*3.2. Macrostructure Examination*

Figure 4 illustrates (a) a macrograph of an example of the produced AMP and (b) the AMPs Diameters/Height (D/H) Ratio as a function of the processing feed speed. The visual inspection of the deposit showed that there is significant flash produced from the deposit, which was restacked to the consumable rod. This may be an indication for an overfed condition, in which the feeding speed for the feedstock consumable material is slightly high [16,29]. The possible decrease of the input material feeding speed would mitigate the generation of this produced excess flash. The generation of excessive flash may require post-processing if the geometric accuracy of the final product is sensitive [4]. The macrostructure cross-sections of the produced AMPs show fully continuous dense structures (Figure 3a) without any physical discontinuities or bonding defects at the layer interfaces, indicating the judicious choice of the processing parameters for the AA2011-T6 and AA2011-O aluminum alloys. It can be seen that the D/H of the produced AMPs increases with increasing feeding speed at constant rotation rate for both the AA2011-T6 and the AA2011-O starting materials, as given in Figure 3b. The AA2011 material plasticity during the FSD process is controlled by the amount of heat input introduced in the vortex zone through the AMPs material building from down to up. This phenomenon appears clearly in the AMP geometry based on the applied processing parameters [24].

**Figure 4.** (**a**) The transverse cross-section macrograph for the AMP with the substrate with the building direction and the interface are indicated for the AMP at a constant rotational spindle rate of 1200 rpm and 6 mm/min feeding speeds, (**b**) AMPs Diameters/Height Ratio against feeding speed for all AMPs produced using the different temper conditions and different processing parameters.

#### *3.3. Microstructure Examination*

FSD as a thermomechanical process is similar to friction stir welding (FSW) [29–32] and processing (FSP) [28] in heat generation, heat dissipation, and heat transfer mechanisms in the stir zone [33–35]. In the AA2011 AMPs, the heat is generated by dynamic contact friction (DCF) between the consumable tool and AA5083 substrate material. Then, it causes severe plastic deformation of the AA2011 material under the applied downward force and transfers it to continuous build by material flow during the stir deposition process. FSW and FSP generate localized grain refinement in the whole nugget zone (NZ) behind the rotating pin tool. The FSD material is analogous to the NZ in FSW and FSP [17,36]. It was found that the presence of a refined, equiaxed grain structure engaged with the formation of high-angle grain boundaries is an indication of the dynamic recrystallization in FSW of AA2219-T8 [9] and FSP of AA2024/Al2O<sup>3</sup> nanocomposite [28].

Moreover, Rutherford et al. [9] reported that the reduction in the average volume fraction and size of the intermetallic particles could also be attributed to the severe plastic deformation of the FSD process. Figure 5 shows the optical micrographs of the initial conditions of AA2011 alloys. The microstructure of the AA2011-T6 alloy shows coarse grains as well as the presence of intermetallics in different shapes: rod-like (R), irregular (I), spherical (S), and almost spherical (A-S), as shown in Figure 5a,c. The microstructure grain size in Figure 5c ranges from 30 ± 3 µm to 150 ± 2 µm with an average grain size of 45 ± 8 µm. [37], whereas the microstructure of the AA2011-O rod alloy shows a relatively smaller grain size (Figure 5b,d) than the AA2011-T6 alloy's grain size. The grain size ranged from 8 ± 2 to 75 ± 3 µm with an average grain size of 16 ± 4 µm. Furthermore, the annealing process causes coarsening of the second phase precipitates compared with the

AA2011-T6 material, transferring their shape from the rod-like shape (Figure 5c) to more spheroidal-shaped precipitates (Figure 5d).

**Figure 5.** Low and high-magnification optical micrographs of the different temper conditions base material: (**a**,**c**) AA2011-T6 and (**b**,**d**) AA2011-O.

#### 3.3.1. AMPs Parts Produced from AA2011-T6 Alloy

The friction-based processes contribute to the increase of temperature of the material in the stirring zone to the temperature range between 60% and 90% of the melting point of the processed material, which is high enough for the recrystallization during the intensive plastic deformation through the solid-state deposition process [38,39]. Figure 6 represents the microstructures of the AA2011-T6 (Figure 6a) and the AMPs deposited at 1200 rpm spindle rotation rate and different feeding speeds of 3 mm/min (Figure 6b), 6 mm/min (Figure 6c), and 9 mm/min (Figure 6d). It can be seen that the coarse grain structure and precipitates of the AA2011-T6 are refined with the applied FSD process parameters at feed speeds of 3, 6, and 9 m/min. The mean measured grain sizes of AA2011-T6 AMPs were 2.9 ± 0.3, 5.3 ± 0.4, and 11.8 ± 0.5 µm at feeding speeds of 3, 6, and 9 mm/min, respectively. It can be said that the reduction in grain size of AMPs deposited at 3, 6, and 9 mm/min feeding speeds reaches the values of 95.3%, 91.5%, and 82.25%, respectively, compared to the grain size of the AA2011-T6 initial material (62 ± 4 µm). The same results of very fine grains and refined second-phase particles are obtained by Dilip et al. [40] for the multi-layer friction deposits of AA2014-T6 (Al–Cu–Mg–Si alloy system). Consequently, the produced AMPs materials undergo continuous dynamic recrystallization and develop very fine equiaxed grains and refined precipitates [3,41]. In addition, Rutherford et al. [9] reported a significant reduction in the intermetallic particles and grain size after FSD processing of AA6061.

#### 3.3.2. AMPs Parts Produced from AA2011-O

Figure 7 represents the microstructures of the AA2011-O (Figure 7a) and the AMPs deposited at 200 rpm spindle rotation rate and different feeding speeds of 1 mm/min (Figure 7b), 2 mm/min (Figure 7c), and 3 mm/min (Figure 7d). A homogenous fine equiaxed structure has been noticed in all conditions due to the stirring of the grains accompanied with dynamic recrystallization during the additive friction-based process [11,20,42]. The mean grain sizes of AA2011-O AMPs were 0.84 ± 0.05, 0.88 ± 0.06, and 0.94 ± 0.08 µm at feeding speeds 1, 2, and 3 mm/min, respectively. It can be reported that the reduction in grain size of AMPs after FSD

reaches not less than the value of ≈98% compared to the grain size of the AA2011-O as-received material (48 ± 4 µm) without any significant difference between the applied feeding speeds.

**Figure 6.** Optical microstructures of the as-received AA2011-T6 (**a**), and AMPs produced t at a rotation rate of 1200 rpm and different feeding speeds of (**b**) 3 mm/min, (**c**) 6 mm/min, and (**d**) 9 mm/min.

**Figure 7.** Optical microstructures of the as-received AA2011-O (**a**) and AMPs produced from AA2011- O at a rotation rate of 200 rpm and different feeding speeds of (**b**) 1 mm/min, (**c**) 2 mm/min, and (**d**) 3 mm/min.

SEM was used to examine the present intermetallic precipitates of the as-received AA2011-T6 rod and the friction stir deposited materials at different conditions. Copper is the principal alloying element in AA2011 (Al–Cu alloys). However, other minor alloying elements (Fe, Ti, Zn, and Pb with traces of Ni, Si, and Mn) can also be specified as given in Table 1. During work hardening, an intermetallic phase (Al2Cu) is precipitated from a supersaturated solid solution. This intermetallic is crystallographically coherent with the Al matrix. Its fine dispersion improves the hardness and strength of the alloy [43]. The nondeformable second-phase precipitates initially present in the base material AA2011 have been fragmented into a smaller size and got uniformly distributed due to the severe plastic deformation involved in the FSD process; see Figure 8. This fragmentation phenomenon is expected in the stir zone of the friction stir welded materials [28,44] and the friction stir deposited materials [6,10,24].

**Figure 8.** Low and high-magnification SEM micrographs of (**a**,**b**) AA2011-T6 base alloy, (**c**,**d**) FSDed of AA2011-T6 at 1200 rpm—3 mm/min, (**e**,**f**) FSDed AA2011-T6 at 1200 rpm—6 mm/min, and (**g**,**h**) FSDed AA2011-O at 200 rpm—1 mm/min.

The fragmentation of the intermetallics may produce different shapes and sizes. The dispersion of micro and nanoparticles in the aluminum matrix affects the mechanical properties of the AMPs [45–47]. Figure 8 shows low and high magnification SEM micrographs of (a) and (b) AA2011-T6 base alloy (c) and (d) FSDed of AA2011-T6 at 1200 rpm–3 mm/min, (e) and (f) FSDed AA2011-T6 at 1200 rpm—6 mm/min, and (g) and (h) FSDed AA2011-O at 200 rpm—1 mm/min. The effect of the stirring process on the fragmentation and distribution of the intermetallic phases can be seen in (Figure 8c–h) compared with the AA2011-T6 base material (Figure 8a,b). Only two types of precipitates were detected in the base material and AMPs; see Figure 9. The EDS analyses of these precipitates are the rod-like shape (R) Al7Cu2Fe (spot 1 analysis in base material; Figure 8b and represented in Figure 9a) and the Al2Cu phase presents in different shapes given the same EDS analyses Figure 9b. These shapes are spherical (S, spot 2 in Figure 8b), small dots (S-D, spot 3 in AMP produced at 1200 rpm and 3 mm/min; Figure 8d), and almost spherical (A-S, spot 4 in AMP produced at 200 rpm and 1 mm/min; Figure 8g). The detected precipitates for both as-received materials and AMPs are consistent with that reported in the literature [37,43]. It can be remarked that the intermetallics are bonded well with the Al matrix for the as-received materials, as shown in Figure 8a,b. Hence, there is no pull-out detected after the grinding and polishing processes. The pull-out of intermetallics is detected for all the AMPs after grinding and polishing (Figure 8c–f). This indicates the weak bond at the interface between the dispersed intermetallics and the Al matrix as a result of subjecting to the FSD thermomechanical process.

**Figure 9.** EDS analyses of precipitates show two types of intermetallic precipitates: (**a**) Al7Cu2Fe (spot 1 analysis in Figure 8) and (**b**) Al2Cu (spot 2, 3, and 4 in Figure 8).

Figures 10 and 11 show the EDS elemental mapping of the AMPs produced from AA2011-T6 and AA2011-O at the processing parameters of 1200 rpm spindle rotational rate and 6 mm/min feeding speed and 200 rpm and 1 mm/min, respectively. Figure 10a shows the SEM image obtained from AMP of AA2011-T6 at 1200 rpm and 6 mm/min. It is indicated that the EDS elemental maps confirm the results of the chemical composition of AA2011 in terms of the overall chemical analysis of the alloy as can be seen in Figure 10b. In terms of the elemental maps of the different elements, it can be observed that the Al (c), Fe (e), and Si (f) are homogeneously distributed in their corresponding maps. However, the Cu map in (d) shows a high density of green-colored points where the particles are outlined in the SEM micrographs. Figure 11a shows the SEM image obtained from AMP of AA2011-O at 200 rpm and 1 mm/min. The elemental mapping showed a homogeneous distribution of all elements and phases with no preferences regions along the specimen (no clusters formation), which can be attributed to the severe homogeneous agitation of the material during the FSD process. It is also confirmed the possibility of the Al7Cu2Fe and Al2Cu intermetallics formation and location.

**Figure 10.** Elemental map distribution of AA2011-T6 AMPs produced at 1200 rpm and 6 mm/min: (**a**) SEM of AMPs, (**b**) map distribution of all alloying elements, and (**c**,**d**) map distribution of Al and Cu, respectively.

**Figure 11.** Elemental map distribution of AA2011-O AMPs produced at 200 rpm and 1 mm/min: (**a**) SEM of AMPs, (**b**) map distribution of all alloying elements, and (**c**,**d**) map distribution of Al and Cu, respectively.

#### *3.4. Hardness*

Hardness is an important mechanical property to evaluate materials, and its value is controlled by the chemical composition, temper condition, and the FSD process parameters. Figures 12 and 13 represent the average hardness values of the initial materials (AA2011 and AA2011-O) and the produced AMPs at different FSD parameters. The average hardness measurements of the AA2011 AMPs show lower hardness values in comparison to the base metal. Moreover, the hardness decreases as the feeding speed increases, as given in Figure 12. The decrease in hardness percentage (hardness loss %) of AA2011-T6 AMPs reaches the highest value of 49% compared to the AA2011-T6 material at processing parameters of a rotational rate of 1200 rpm and feeding speed of 9 mm/min. At the same rotation rate, the applied feeding speed of 3 mm/min produces an AMP with a hardness loss of 39% compared to the base material. This loss in hardness accompanying the additive manufacturing process was also reported by Dilip et al. [6] for the AMPs of AA2014. They concluded that the friction deposits showed inferior hardness because of their overaged microstructure.

**Figure 12.** Average hardness and a hardness loss percentage of the AA2011-T6 and its AMPs produced at a rotational rate of 1200 rpm and feed speeds of 3, 6, and 9 mm/min.

**Figure 13.** Average hardness and a hardness increase percentage of the AA2011-O and its AMPs produced at a rotational rate of 200 rpm and feed speeds of 1, 2, and 3 mm/min.

The average hardness values of AA2011-O alloy and the AMPs at a rotation rate of 200 rpm and feeding speeds of 1, 2, and 3 mm/min are illustrated in Figure 13. The hardness value of the AMPs produced at the applied feeding speeds is higher than that of the initial condition AA2011-O. This increase in hardness slightly decreases with the increasing feeding speed. The increase in the hardness value of AMP produced at a 1 mm/min feeding speed attains 163% compared with the AA2011-O starting material, whereas the improvement reaches 142% at the applied feeding speed of 3 mm/min.

The high hardness of the as-received AA2011-T6 (130 ± 3 HV) is ascribed to the high internal stresses stored in the material due to the previous production process, such as cold working. Afterward, the annealing process at 415 ◦C for 2.5 h followed by slow furnace cooling contributed to the stress relief of the alloy, coarsening, and softening of the second phase precipitates. Thus, the hardness of the alloy decreased after annealing, despite the reduction in grain size. It is well known that two mechanisms are controlling the hardness of this alloy. The first one is the reduction in grain size, which is directly proportional to the hardness increase. The second one is the morphology and dispersion of the second phase precipitates. In AMPs fabricated using the hard rods of AA2011-T6, the hardness decreased after deposition because the dominant mechanism that affects the hardness was the fragmentation of the hard precipitates. On the contrary, the hardness of the AMPs manufactured using the soft AA2011-O rods increased after friction deposition as the dominant mechanism, in this case, was the grain size reduction.
