**2. Materials and Methods**

The base metals that were used in this study were three-mm thick commercial pure copper (99.9% purity, annealed, JIS C1100), brass (JIS C2600), AA1050H, and AA6061-T6 aluminum rolled sheets. The chemical compositions of raw materials that are given in Table 1 were determined by inductively coupled plasma-atomic emission spectrometry (ICP-AES). The base metals were machined into rectangular specimens with dimensions of 80 (*l*) × 30 (*w*) × 3 (*t*) mm for the FSW process. The surfaces of FSW-joined specimens were ground with 2000-grit SiC papers, and then ultrasonically cleaned with acetone prior to welding. The friction stir welded AA1050H/C1100, AA6061-T6/C1100, and AA1050H/C2600 dissimilar joints were carried out in this study, and these joints were labeled as A1/C1, A6/C1, and A1/C2 specimens in the following, respectively.

**Table 1.** Chemical compositions of the used Al and Cu base metals (in wt.%).


Figure 1a schematically illustrates the FSW process of Al–Cu joints. The welding tool that was used in this study was made of AISI H-13 tool steel with a shoulder that was 15 mm in diameter and a stirring probe with a four-mm diameter and two-mm depth, as shown in Figure 1b. Pure Cu and brass plates were always positioned at the advancing side (AS). The tool was in the center of the Al–Cu joints, and it was not shifted to the Al or Cu side. According to our preliminary trial experiments of several Al–Cu joints and the macrograph/micrograph inspections for the welds of various parameters, optimum welding parameters were obtained for the welds with an optimal tensile strength and defect-free microstructures in the welding zone at the stable part of the seam for the present study. Based on our preliminary trials and the literature review [22], a low transverse speed and high tool rotational speed are usually required for obtaining defect-free joints. The tool rotational speed was set at 3000 rpm. The rotating tool was tilted 1.5◦ opposite to the welding direction, and the stirring probe moved along the butt line of the Al–Cu joint specimens at a constant traverse speed of about 60 mm min−1. The downward push pressure was controlled at about 45 MPa. During the FSW process, the downward push pressure was maintained for an appropriate time to generate sufficient frictional heat. The frictional heat softened the Al–Cu joint specimens, and the stirring probe caused material plastic flow in both circumferential and axial directions. The welded direction (WD), normal direction (ND), and transverse direction (TD) of the FSW Al–Cu dissimilar joints were also defined in Figure 1a. After the friction stir welding process, the upper and lower surfaces of welded specimens were carefully ground with 2000-grit SiC abrasive paper to eliminate the defects and stress concentrators located on the surface of the seam.

**Figure 1.** Schematic illustrations of the (**a**) friction stir welding (FSW) process of Al–Cu dissimilar joints, (**b**) the welding tool geometries used for FSW, and (**c**) the dimensions of the tensile specimen prepared from welded samples.

The FSW joined A1/C1, A6/C1, and A1/C2 dissimilar joints were cross-sectioned along the TD. For the study of the welding microstructures, the cross-sections (on the WD plane) taken at the stable part of the seam were ground and polished with a diamond polishing agent. The specimens were etched in a Keller's solution for the Al alloy side, and the Cu alloy side was etched with a solution of five grams of FeCl3, 50 mL of HCl, and 100 mL of H2O. The phase composition within the welding zone (WZ) of the Al–Cu dissimilar joints were identified by an X-ray diffractometer (XRD, Bruker D8A25, Bruker Corp., Karlsruhe, Germany), using CuKα radiation at 40 kV and 40 mA with a scan speed of 3◦ (2θ) min−<sup>1</sup> (step size, 0.02◦). Microstructures of the A1/C1, A6/C1, and A1/C2 dissimilar joints were observed by the backscattering electron image (BEI) taken with a scanning electron microscopy (SEM, JEOL/JSM-6360, JEOL Ltd., Tokyo, Japan). SEM equipped with an energy-dispersive X-ray

spectroscopy (EDS) and electron probe micro-analyzer (EPMA, JEOL JXA-8530F, JEOL Ltd., Tokyo, Japan) were used to identify the elemental compositions and distribution of compounds formed within the WZ of FSW Al–Cu dissimilar joints for the investigation of metallurgical interactions.

The Micro-Vickers hardness test (HV) was applied to evaluate the variations of microhardness after the FSW process. The Micro-Vickers hardness test across the cross-sections of the A1/C1, A6/C1, and A1/C2 joints was applied using a Vickers indenter (Future-Tech Corp., Kawasaki, Japan) with a 50-g load for 10 s of dwell time. Each measured microhardness datum was the average of at least three tests.The tensile strength of the FSW A1/C1, A6/C1, and A1/C2 dissimilar joints was measured according to the standard tension testing of ASTM E8M-11. Uniaxial tensile tests of the FSW Al–Cu joints were carried out in the directions perpendicular to the WD. Figure 1c shows the dimensions of the tensile specimens prepared from welded samples. The specimens were tested at room temperature with an initial strain rate of 8 × <sup>10</sup>−<sup>4</sup> <sup>s</sup>−<sup>1</sup> per mm. The tensile strength measurements of each condition (A1/C1, A6/C1, and A1/C2 joints) were performed on 20 test specimens (*n* = 20) for the statistical significance of following Weibull statistical analysis. The fracture surfaces and sub-surfaces of the FSW Al–Cu joints were further examined using SEM/BEI with EDS mapping to analyze the fracture morphologies and behaviors.
