4.1 Tensile Properties
The average tensile value of the samples taken as sample 1 horizontally is 255.3 MPa and the average tensile value in the vertical is 166 MPa. While the average elongation value of the samples in the horizontal direction is %40, the elongation value of the samples in the vertical direction is %16.3. As a result of the tensile test, the horizontal tensile value of 258 MPa and the vertical tensile value of 160 MPa were obtained for sample 2. The average elongation of the horizontal samples is %37 and the vertical is %18. It is close to the mechanical properties of ER5356 welding wire [24]. ER5356 has yield strength is 110-120 MPa, tensile strength is 240-296 MPa and elongation is %17-26. The results obtained are close to other studies. Köhler et al [25] produced 5356 and 4047 aluminum alloys using pulsed cold metal transfer (CMT-P). The walls produced from these aluminum alloys were compared with each other. In both parts, they observed that the tensile strength in the horizontal direction was higher than the vertical. The difference between the mechanical properties of the horizontal and vertical tensile specimens are shown in Figure 4.
Fig.4 Horizontal and vertical tensile values of sample 1 and sample 2.
When the tensile, yield and elongation values of the two samples were examined, it was observed that there was no significant difference between them. It can be said that welding performed only in the horizontal direction at sample 2 provides higher strength although there is no significant differences. Horizontal tensile values for both samples were found close to other studies and the tensile value of ER5356. The fact that the tensile and yield strengths in the horizontal direction are significantly higher than those in the vertical is also a situation encountered in other studies [26]. For instance Suryakumar et al [27] produced parts using the waam method with copper-clad steel filler wire and examined their mechanical properties. Similar to Al parts, they have reported that higher tensile strength is obtained in the horizontal direction than in the vertical direction in steel materials too. Fang et al [28] produced 2219 aluminum alloy in different CMT modes. In their study, they stated that higher mechanical properties can be obtained according to the changes in the mode of the CMT method. They examined the mechanical and microstructural properties of the parts according to the mode change. As a result of their studies, they observed that the horizontal tensile strengths were higher than the vertical in almost all modes. They stated that the difference in tensile strength between horizontal and vertical samples could decrease, although the anisotropic property continued with the change of parameters.
When the given examples were evaluated the same result was observed in different materials and different welding methods. This shows that the samples perpendicular to the weld seam direction do not have the same homogeneity and contain microstructural differences. As it is known, in the welding process, each pass is applied in successive passes or in multi-pass welding applications [29]. It makes heat treatment on the previous pass and causes it to be softer. As a result, different microstructural transformations occur due to non-homogeneous heat distribution during multi-pass welding and this affects the mechanical properties.
4.2 Microhardness
In order to examine the hardness variation between layers, microhardness measurements were made from the upper and lower layers of the samples. The results of the microhardness values are shown in Figure 5. Sample 1 had an average hardness of 84 HV0.2 in the lower layer, while the hardness value in the upper layer was 101.8 HV0.2. In Sample 2, 105.38 HV0.2 was obtained in the lower layer and 109.28 HV0.2 in the upper layer. In both samples, higher hardness value was observed in the upper layers. This is due to heat accumulation. As the layers are built on top of each other, the surface of the substrate is exposed to high temperature. As the height increases, the heat accumulation increases as the layers overlap. Heat affects hardness negatively. The top layer cools and heats up faster while the bottom layer cools very slowly. Hence higher hardness values are observed in the upper layers due to the cooling rate. Singh et al [30] investigated hardness of samples in their review article on WAAM process strategies and challenges. They reported that as the deposition rate increased, they obtained higher hardness values in the upper regions.
Fig.5 Vickers microhardness of WAAM AA5356 samples
4.3 Microstructure
The biggest difficulties encountered in the production of aluminum alloys with WAAM are the appearance of faulties such as pores, cracks and oxidation. Process parameters such as voltage, current, wire feed speed, interpass temperature, shielding gas flow, and others can alter the effects of oxidation and other problems. The effects of oxidation change the properties of the material, such as mechanical properties, which affect part quality. Although these effects are tried to be minimized by optimizing the parameters, many studies have stated that this problem is inevitable for the production of aluminum alloys by the WAAM process. For example, Hauser et al [31] investigated the oxidation problem in the production of AW4043/AlSi5(wt%) by the waam process. They supported experimental observations with CFD (Computational Fluid Dynamics) simulation. They examined the effects of parameters on oxidation formation and reported that the thickness of the layer varies depending on the change of the CMT mode. By optimizing the parameters, the effect of oxidation can be reduced but oxidation is inevitable in the production of aluminum by this method. Also Ding et al [32] investigated the optimization of the oxidation problem in waam and the effects of surface oxidation. They revealed that oxidation does not only affect the mechanical and microstructural properties. Surface oxide can cause problems in the progress of the process affect the deposition rate and reduce the deposition efficiency. Hence oxidation is one of the main problems of this process and oxidation optimization studies are envisaged for the widespread use of WAAM technology in many areas of use.
When heated aluminum comes into contact with the ambient air and the oxygen it contains, a strong aluminum oxide layer (Al2O3) is formed as an exothermic reaction. Al2O3 is a very durable oxide that gives aluminum its excellent corrosion resistance. While oxides of most other metals melt at or below the temperature of their metals, Al2O3 has a very high melting point of 2060 °C, compared to the pure metal melting at 660 °C [31]. Hence this situation is frequently encountered in the production of aluminum alloys with WAAM. The oxide layer formed on the parts are given in Figure 6. The microstructures of sample 1 and 2 are shown in Figure 6a) and Figure 6b), respectively. In the microstructure images, the oxide layer on the aluminum parts can be clearly seen. Even if the process is protected by the inert gas, a strong surface oxide is formed when the aluminum which is heated due to the rising heat during welding, comes into contact with the oxygen in the environment. Especially Al-Mg alloys are easily oxidized when heated the high corrosion resistance of 5xxx series alloys is associated with this situation. Easily oxidized 5356 samples were exposed to constant heat input since they were produced by welding on each other. WAAM process accelerated the oxidation process and a surface oxide layer was formed in a short time. EDS analysis of the samples is given in Figure 6c). In the EDS analysis of certain parts of the part surfaces, Al, O and Mg peaks and their weight ratios are seen. In accordance with the wire content, 5356 Al alloy contains more than 5% Mg. Besides Mg, O peaks were observed due to the Al2O3 layer formed on the surface of the parts [31,33].
Fig.6 Al2O3 layer formed on the surface of the part; a) Sample 1, b) Sample 2, c) EDS analysis.
4.4 Electron Backscatter Diffraction (EBSD) Analysis
In order to examine the effect of the current on the mechanical and microstructure properties of samples, the samples were taken from the top and bottom layers and these samples were analyzed by EBSD analysis. EBSD maps are shown in Figure 7. The coarse grain structure formed in the lower regions of the first layers of Sample 2 is shown in Figure 7a). Upper layers of sample 1 are shown in Figure 7b). Fine grain structures formed in the upper layers of Sample 2 are shown in Figure 7c). Compared to Figure 7b and c, it is seen that the current range has not a significant effect on the grain size. However, grain size differences were observed between the layers as can seen in Figure 7a) and c. Coarse grain structure was observed in the lower layers otherwise fine grained structures were observed in the upper layers [34,35].
Fig.7 Examination of grain structure by EBSD analysis; a) Bottom layers of sample 2, b) Top layers of sample 1, c) Bottom layers of sample 2
EBSD analysis results are in accordance with tensile test results and Hall-Petch rule. Hall-Petch rule was shown in equation 1. Similar results are observed in many studies [36,37]. As the grain size increases, the tensile value decreases. When compared to the upper layers, coarse grain structure and low tensile strength were detected in the lower layers.
4.5 Atomic Force Microscopy (AFM)
The milling operation after the WAAM process only visually helped to obtain a relatively flat surface compared to the initial state of the specimens. When the surface topography was examined, elevations and depressions were found between the source layers as given Figure 8. Surface properties of sample 1 are given Figure 8 a) and sample 2 in Figure 8 b). Since the WAAM technology is based on a multi-pass welding process, the upper part of the previous layer is partially melted during each new layer deposition. Surface roughness occurs in the part that is formed by coming together from many layers. Since WAAM is an AM method, finishing process are often required and surface roughness can be improved by machining methods such as milling. However in useage areas where surface roughness is an important criterion, it is necessary to optimize the parameters of finishing process or use different methods to improve the surface properties [38,39].
Fig. 8 AFM images; a) Sample 1 b)Sample 2.