Gihad Kerrar, Nesar Merah, Abdelrahman Nasr Shuaib,
Fadi AL-Badour and Abdelaziz Bazoune
Abstract In this paper, results of performing successful butt-welding of aluminum
grade Al6061-T6 to commercial pure copper using the relatively new friction stir welding (FSW) process are presented and discussed in conjunction with finite element results. A sound weld joint between these dissimilar materials has been achieved only when the pin tool was offset a certain distance from the center of the weld line and the harder material (copper) was placed at the advancing side of the FSW tool. On the other hand, the tool offset was not required when joining the similar materials of aluminum to aluminum and of copper to copper. A combined use of temperature distribution of the aluminum to copper weld nugget determined by a finite element model, the elemental concentration of copper and aluminum in the weld nugget from EDS analysis, and aluminum-copper phase diagram have been used to identify the phases present in the different weldment zones. The aluminum-copper joint was found to consist of several single and two phase intermetallic compounds such as Al2Cu, Al4Cu9, and (Al + Al2Cu). The results revealed higher hardness values for the weld nugget compared with the two base metals.
Keywords Welding dissimilar metals and alloys Friction stir welding Sound
weld joint Joining aluminum to copper
1 Introduction
It is possible to partially or completely replace copper with aluminum for several
engineering applications because aluminum shares with copper some similar physical properties, e.g. electrical conductivity and because aluminum has lower price and lower density. Therefore, researchers are striving to join these two metals
G. Kerrar N. Merah (&) A.N. Shuaib F. AL-Badour A. Bazoune
Mechanical Engineering Department, King Fahd University of Petroleum and Minerals (KFUPM), Dhahran 31261, Saudi Arabia e-mail: nesar@kfupm.edu.sa
T. Boukharouba et al. (eds.), Applied Mechanics, Behavior of Materials, and Engineering Systems, Lecture Notes in Mechanical Engineering, DOI 10.1007/978-3-319-41468-3_10 130 G. Kerrar et al.
together, while facing many challenges, including differences in chemical compo-
sitions and physical and mechanical properties. Based on its relative advantages against conventional fusion welding processes, researchers investigated the possi- bility of using FSW technique to replace the conventional joining methods of dissimilar metals and alloys, such as aluminum to copper, and achieve better quality weld joints [1–3]. Galvao et al. [4, 5] and Xue et al. [5, 6] analyzed the effects of tool design and working conditions on the material flow during FSW of aluminum grade AA5083-H11 to copper. They reported that the relative location of material with respect to the advancing side of the tool as well as the tool geometry have significant effects on the shape of Al–Cu interaction regions and the intermetallic compounds formation. Shin et al. [7] performed an experimental study for dis- similar welding of BMG alloy to lightweight crystalline alloys, and studied the influences of tool geometry having round and triangular shaped pins. They con- cluded that the triangular pin tool produced more chips, and its cutting effect resulted in lowering the tool axial load and workpiece temperature in the early stage of the process as compared to the round pin case. Sun et al. [8] who analyzed flat FSSW for dissimilar metals, Al 6061-T6 to mild steel succeeded to remove the keyhole produced by FSSW using a dent in the back plate and rotating pinless flat shoulder. However, they relied on physical bonding which resulted in weak joint strength. This paper presents the results of performing successful butt-welding of alu- minum grade Al6061-T6 to commercial pure copper using friction stir welding (FSW) process. A finite element model was developed based on earlier work by Al-Badour et al. [9] to estimate the temperature distributions in the dissimilar materials and help explain the observed nuggets structures.
2 Experimental Procedure
Pure commercial copper and Al6061-T6 plates measuring 100 mm 50 mm
4 mm were butt jointed using FSW process. The FSW tool used in this investigation was manufactured from tool steel (H13 RC 50–55), with 5 mm threaded pin diameter, 3.8 mm length, and had a scrolled shoulder of 11.52 mm diameter. Table 1 illustrates the welding conditions used in the investigation. The rotational speed of the tool was maintained at 900 rpm for all test conditions. All tests were
Table 1 Welding parameters and conditions
Condition Welding speed (mm/min) Tool offset (mm) Advancing side material 1 150 0 Al6061-T6 2 20 0 Cu 3 40 0 Cu 4 40 2 Al6061-T6 5 40 2 Cu Experimental and Numerical Investigations of Friction … 131
performed on a fully instrumented experimental model RM-1 friction stir welder
manufactured by MTI. The weld beads were cross-sectioned, polished, and etched using a solution of 1 g of FeCl3, 10 ml HCl, and 100 ml distilled water to reveal copper side microstructure first and then using a solution of 3 ml HNO3, 6 ml HCl, 6 ml HF and 150 ml of distilled water to reveal the Al6061-T6 side microstructure. The weld joints quality was evaluated by examining the optical microscope images of their cross sections to identify the presence or lack of voids, cracks and dis- continuities. Using 300 g load and 10 s dwell time, Vicker’s hardness readings were taken across the weld joint at two locations, 1.5 mm and 3.00 mm below the top surface of the plate. Energy dispersive spectroscopy (EDS) analysis was per- formed at various locations of the weld joint to identify the elemental distribution of copper and aluminum.
3 Finite Element Model
A modified coupled Eulerian Lagrangian (CEL) finite element model that was developed by Al-Badour et al. [9], was used to simulate the FSW process of joining aluminum grade Al6061-T6 to pure copper. The modified model, which is imple- mented and solved in Abaqus environment using Explicit Solver [10], considers a full geometrical model with threaded feature for the pin tool. Johnson-Cooks- empirical law [11] in Eq. (1), is used to represent the plastic flow stress, r0 , of copper and Aluminum 6061-T6. c ln e_ pl
T Tref m
r0 ¼ A þ Benpl 1 þ 1 ð1Þ e_ 0 Tmelt Tref
where enpl is the effective plastic strain, e_ pl the effective plastic strain rate, e_ 0 nor- malizing strain rate (typically 1.0 s−1), A, B, C are material constants, n parameter takes into account the effect of strain hardening, m models the thermal softening effect and c represents the strain rate sensitivity. Tref is the temperature where A, B and n are evaluated while Tmelt is the material solidus temperature. Johnson- Cook’s-empirical law constants for copper and aluminum are given in Table 2. Coulomb’s friction Law with an average coefficient of friction µ = 0.5 was used to represent the interaction between the Eulerian (aluminum and copper) and Lagrangian (tool) domains [12]. As the Eulerian mesh is rigid, velocity constrains around the Eulerian domain were applied to avoid material escape. The tool
Table 2 Copper Johnson-Cook’s parameters [10]
Material A (MPa) B (MPa) C n m Tm (K) T0 (K) Copper 90 292 0.025 0.31 1.09 1356 298 Al 6061-T6 324 114 0.002 0.42 1.34 855 298 132 G. Kerrar et al.
Fig. 1 Al6061-T6/Cu CEL
model materials locations and N Cu/ mesh generation
Al6061- T6/Rs V
rotational speed and welding speed were 900 rpm and 40 mm/min, respectively. Moreover, the effect of tool offset on the weld joint quality was studied using 2 mm offset value to the retreating side. Figure 1 shows a typical model mesh of the Eulerian and Lagrangian domains.
4 Results and Discussions
4.1 Optimizing Welding Parameters
This section covers the procedure for finding the friction stir welding conditions that produce defect free weldments of aluminum to copper butt joint. Welding tests were first conducted to identify the FSW parameters that produced sound weld joints for each of Aluminum grade Al6061-T6 to aluminum grade Al6061-T6 and welding copper to copper. The goal is to use the results of these tests to find starting process parameters for welding aluminum to copper. Defect free weld joints of Al6061-T6 to Al6061-T6 were obtained when welding was performed at tool rotation speed of 900 rpm, welding speed of 175 mm/min, plunging depth of 3.8 mm, and tool tilt angle of 3°. Likewise, the welding parameters for obtaining sound Cu to Cu weldments were found to be 900 rpm rotational speed, 125 mm/min welding speed, 3.8 mm plunging depth, and 3° tool tilt angle. These optimum process parameters of welding copper to copper and aluminum to alu- minum have been used as initial trial parameters for welding aluminum to copper. The first trial test for welding aluminum to copper used the test condition 1 of Table 1. The 150 mm/min welding speed of this condition is the average value of the welding speeds used in joining copper-to-copper and aluminum-to-aluminum. No welding was achieved during the first test condition and the tool pin failed by fracture after travelling a distance of 50 mm only. The tool failure is attributed to an excessive cross feed (traverse) force of 7 kN that overloaded the tool. This high level of traverse force at the relatively fast welding speed of 150 mm/min was caused by a stronger weld bead that was subjected to faster cooling rate. In order to avoid tool failure and to obtain sound weld quality, several trials were performed under lower welding speed conditions of 80, 40, and 20 mm/min. Furthermore, the effect of having the softer material (aluminum) or the harder material (copper) at the Experimental and Numerical Investigations of Friction … 133
advancing side of the tool and of using tool offset relative to the center of the weld bead was explored at these welding speeds. The conditions which resulted in lower welding forces and acceptable weldment quality are: using lower welding speed, placing the hard material on the advancing side, and using a 2 mm offset of the tool. However, placing copper at the retreating side resulted in higher forces, even with the presence of an offset.
4.2 Quality of the Friction Stir Welded Al6061-T6 to Cu
Joint
The effects of test conditions on the quality of the friction stir welded Al6061-T6 to Cu joints, with respect to the presence or lack of discontinuities, in the cross sections of the joints are revealed in the optical microscope images of the sectioned, polished, and etched weldments shown in Fig. 2. The two dissimilar material joints in Fig. 2a, b were both made using 900 rpm tool rotation, no tool offset, 3.78 mm
(a) (b) Voids
(c) Cracks (d)
Fig. 2 Cross sections of the friction stir welded Al6061-T6 to Cu joint at different welding conditions, a condition 2, b condition 3, c condition 4, d condition 5 134 G. Kerrar et al.
penetration depth, 3° tilt angle, and with copper placed at the advancing side of the pin tool. However, the joint of Fig. 2a was performed at 20 mm/min and that of Fig. 2b at 40 mm/min. The lack of tool offset under these two conditions appears to be the reason for the presence of voids in the Al6061-T6-Cu joint. Evidence of presence of cracks appeared in a magnified image of the weldment of Fig. 2c which was made under condition 3 in Table 1. This weldment was performed under the conditions of Fig. 2b, except for the fact that the weld illus- trated in Fig. 2c is made using 2 mm tool offset and the softer Al60-T6 was placed on the advancing side. This lead to the conclusion that placing the softer material on the advancing side produces cracks. Figure 2d shows aluminum to copper weld- ment with no defects. This weldment was produced at the same welding conditions of weldment in Fig. 2c, except that the harder copper material was placed at the advancing side of the tool where more heat is generated by the tool compared with the retreating side. This provided copper with enough heat to soften and at the same time compensated for the heat loss arising from the higher thermal conductivity of copper. The relatively lower amount of heat generation when Al6061-T6 was placed on the retreating side allowed it to be softened below melting point. These results are supported by those of the finite element analysis. It can now be concluded from the preceding that it is possible to weld copper to Al6061-T6 and obtain sound weldments with relatively low welding force when copper is placed at the advancing side of the tool and by using tool offset of 2 mm towards the retreating side. The results of EDS analysis performed on the interface of the sound weld revealed the presence of Al2Cu intermetallic compound of composition 71.0 wt% of Al and 29.0 wt% of Cu.
4.3 Phases in the Copper-to-Aluminum Weld Nugget
The approach to identifying the phases present in the aluminum-to-copper weld
joint nugget is to simultaneously utilize the following: • Elemental composition and relative concentrations of aluminum and copper at various locations of the nugget using EDS analysis. • The temperature distribution of the weld nugget obtained from the finite element model. • Phases and intermetallic compounds from Cu–Al phase diagram at various compositions and temperatures. For example, to identify the phase or intermetallic compound present at a certain location of the nugget, find the concentration of Cu and its ratio to aluminum using EDS analysis, determine the temperature at the location, and then use the phase diagram of Al–Cu to identify the phase at that location. This process is repeated for all points of interest. Figure 3 shows the temperature distribution within the weld Experimental and Numerical Investigations of Friction … 135
Fig. 3 CEL model temperature distributions, condition 3
Fig. 4 Aluminium to copper
weld joint and EDS analysis locations, condition 3
11 1 10 9 8 7 6 5 4 3 2
joint obtained from the finite element analysis and Fig. 4 shows the sound aluminum to copper weld joint together with some of the locations where EDS analysis was performed. The aluminum copper phase diagram is shown in reference [13]. The concentrations of copper and aluminum where EDS analyses were performed are illustrated in Table 3, together with the corresponding intermetallic phases present in the copper side of the sound aluminum to copper weld nugget; and Table 4 is for the aluminum side. The elements and phases present on the copper side of the nugget include single phases of Al2Cu, AlCu, Al4Cu9, and Solid Solution Al(Cu) in addition to a two phase (Al + Al2Cu). Similar phases were also identified on the aluminum side of the sound nugget as shown in Table 4. The presence of intermetallic 136 G. Kerrar et al.
Table 3 Spectrum analysis Al wt% Cu wt% Spectrum Phase presents
toward the Cu side 69.55 30.45 1 Al2Cu 76.17 23.14 2 Al2Cu 92.87 6.36 3 Solid solution Al (Cu) 86.74 12.46 4 Al + Al2Cu 84.31 14.95 5 Al + Al2Cu 75.23 24.04 6 Al2Cu 67.1 32.05 7 Al2Cu 72.05 27.23 8 Al2Cu 82.33 16.89 9 Al + Al2Cu 52.66 46.75 10 AlCu 29.78 70.22 11 Al4Cu9
Table 4 Spectrum analysis Spectrum Al wt% Cu wt% Phase presents
toward the Al6061-T6 side 1 94.44 4.75 Solid solution Al (Cu) 2 19.05 80.95 Al4Cu9 3 90.89 6.78 Solid solution Al (Cu) 4 93.54 5.65 Solid solution Al (Cu) 5 64.32 34.96 Al2Cu 6 93.72 5.35 Solid solution Al (Cu) 7 71.71 27.62 Al2Cu 8 61.30 37.85 Al2Cu 9 55.38 44.62 AlCu
compounds and grain size refinement in the weld nugget caused the measured hardness values in the stir zone and surrounding areas to reach values of 200– 250 HV. Note that the base metal hardness values vary between 47.8–68 HV for aluminum and 78.4–97 HV for copper.
5 Conclusion
The friction stir welding of aluminum grade Al6061-T6 to pure copper was inves- tigated using experimental and numerical approaches and the conditions that resulted in sound weld quality were identified. The following specific conclusions are drawn from the results of the present combined experimental and numerical study: • A sound weld joint between these dissimilar materials has been achieved at lower welding speeds only when the pin tool was offset by 2 mm from the center of the weld line and the harder material (copper) was placed at the advancing side of the tool. Experimental and Numerical Investigations of Friction … 137
• A combined use of temperature distribution of the aluminum to copper weld
nugget determined by a finite element model, the elemental concentration of copper and aluminum in the weld nugget from EDS analysis, and aluminum-copper phase diagram have been used to identify the phases present in the weldment. • The aluminum-copper joint was found to consist of several single and two phase intermetallic compounds such as Al2Cu, Al4Cu9, Al + Al2Cu. • The presence of intermetallic compounds and grain size refinement in the weld nugget caused hardness values in the stir zone and surrounding areas to reach more than twice the hardness values of the base metals.
Acknowledgments The authors acknowledge the support provided by King Abdulaziz City for Science and Technology (KACST) through the Science and Technology Unit at King Fahd University of Petroleum and Minerals (KFUPM) for funding this work through project No. NSTP11-ADV2130-04 as part of the National Science, Technology and Innovation Plan.
References
1. Li, X., Zhang, D., Qiu, C., Zhang, W.: Microstructure and mechanical properties of dissimilar pure copper/1350 aluminum alloy butt joints by friction stir welding. Trans. Nonferrous Met. Soc. China 22(6), 1298–1306 (2012) 2. Galvão, I., Verdera, D., Gesto, D., Loureiro, A., Rodrigues, D.M.: Influence of aluminum alloy type on dissimilar friction stir lap welding of aluminium to copper. J. Mater. Process. Technol. 213(11), 1920–1928 (2013) 3. Ouyang, J., Yarrapareddy, E., Kovacevic, R.: Microstructural evolution in the friction stir welded 6061 aluminum alloy (T6-temper condition) to copper. J. Mater. Process. Technol. 172 (1), 110–122 (2006) 4. Galvão, I., Leal, R.M., Loureiro, A., Rodrigues, D.M.: Material flow in heterogeneous friction stir welding of aluminium and copper thin sheets. Sci. Technol. Weld. Join. 15(8), 654–660 (2010) 5. Xue, P., Ni, D.R., Wang, D., Xiao, B.L., Ma, Z.Y.: Effect of friction stir welding parameters on the microstructure and mechanical properties of the dissimilar Al–Cu joints. Mater. Sci. Eng. A 528(13–14), 4683–4689 (2011) 6. Galvão, I., Loureiro, A., Verdera, D., Gesto, D., Rodrigues, D.M.: Influence of tool offsetting on the structure and morphology of dissimilar aluminum to copper friction-stir welds. Metall. Mater. Trans. A 43(13), 5096–5105 (2012) 7. Shin, H.-S.: Tool geometry effect on the characteristics of dissimilar friction stir spot-welded bulk metallic glass to lightweight alloys. J. Alloys Compd. 586, S50–S55 (2014) 8. Sun, Y.F., Fujii, H., Takaki, N., Okitsu, Y.: Microstructure and mechanical properties of dissimilar Al alloy/steel joints prepared by a flat spot friction stir welding technique. Mater. Des. (2012) 9. Al-Badour, F., Merah, N., Shuaib, A., Bazoune, A.: Coupled Eulerian Lagrangian finite element modeling of friction stir welding processes. J. Mater. Process. Technol. 213(8), 1433– 1439 10. Abaqus ‘software’ 6(11–2) (2011) 11. Johnson, G.R., Cook, W.H.: A constitutive model and data for metals subjected to large strains, high strain rates and high temperatures. In: 7th International Symposium on Ballistics, Netherlands, pp. 541–547 (1983) 138 G. Kerrar et al.
12. Al-Badour, F., Merah, N., Shuaib, A., Bazoune, A.: Thermo-mechanical finite element model of friction stir welding of dissimilar alloys. Int. J. Adv. Manuf. Technol. 72(5–8), 607–617 (2014) 13. Kutz, M. (ed.): Mechanical Engineers Handbook, 2nd edn. Wiley (1998)
