3.1 Microstructure
The morphology of the top and bottom surface of the weld in both configurations are observed and shown in Fig. 3. In the weld of Al/Cu, the top surface reveals a clean weld surface without visible spatters on the weld surface. In contrast, a visible heat-affected zone tends to enlarge at the end of the welding line in the weld of Cu/Al. From the bottom view, heat-affected evidence is visible from the laser power of 120 W when Al is on top of Cu. Meanwhile, in the weld of Cu/Al, full penetration can be observed from the laser power of 180 W. Overall, with the growth of the visible defect on the bottom view as the laser power increases, one can be withdrawn that the penetration increases with increasing laser power.
Figure 4 exhibits the observation of cross-section view in both welds of Al/Cu and Cu/Al. In the case of Al/Cu, the observation result demonstrates an increase of penetration depth when laser power increases. Moreover, dense porosities are observed at the laser power of 140 W and 160 W mainly in the upper material (Al). At the laser power of 140 W, porosities are found to lie on the interface line of the welded materials, while pores are formed around visible cracks initiated from the Al material when laser power of 160 W is used. In the case of Cu/Al weld, the cross-section reveals fluctuated weld surface as well as loss of material in the weld zone. As the laser power increases, the loss of material rises. A reliable explanation for this is the high reflectance of Cu [36] together with high energy input (i.e., laser power). Besides, porosities are also detected mainly in the Al side.
The microstructure of the interface in the weld of Al/Cu and Cu/Al is shown in Fig. 5. In both welds, the dendritic structure is observed to form and expand towards the Al side. The chemical composition of the dendritic structure has been examined and shown in Table 4. The result reveals that the dendritic structure reveals the structure for the IMCs of Al2Cu. Meanwhile, a thicker layer of dendritic structure can be detected in the case of Al on top in comparison with the weld of Cu/Al. To understand the behaviour of microstructures in the weld zone Kirkendall effect has been considered [37]. A reliable explanation for the expansion of dendritic structure into the Al side relates to the phenomenon of the Kirkendall effect which is demonstrated by Smigelskas and Kirkendall [37]. The Kirkendall effect expresses the motion of the interface between two metals due to the difference in the diffusion rate of the metal atoms. The illustration for the effect is indicated in Fig. 6. During the welding process, the atoms of Al and Cu start to diffuse to each other. The diffusion rate of Cu is generally five times higher than that of Al, and the diffusion rate of Cu is linearly proportional to temperature shown in Fig. 6(a) [38]. The interdiffusion of Al and Cu leads to the formation of Al-Cu intermetallic compounds shown in Fig. 6 (b). Together with the low diffusion rate of Al, the spread of IMCs will tend to be on the Al side shown in Fig. 6(c).
Table 4
Chemical composition on the cross-section of the points in
Figure 5.
Point
|
Al (at%)
|
Cu (at%)
|
Possible phase
|
1
|
61.16
|
38.84
|
Al2Cu
|
2
|
83.88
|
16.12
|
Al + IMCs
|
3
|
87.52
|
12.48
|
Al + IMCs
|
4
|
69.86
|
30.14
|
Al2Cu
|
5
|
83.65
|
16.35
|
Al + IMCs
|
6
|
89.55
|
10.45
|
Al + IMCs
|
3.2 Mechanical properties
a. Hardness profile
Vickers hardness measurement is conducted vertically in the cross-section of the weld from the upper material to the lower material. The hardness measurement of the weld of Al/Cu is shown in Fig. 7. Within the upper material (Al) in the weld configuration of Al/Cu, the hardness increases as the distance is close to interface. Meanwhile, the hardness distribution in the weld with the laser power of 80 W remains similar to the base materials (Al: 60 HV, Cu: 117 HV). At the laser power of 100 W, a sharp increase of hardness takes place at the interface. Meanwhile, a significant rise in hardness happens at the distance of 50 \({\mu }\text{m}\) from the interface when the laser powers of 120 W, 140 W, and 160 W are used. In other words, the peak hardness at each laser power shifts towards the Al side when the laser power increases. Furthermore, with the increment of the laser power, the peak hardness of individual laser power rises. In particular, the maximum hardness at the laser power of 100 W and 120 W are 625.9 HV and 880 HV, respectively. Meanwhile, the peak hardness obtained at the laser power of 140 W is 890 HV and at the laser power of 160 W is 995.7 HV. In the lower material (Cu), the hardness decreases significantly at the distance of 50 \({\mu }\text{m}\) and 100 \({\mu }\text{m}\) from the interface. From the distance of 150 \({\mu }\text{m}\) towards the bottom, there is an insignificant variation of the hardness among different laser powers.
Figure 8 indicates the hardness profile of the weld with Cu on top of Al. Overall, the material surrounding the interface is reported with high hardness. Within the upper material (Cu) at all laser powers, hardness increases as the distance is close to the interface. Peak hardness values at the laser power of 180 W and 200 W are measured at the interface of Al and Cu. Meanwhile, the maximum hardness values at the laser power of 220 W and 240 W are detected in the Al side at the distance of 50 \({\mu }\text{m}\) and 100 \({\mu }\text{m}\) from the interface, respectively. On the Al side, it is noticed that the high hardness zone expands together with the increase of the laser power. In particular, at the laser power of 200 W, the material with the hardness exceeding 300 HV remains till the distance of 50 \({\mu }\text{m}\) from the interface. In comparison, at the laser power of 220 W and 240 W, the high hardness zone remains within the distances of 200 \({\mu }\text{m}\) and 250 \({\mu }\text{m}\) from the interface, respectively.
b. Mechanical strength
The mechanical strength of the weld is evaluated by the shear strength test. Figure 9 expresses the ultimate loading forces of each weld according to laser powers. It is clear that the weld of Al/Cu achieves higher weld strength than the weld of Cu/Al. Furthermore, an optimal laser power at which the maximum mechanical strength can be achieved is also recognized in both cases. In particular, in the weld of Al/Cu, the weld attains the average shear loading force of 246.82 N at the laser power of 80 W. The low strength of the weld with the laser power of 80 W can be explained by the lack of penetration in the weld. The ultimate loading force of 297.59 N which is higher than the shear strength of base Al (260 N) and lower than the strength of base Cu (490 N) is recorded at the weld with the laser power of 100 W. As for laser power of 120 W, 140 W, and 160 W, the recorded loading forces declines, respectively. The downturn of the shear strength of the weld when the laser power increase is related to the increase of the IMCs formation in the weld. As shown in Fig. 10, the intermixture of Al and Cu in the interface of the weld Al/Cu increases with the increase of laser power.
When the weld is conducted with the configuration of Cu/Al, the shear strength of the weld is obtained with a similar tendency as in the weld of Al/Cu with the given laser power range. At the laser power of 180 W, the average ultimate loading force recorded is 64.85 N. When the weld is conducted with the laser power of 200 W, the average strength reaches 83.8 N. Afterward, the average of the ultimate loading force reduces as the laser power increases. The lowest strength of the weld is found at the laser power of 260 W with the average loads that the weld can endure is 57.64 N. It is important to note that the explanation for the reduction of the weld strength in Cu/Al welds relies on the loss of the weld metal in the weld zone due to evaporation. As shown in Fig. 4, the loss of the weld metal in the cross-section view increases as the laser power increases. The loss of the weld metal will weaken the connection strength.
Moreover, the fracture behavior of Al/Cu and Cu/Al welds are significantly different. Figure 11 demonstrates the shear strength test result for the weld of Al/Cu and Cu/Al with the presence of elongation. Overall, the weld of Al/Cu results in a longer elongation before it fractures than the weld of Cu/Al. Moreover, the fracture of most of the weld in the configuration of Cu/Al happens at the ultimate load points, while there is elongation after the ultimate load in the case of Al/Cu weld. In short, the weld of Al/Cu is more ductile than the weld of Cu/Al.