng. A537, 11-24. Zhang, H.J., Wang, F.J., Xi, T., Zhao, J., Wang, L.J., Gao, W.G., 2015. A novel quality evaluation method for resistance spot welding based on the electrode displacement signal and the Chernoff faces technique. Mech. Syst. Signal. Pr. 42-43, 431-443. Zhang, H.Q., Wei, A.J., Qiu, X.M., Chen, J.H., 2014. Microstructure and mechanical properties of resistance spot welded dissimilar thickness DP780/DP600 dual-phase steel joints. Mater. Des. 54, 443-449. Zhao, D.W., Wang, Y.X., Zhang, L., Zhang, P., 2013. Effects of electrode force on microstructure and mechanical behavior of the resistance spot welded DP600 joint. Mater. Des. 50, 72-77. Zhou, M., Zhang, H., Hu, S.J., 2003. Relationships between quality and attributes of spot welds. Weld. J. 82 (4), 72-77.

14

Figure Captions Fig. 1 Schematic of the welding process and the specimen dimensions (mm) Fig. 2 Macrostructure and microstructure of joint with interlayer: (a–f) Mg side; (g–l) steel side Fig. 3 Macrostructure and microstructure of the joint without an interlayer: (a–f) Mg side; (g–j) steel side Fig. 4 Growth of the weld nugget for the joint with an interlayer at 18 kA for different welding times Fig. 5 Hardness distribution in the weld cross-section for joints with and without an interlayer Fig. 6 Impact of the welding current upon the (a) peak load and (b) energy absorption of joints with and without an interlayer Fig. 7 Impact of the welding current upon the button size of the joints with and without an interlayer Fig. 8 Typical tensile curves for the joints with and without an interlayer Fig. 9 Fracture morphology of the joint with an interlayer: (a–d) steel side; (e–h) Mg side Fig. 10 EDX results from the reaction layer in the joint with an interlayer Fig. 11 Fracture morphology of the traditional RSW joint: (a–d) steel side; (e–h) Mg side

15

Fig. 1 Schematic of the welding process and the specimen dimensions (mm)

16

Fig. 2 Macrostructure and microstructure of joint with interlayer: (a–f) Mg side; (g–l) steel side 17

Fig. 3 Macrostructure and microstructure of the joint without an interlayer: (a–f) Mg side; (g–j) steel side

18

Fig. 4 Growth of the weld nugget for the joint with an interlayer at 18 kA for different welding times

19

Fig. 5 Hardness distribution in the weld cross-section for joints with and without an interlayer

20

Fig. 6 Impact of the welding current upon the (a) peak load and (b) energy absorption of joints with and without an interlayer

21

Fig. 7 Impact of the welding current upon the button size of the joints with and without an interlayer

22

Fig. 8 Typical tensile curves for the joints with and without an interlayer

23

Fig. 9 Fracture morphology of the joint with an interlayer: (a–d) steel side; (e–h) Mg side

24

Fig. 10 EDX results from the reaction layer in the joint with an interlayer

25

Fig. 11 Fracture morphology of the traditional RSW joint: (a–d) steel side; (e–h) Mg side

26

Table 1 Composition of base metals and interlayer (Wt%) Si AZ31B 0.1

Mg

Mn

C

Balanced 0.2–0.5 ——

Zn

Fe

0.5–1.5 0.005

Al

P

S

2.5–3.5 0.015 0.005

DP600

1.00 2.31

1.52

0.079 ——

Balanced 0.023

——

——

Q235

0.01 ——

0.39

0.01

——

Balanced ——

0.03

0.025

27

Table 2 Elemental composition of different positions in Fig. 9 (Wt%) Positio

Fe

Mg

Al

Zn

Mn

Si

1

8.85

7.06

——

84.1

——

——

2

33.67

62.8

3.53

——

——

——

3

——

96.4

2.5

1.1

——

——

4

35.04

53.36

7.56

——

3.2

0.83

5

——

58.39

2.33

39.28

——

——

6

——

98.0

1.5

0.5

——

——

7

——

96.2

2.6

1.2

——

——

n

28

Table 3 Elemental composition of different positions in Fig. 11 (Wt%) Positio

Fe

Mg

Al

Zn

Mn

Si

1

——

96.3

2.6

1.1

——

——

2

——

95.9

2.9

1.2

——

——

3

——

98.0

2.0

——

——

——

4

80.8

6.79

8.22

——

3.46

0.73

5

——

99.0

0.8

0.2

——

——

6

——

97.6

1.7

0.7

——

——

7

——

96.6

2.5

0.9

——

——

n

29