Microstructure Characterization and Mechanical Properties of Al6061 Alloy Fabricated by Laser Powder Bed Fusion

Abstract

Processing high-performance aluminum alloys, including 6xxx and 7xxx series, via laser additive manufacturing (AM) processes poses significant challenges, primarily due to the rapid cooling rates inherent in these processes, which often result in solidification cracking and metallurgical defects. This study aimed at producing dense, crack-free samples of Al6061 alloys, using the laser powder bed fusion (L-PBF) process. Taguchi’s method of design of experiments was employed to study the effects of laser power, scanning speed, and hatch spacing on the L-PBF process parameters for Al6061. Two types of samples were fabricated: cubic samples for density and microstructural analyses; and dog bone samples for tensile testing. The microstructure, density, mechanical properties, fractography, and material composition of the L-PBF Al6061 parts were investigated. Based on our experimental findings, an optimal process window is suggested, with a laser power of 200–250 W, scanning speed of 1000 mm/s, and hatch spacing of 140 µm, resulting in complete melting within the energy density range of 44–50 ${\mathrm{J}/\mathrm{mm}}^3$. This work demonstrates that adjusting processing conditions—specifically, increasing the energy density from 25.51 ${\mathrm{J}/\mathrm{mm}}^3$ to 44.64 ${\mathrm{J}/\mathrm{mm}}^3$—leads to a reduction in porosity from approximately 5% to below 1%, significantly improving the density and quality of the parts fabricated using L-PBF.

1. Introduction

Metal additive manufacturing (AM) methods have gained attention for their ability to create complex and customized parts layer by layer directly from part designs, thereby reducing material waste and assembly steps [1]. Laser powder bed fusion (L-PBF) is one of the most common metal AM methods already being adopted in aerospace, automotive, energy, and medical applications to produce high-performance parts [2,3]. The L-PBF process starts with scanning a laser beam over a layer of powder with a predefined thickness at a controlled speed, to selectively melt specific regions. Subsequently, the build plate is lowered by the layer thickness to allow the spreading of a new layer, and the process is repeated until the part is completely built [4].

Aluminum (Al) alloys are used in various industries—such as the energy, aerospace, automotive, and defense industries—due to their good strength, light weight, and low price compared to other metals [5,6]. The most common Al alloys in L-PBF are the Al–Si family (e.g., AlSi10Mg), as they offer excellent weldability and printability [7]. However, these alloys exhibit relatively low yield strengths (150–300 MPa) that fall short in applications where high performance is needed [8,9]. Processing high-performance Al alloys, including 6xxx and 7xxx series, via the L-PBF process poses considerable challenges, primarily due to the steep cooling rates associated with these processes, which often lead to solidification cracking and high thermal stresses [10,11,12,13]. These high-performance Al alloys also have a high temperature interval between solidus and liquidus temperatures because of their hypoeutectic composition, which increases their susceptibility to solidification cracking [14,15,16]. According to the literature, hot cracks are commonly attributed to solidification cracking, which occurs in the final stages of solidification, or to liquation cracking, which takes place at the grain boundary, because of elemental segregation [17,18]. In the case of Al6061, the dominant cracking mechanism identified is solidification cracking, as detailed in the open literature [19].

Al alloys, due to their significant reflectivity and thermal conductivity, are recommended to be fabricated using slower scanning speeds combined with high laser powers in laser AM [20]. These parameters are crucial because they significantly influence the quality of the final parts, including part density, microstructure, and mechanical properties [20]. Many studies in recent years have shown the formation of cracks during L-PBF of various Al alloys, such as Al7075 and Al6061 [21]. For example, Kaufmann et al. [22] reported that cracks were present in the L-PBF of Al7075 specimens fabricated at different laser powers, scanning speeds, and build-plate temperatures. Similarly, Maamoun et al. [23] found that solidification cracking was inevitable during L-PBF of Al6061, and the lowest crack density was obtained at a laser power of 500 W and a scanning speed of 1200 mm/s.

Several approaches have been discussed in the literature, to eliminate solidification cracking in high-performance Al alloys, including optimizing process parameters [14,21] and adding alloying elements [24,25,26,27,28]. Among these strategies, heating the substrate, i.e., build plate, shows beneficial effects, in terms of reducing cracks and residual stresses, since it reduces the thermal gradient [29]. Using platform preheating at 500 °C and a set of process parameters of 400 W and 1400 mm/s, Uddin et al. [30] obtained crack-free samples of Al6061 by using the L-PBF process. In another study by Aboulkhair et al. [31], the effect of hatch spacing on porosity formation in AlSi10Mg was investigated. The results showed that porosity increased with larger hatch spacing, as the fraction of gaps grew, due to insufficient overlap between the scan tracks. On the other hand, adjusting the composition of Al6061 via adding strengthening alloying elements and grain refiners leads to crack reduction, as mentioned in the open literature [32,33,34]. Chen et al. [35] showed that the addition of yttria-stabilized zirconia (YSZ) to Al6061 results in the almost complete elimination of cracking during the L-PBF process, due to the significant reduction in grain size and the formation of fine equiaxed grains along the melt pool boundaries. This grain refinement helps to accommodate the tensile strains generated during solidification, thereby suppressing the propagation of cracks along the grain boundaries. Moreover, Mehta et al. [36] investigated the effect of Zr on the hot-cracking susceptibility of Al6061 fabricated with L-PBF and found that the addition of 1 wt% Zr to Al6061 led to mitigation of crack formation as well as the reduction of internal porosity because of significant grain refinement. However, these internal cracks and porosity could be reduced further via optimizing the process conditions. Modifying the alloy by doping with alloying elements and optimizing the process conditions are promising approaches for reducing cracking in Al6061 during L-PBF [19].

Key process parameters in the L-PBF process include laser power, scanning speed, hatch spacing, and layer thickness [31]. These key parameters significantly influence the solidification mechanism, thermal gradient, and, consequently, susceptibility to cracking [37]. Therefore, identifying an optimal process map is essential for producing crack-free parts in the L-PBF AM process, particularly when working with high-performance alloys such as Al6061. The current literature shows a gap in understanding the optimal L-PBF process window for Al6061, considering a range of influential parameters. While previous studies have focused on reducing defects in Al6061 with alloying elements, such as yttria-stabilized zirconia (YSZ) [35] and zirconium (Zr) [36], this study extends the current understanding by examining the impact of key process parameters, such as laser power, scanning speed, and hatch spacing, on the density, microstructure, and mechanical properties of typical Al6061 without the addition of any alloying elements.

The present work aimed at finding the process window to produce fully dense and defect-free parts of Al6061. A design of experiment (DOE) approach was developed to investigate the effect of individual L-PBF process parameters and the volumetric energy density on Al6061 part quality, by evaluating their density, microstructure, and mechanical properties.

2. Methodology

2.1. Powder Feedstock

In this study, gas-atomized Al6061 powder, supplied by Valimet Inc., was used as the feedstock material. The powder size distribution (PSD) was measured using an automated particle size analyzer (Morphologi G3, Malvern Panalytical, UK). Powder morphology and chemical composition were analyzed using a field-emission scanning electron microscope (FE SEM, Zeiss Sigma, Germany) equipped with energy-dispersive X-ray spectroscopy (EDS, AZtec version 3.1 software, Oxford Instrument). X-ray diffraction (XRD) was performed using an automated diffractometer (Ultima IV, Rigaku, Japan) for phase identifications of the powder. The powder samples were exposed to ${\mathrm{Cu-K}}_{\alpha }$ radiation ($\lambda$ = 1.54056), using scanning speeds of 2°/min.

2.2. L-PBF Process Conditions

Samples were fabricated using an EOS M290 machine equipped with a 400 W Yb-fiber laser and a laser spot size of 100 µm. The build plate was preheated to 180 °C, and each layer was rotated by 67°. The layer thickness was maintained constant at 0.04 mm. A DOE, based on the L9 orthogonal array of the Taguchi method, was developed to investigate three factors: laser power, P (W); scanning speed, V (mm/s); and hatch spacing, h (µm); each at three levels, as outlined in

Table 1. L-PBF process parameters used in this study.

Sample	Laser Power (W)	Scanning Speed (mm/s)	Hatch Spacing (µm)	Energy Density (J/mm3)
1	150	600	140	44.64
2	150	1000	120	31.25
3	150	1400	100	26.78
4	200	600	120	69.44
5	200	1000	100	50.00
6	200	1400	140	25.51
7	250	600	100	104.16
8	250	1000	140	44.64 *
9	250	1400	120	37.20

* indicates higher laser power for the same energy density.

. The levels of each parameter were determined using data reported in the open literature [19,33,34,35,36]. Taguchi’s experimental design uses orthogonal arrays to arrange process parameters and determine the levels at which they should be adjusted. One of the primary goals of Design of Experiments (DOEs) is to design experiments in a cost-effective manner [38]. The choice of the appropriate DOE depends on the objectives of the experiment, the type of problem, and factors such as the number of experiments that can be conducted and the parameters to be investigated. By using orthogonal arrays, the Taguchi method reduces the number of experiments required while still providing reliable insights into the effects of multiple factors [39]. Given the complexity of the AM process, where numerous parameters influence the final part quality, it is common to focus on key AM factors, such as laser power, scanning speed, hatch spacing, and layer thickness. Therefore, volumetric laser energy density is often used as a consolidated parameter to represent these factors, even though it does not fully account for material properties [40]. An energy density-based approach allows for the identification of complete melting with an optimal range of process parameters, in which the final part achieves minimal porosity. Within that range, dense parts are fabricated, whereas lower or higher energy densities are likely to lead to undesirable defects, such as lack of fusion pores or keyhole porosity, respectively [41]. This energy density-based approach has been widely adopted in the literature for laser AM processes such as the L-PBF process. For instance, Maamoun et al. [23] and Roberts et al. [32] optimized the fabrication of Al6061 alloy based on energy density to minimize defects and optimize material properties in their studies. These researchers demonstrated that adjusting process parameters to achieve the optimal energy density range resulted in parts with reduced porosity and improved mechanical performance [23,32].

This DOE generated 9 experimental runs, covering a range of laser energy densities from 25.51 to 104.06 ${\mathrm{J}/\mathrm{mm}}^3$. Two types of samples were fabricated: cubic samples for density and microstructural analyses, and dog bone samples for tensile testing. Figure 1 shows the fabricated samples and the dimensions of the cube and dog bone specimens [38,39].

2.3. Microstructural Analysis and Phase Identification

As-built cubes were sectioned both along and perpendicular to the build direction. Specimens were ground using 600- and 1200-grit SiC paper and sequentially polished with 9 μm, 6 μm, 3 μm, and 1 μm diamond suspensions. The microstructure was analyzed using a field-emission scanning electron microscope (FE SEM, Zeiss Sigma) equipped with electron-backscattered diffraction (EBSD). To perform EBSD, the surface of the samples was mirror-polished using ion beam polishing with a plasma focused ion beam (PFIB, Helios Hydra, Thermo Fisher, USA). An area of 500 $\mathsf{\mu }$m in diameter was polished using a 30 kV, 60 nA Xe beam with spin milling. Three EBSD maps, each with an area of 400 × 300 $\mathsf{\mu }$${\mathrm{m}}^2$ and a step size of 0.75 $\mathsf{\mu }$m, were acquired with 15 kV, a 60 $\mathsf{\mu }$m aperture, and high current mode. Post-processing was carried out using Channel 5 software. Additionally, crack density was measured using ImageJ software (version 1.54g).

Moreover, the polished samples were prepared for XRD analysis, to detect the different phases and intermetallics present in the samples.

2.4. Density Measurement

Density measurements were performed on all cubic samples, to assess their porosity, using the Archimedes principle, following ASTM B962-23 [42]. The measured densities were compared to the reference value for bulk Al6061, which is 2.7 ${\mathrm{g}/\mathrm{cm}}^3$ [43], and presented in relative density form. The density of each sample was calculated using Equation (1):

$$\rho ={\rho }_w\times \left({\displaystyle \frac{W_1}{W_1-W_2}}\right)$$

(1)

where $W_1$ is the weight of the dry sample, $W_2$ is the immersed weight of the sample, and ${\rho }_w$ is the water density. The measurements were carried out using a scale with a resolution of 0.001 $\mathrm{g}$, equipped with an Archimedes measuring kit. Five measurements were taken for both the dry and immersed mass of each cubic specimen.

2.5. Mechanical Properties

The Vickers hardness values of the as-built specimens were measured on the polished surfaces under an applied load of 100 gf (HV0.1) for 10 s using Wilson VH1102 micro-hardness tester (Buehler Wilson, IL, USA). At least ten HV0.1 measurements were made on the polished cross-section of the samples. Rectangular dog bone-shaped tensile specimens were fabricated according to the ASTM E8/E8M standard [44]. The L-PBF Al6061 samples were built directly on the build platform with no additional supports. Tensile tests were conducted using the Instron-5966 testing system, equipped with a laser extensometer to measure displacement across the gauge section at a strain rate of 0.25 mm/min.

3. Results and Discussion

3.1. Powder Characterization

The morphology of the Al6061 powders is shown in Figure 2; “virgin powder” refers to as-atomized powder that has not been recycled or reused in the L-PBF process; this powder consists of nearly regular spherical particles with acceptable powder size distribution (PSD); “recycled powder”, on the other hand, refers to powder collected after the L-PBF build. Although recycled powder typically features agglomerated spherical particles in AM, no significant changes in morphology were observed between the virgin and recycled Al6061 powders. The chemical composition of the powders is detailed in

Table 2. Elemental composition of Al6061 feedstock.

Element [wt%]	Cr	Cu	Fe	Mg	Mn	Si	Ti	Zn	Al
Nominal composition	0.04–0.35	0.15–0.4	≤0.70	0.80–1.20	≤0.15	0.40–0.80	≤0.15	≤0.25	Bal.
Virgin powder composition	0.09	0.18	0.29	0.95	0.00	0.60	0.00	0.02	Bal.
Recycled powder composition	0.21	0.21	0.14	0.98	0.00	0.44	0.00	0.00	Bal.

. Figure 3 shows the EDS maps of the virgin powder, highlighting the uniformity of the elements as a result of the gas atomization process. A particle size distribution test indicated that the particle sizes ranged from 15 to 40 $\mathsf{\mu }$m ($D_{10}$ = 14.57 $\mathsf{\mu }$m, $D_{50}$ = 25.43 $\mathsf{\mu }$m, $D_{90}$ = 38.43 $\mathsf{\mu }$m), as shown in Figure 4a. Additionally, Figure 4b presents the XRD patterns of the Al6061 powder. The dominant crystal phase for the powder is $\mathsf{\alpha }$-Al (FCC) (powder diffraction file card: 04-003-1376), with three major peaks at around 45° (111), 52° (200), and 77° (220), indicating a strong presence of $\mathsf{\alpha }$-Al.

3.2. Characterization of Process-Induced Defects

Figure 5, Figure 6 and Figure 7 show the XY and XZ cross-sectional micrographs of the as-built Al6061 at different energy densities. Although the focus of this study was on the effect of energy density, it is important to clarify that these energy densities resulted from varying combinations of process parameters—specifically, laser power, scanning speed, and hatch spacing. As shown in Table 1, each combination of these parameters led to different energy densities, which, in turn, affected the defect formation in the parts. At low energy densities, lack of fusion pores was evident, as shown in Figure 5a–e. These images highlight areas with unmelted powder visible around the lack of fusion pores. This type of defect was typically observed at lower energy densities, where insufficient energy was applied to fully melt the powder, resulting in poor bonding between powder layers. The images in the XZ plane (Figure 5f–j) further confirm this, as irregular defects are visible. The defect area percentage was calculated using ImageJ software, but it is important to note that there are limitations to this technique, including errors due to defect overlap, defect depth, sampling bias, image resolution, and the use of a single surface image [45]. The calculated defect percentages are consistent with the SEM images (Figure 5, Figure 6 and Figure 7), which show that in the lack of fusion mode, the defect percentage was higher. This percentage decreased in the optimal energy density range, and further increases in energy density led to an increase in defects.

Parts that were produced at 44.64 ${\mathrm{J}/\mathrm{mm}}^3$ showed significantly fewer pores and cracks, and those at 50 ${\mathrm{J}/\mathrm{mm}}^3$ were defect-free, as shown in Figure 6a,b. At the same energy density of 44.64 ${\mathrm{J}/\mathrm{mm}}^3$, the samples in Figure 5e, produced at lower laser power, showed more defects than those in Figure 6a, produced at higher laser power. The increased defects in Figure 5e were likely due to insufficient melting, while the higher laser power in Figure 6a provided enough energy to reduce defects, despite the same energy density. Similar to the XY cross-sections, fewer defects were observed at energy densities of 44.64 ${\mathrm{J}/\mathrm{mm}}^3$ and 50 ${\mathrm{J}/\mathrm{mm}}^3$ (Figure 6c,d). This range of energy density provides enough energy to induce stable melting, where the material has fully melted, minimizing the formation of defects. The sample fabricated at the highest energy densities of 69.44 ${\mathrm{J}/\mathrm{mm}}^3$ and 104.16 ${\mathrm{J}/\mathrm{mm}}^3$ showed keyhole pores, as shown in Figure 7a–d. Both the XY and XZ plane images reveal the formation of keyhole porosity, which was a result of the higher energy density. These pores formed due to the high penetration of the laser into the deposited layers, leading to an excessively deep melt pool [46]. Figure 7e shows the different types of defects, including cracks, unmelted powder, lack of fusion, and keyhole pores. Overall, the number of defects decreased with an increase in laser power but increased with scanning speeds up to 1400 mm/s, due to lack of fusion pores. At high laser power, there was a slight increase in internal pores and cracks. A defect-free sample was achieved at a scan speed of 1400 mm/s and laser power of 200 W.

3.3. Crystallographic Characterization

Figure 8 shows the XRD results of the LPBF-fabricated Al6061 alloy at various energy densities: 25.51 ${\mathrm{J}/\mathrm{mm}}^3$, 50 ${\mathrm{J}/\mathrm{mm}}^3$, and 104.16 ${\mathrm{J}/\mathrm{mm}}^3$. The XRD patterns predominantly feature peaks corresponding to the $\mathsf{\alpha }$-Al (FCC) phase. Notably, the XRD spectrum indicated a preferred grain orientation along the (200) direction, suggesting a significant texturization of grains in this orientation. In L-PBF, the repeated cycles of melting and solidification create a high thermal gradient primarily along the build direction. In cubic metals, grain growth is favored along the $\langle 100\rangle$ direction, due to its lower packing efficiency. As a result, most grains tend to grow in the $\langle 100\rangle$ crystallographic direction [36]. As the energy density increased to 104.16 ${\mathrm{J}/\mathrm{mm}}^3$, additional phases, such as ${\mathrm{Mg}}_2$Si and ${\mathrm{AlMg}}_4$${\mathrm{Si}}_3$, emerged in the XRD patterns. This was likely a result of the higher temperatures generated by the increased energy density, which provided sufficient thermal energy for the precipitation of these phases [36].

3.4. Density

Figure 9 shows the relative density as a function of laser energy density. Within the evaluated parameters, the optimal value appears to be 44.64 ${\mathrm{J}/\mathrm{mm}}^3$ with a relative density of 99%. At this energy density, variations in laser power were observed to influence the material’s relative density significantly, as shown in samples 1 and 8. Despite maintaining the same energy density, configurations with lower laser power resulted in a lower relative density. This phenomenon can be attributed to the incomplete melting of the powder particles. Lower laser power, even when adjusted to achieve the optimal energy density, may not have provided sufficient energy distribution to fully melt the material, leading to lack of fusion defects. These defects, which contributed to the observed reduction in density, are clearly demonstrated in SEM Figure 5. Lower energy densities are associated with higher porosity, primarily due to the presence of lack-of-fusion pores. As the energy density increases, the melting and flow of the molten material improve, helping to fill these pores and increase the overall part density. However, at very high energy densities, microsegregation and vaporization of low-melting-point elements can occur, leading to a reduction in part density [47].

3.5. Microstructure and Crystallographic Texture

Figure 10 presents the electron backscatter diffraction (EBSD) inverse pole figure (IPF) X maps, along with the corresponding pole figure (PF) plots and IPF plots in the XY plane, for as-fabricated Al6061 samples fabricated at energy densities of 31.25 ${\mathrm{J}/\mathrm{mm}}^3$, 44.64 ${\mathrm{J}/\mathrm{mm}}^3$, and 104.16 ${\mathrm{J}/\mathrm{mm}}^3$. The band contrast (BC) images reveal the presence of cracks and porosities in all three samples. The samples fabricated at 31.25 ${\mathrm{J}/\mathrm{mm}}^3$ and 104.6 ${\mathrm{J}/\mathrm{mm}}^3$ exhibited a higher number of defects, which is consistent with the observations from the SEM images shown in Figure 5c and Figure 7b. According to Sonawane et al. [19], cracks in Al6061 typically initiate at high-angle grain boundaries (HAGBs) where the misorientation angles exceed 15 °C. This aligns with what was observed in the Al6061 samples in the current study, as shown in Figure 10. The pole figures were calculated based on the crystallographic orientations determined in the analyzed regions, and the PF and IPF plots indicate a texture randomization for all three samples. Notably, the texture strength increased from the sample fabricated at an energy density of 31.25 ${\mathrm{J}/\mathrm{mm}}^3$ to the sample fabricated at 104.16 ${\mathrm{J}/\mathrm{mm}}^3$, as demonstrated by the increasing maximum texture intensity. This trend is further supported by the more randomized color distribution in the IPF map of Figure 10(1b). The grain size ranges and average grain sizes for the three samples were as follows: For the sample fabricated at an energy density of 31.25 ${\mathrm{J}/\mathrm{mm}}^3$, the grain size ranged from 0.5 $\mathsf{\mu }$m to 121.5 $\mathsf{\mu }$m, with an average size of 61 $\mathsf{\mu }$m. For the sample at 44.64 ${\mathrm{J}/\mathrm{mm}}^3$, the grain size ranged from 0.5 $\mathsf{\mu }$m to 129.5 $\mathsf{\mu }$m, with an average of 65 $\mathsf{\mu }$m. Finally, for the sample at 104.6 ${\mathrm{J}/\mathrm{mm}}^3$, the grain size ranged from 0.5 $\mathsf{\mu }$m to 103.5 $\mathsf{\mu }$m, with an average size of 52 $\mathsf{\mu }$m. Although the average grain size number was lower for the sample fabricated at 104.6 ${\mathrm{J}/\mathrm{mm}}^3$, the number of fine grains in the sample fabricated at 31.25 ${\mathrm{J}/\mathrm{mm}}^3$ was higher. Specifically, the sample fabricated at 31.25 ${\mathrm{J}/\mathrm{mm}}^3$ had a higher frequency of grains below 20 $\mathsf{\mu }$m, with a total of 84, compared to the sample fabricated at 104.6 ${\mathrm{J}/\mathrm{mm}}^3$, which had a frequency of 66 below 20 $\mathsf{\mu }$m. The distribution of grain sizes in the 31.25 ${\mathrm{J}/\mathrm{mm}}^3$ sample showed a greater proportion of small grains, which could contribute to improved mechanical properties, as fine grains typically enhance strength [23]. In contrast, the broader distribution of larger grains in the 104.6 ${\mathrm{J}/\mathrm{mm}}^3$ sample could contribute to weaker grain boundaries and reduced mechanical strength, despite its lower average grain size. This observation is consistent with the results of the hardness and tensile tests shown in Figure 11 and Figure 12, respectively, where the sample with more fine grains (31.25 ${\mathrm{J}/\mathrm{mm}}^3$) exhibited higher hardness and tensile strength, while the sample with larger and more randomly distributed grains (104.6 ${\mathrm{J}/\mathrm{mm}}^3$) showed lower mechanical performance.

Figure 10. (a) Band contrast (BC). (b) EBSD maps. (c) IPF-X maps of XY plane, the corresponding pole figures (PF), and the inverse pole figures (IPF). (d) Grain size distribution of as-built Al6061 at energy densities of (1) 31.25 ${\mathrm{J}/\mathrm{mm}}^3$, (2) 44.64 ${\mathrm{J}/\mathrm{mm}}^3$, and (3) 104.16 ${\mathrm{J}/\mathrm{mm}}^3$.

Figure 10. (a) Band contrast (BC). (b) EBSD maps. (c) IPF-X maps of XY plane, the corresponding pole figures (PF), and the inverse pole figures (IPF). (d) Grain size distribution of as-built Al6061 at energy densities of (1) 31.25 ${\mathrm{J}/\mathrm{mm}}^3$, (2) 44.64 ${\mathrm{J}/\mathrm{mm}}^3$, and (3) 104.16 ${\mathrm{J}/\mathrm{mm}}^3$.

3.6. Microhardness

Figure 11 shows the hardness measurements of all the fabricated Al6061 test samples along the XZ and XY planes. The XZ plane consistently exhibited greater hardness than the XY plane in all samples. This trend was consistent with Han et al.’s [48] observation that the XZ plane has a slightly higher melt pool boundary density compared to the XY plane, in addition to the intrinsic anisotropy introduced by the LPBF process. The microhardness along the XZ plane increased as the energy density increased. This increase in microhardness correlated with a reduction in defects and a more uniform microstructure, as seen in the SEM micrographs. However, beyond a certain point, further increases in the energy density resulted in a gradual decrease in the microhardness, due to over-melting. Along the XY plane, the microhardness gradually increased as the energy density increased from 25.51 to 31.25 ${\mathrm{J}/\mathrm{mm}}^3$, but further increases in the energy density decreased the microhardness. This reduction in hardness is attributable to the growth of the melt pool and the corresponding decrease in the thermal gradient, which lowered the thermal and residual stresses. Consequently, the higher temperature and slower cooling rate led to increased grain size due to extended melt time, further reducing the hardness [49]. Similarly, Maamoun et al. [23] observed a comparable trend in their study, where further increases in energy density resulted in a decrease in hardness, primarily due to an increased solidification rate, which led to a coarser grain structure. Overall, the average microhardness measurements showed high standard deviation due to the effect of defects formed within the parts, and, thus, the results from the microstructure inhomogeneity [23].

Table 3. A summary of hardness values reported in the open literature and this study for L-PBF Al6061 parts.

Material	Energy Density (${\mathbf{J}\mathbf{/}\mathbf{mm}}^{\mathbf{3}}$)	Hardness (HV)	Ref.
Al6061-(Preheat = 200)	50	84	[23]
Al6061-(Preheat = 200)	123.3	71	[23]
Al6061-(Preheat = 200)	76.9	77	[23]
Al6061-(Preheat = 500)	20.41	54	[30]
Al6061	20.41	90	[30]
Zr-Al6061	79.8	116	[34]
Zr-Al6061	64.1	88.6	[36]
Al6061-(Preheat = 180)	50	100.3	This work

summarizes the hardness values of the Al6061 samples in the current study, compared to those reported in the literature. The higher hardness observed in our samples can be attributed to several key factors. The combination of laser power and scanning speed in our study was optimized to achieve an ideal energy density, leading to fewer defects compared to those reported by other researchers. Additionally, the preheating of the powder bed significantly mitigated the thermal barriers caused by the large thermal conductivity of Al6061, reducing the occurrence of thermal defects and enhancing the overall hardness of the fabricated parts [30].

Figure 11. Vickers hardness in XY and XZ planes at different energy densities.

Figure 11. Vickers hardness in XY and XZ planes at different energy densities.

3.7. Tensile Mechanical Properties

The tensile mechanical properties of the as-fabricated material at various energy densities are given in Figure 12. Samples manufactured at an energy density of 26.78 ${\mathrm{J}/\mathrm{mm}}^3$ exhibited the highest tensile strength, reaching 83 Mpa. In contrast, parts manufactured at 44.64 and 104.16 ${\mathrm{J}/\mathrm{mm}}^3$ had the lowest tensile strengths, at 45 MPa and 49 MPa, respectively. Overall, increasing energy density leads to lower ultimate tensile strength (UTS). As the energy density rises, overheating may occur, resulting in a larger average grain size, which contributes to the reduction in UTS [50]. The results from the EBSD analysis shown in Figure 10 indicate that the sample produced at the lower energy density (31.25 ${\mathrm{J}/\mathrm{mm}}^3$) exhibited smaller grain sizes compared to the sample produced at 44.64 ${\mathrm{J}/\mathrm{mm}}^3$. This smaller grain size in the 31.25 ${\mathrm{J}/\mathrm{mm}}^3$ sample was associated with higher hardness and tensile properties, whereas the sample produced at 44.64 ${\mathrm{J}/\mathrm{mm}}^3$, which had larger grain sizes, showed lower mechanical properties. As the energy density increased further to 104.6 ${\mathrm{J}/\mathrm{mm}}^3$, the crack formation along the grain boundaries became more prominent, leading to a further decrease in both hardness and tensile strength. The mechanical properties derived from the tensile tests are presented in

Table 4. Tensile mechanical properties for Al6061 samples produced at different energy densities.

Sample	Energy Density (J/mm3)	UTS (MPa)	Elongation (%)
6	25.51	67.19	1.56
3	26.78	83.37	2.54
2	31.25	74.25	2.28
9	37.20	52.52	1.43
1	44.64	64.35	1.56
8	44.64 *	45.19	1.07
5	50.00	57.60	1.81
4	69.44	53.90	2.56
7	104.16	48.99	1.24

* indicates higher laser power for the same energy density.

.

Figure 12. Engineering stress–strain curves of the as-built Al6061 samples fabricated at different energy densities [51].

Figure 12. Engineering stress–strain curves of the as-built Al6061 samples fabricated at different energy densities [51].

Figure 13 shows the fractography of the samples at energy densities of 26.78, 50, and 104.6 ${\mathrm{J}/\mathrm{mm}}^3$. All these samples underwent brittle fracture, as also observed by Qbau et al. [51] in their examination of L-PBF Al6061, which fractured immediately, indicating brittle fracture behavior. The defects observed on the fracture surfaces of these specimens included unmelted powder, cracks, and porosity. During tensile testing, microcracks formed and propagated along the coarse grain boundaries, which led to the distinct brittle fracture characteristics. In particular, long cracks were observed along the grain boundaries in the sample that was produced at the high energy density of 104.16 ${\mathrm{J}/\mathrm{mm}}^3$, as shown in Figure 10(3a). This is likely attributable to weak metallurgical bonding at the grain boundaries in these samples leading to a reduction in the mechanical properties [52].

3.8. Predictions on the Hot Cracking Sensitivity of Al6061

The susceptibility to solidification cracking of Al6061, Al7075, and AlSi10Mg alloys processed by Laser Powder Bed Fusion (L-PBF) was assessed using the Kou criterion, based on the Gulliver–Scheil model. This model predicts hot cracking susceptibility through the steepness of the solidification curve towards the final stages of solidification (i.e., as $f_s\to 1$) [36,53,54]. The solidification curves, generated using Thermo-Calc software version 2024a with the TCAL7 database, depict temperature as a function of the solid fraction.

As illustrated in Figure 14, Al6061 exhibited a sharp decrease in temperature near the end of solidification, indicating high cracking susceptibility. In contrast, AlSi10Mg, which typically processes without solidification cracks in L-PBF, showed a flatter solidification curve, correlating to a low cracking susceptibility index [55]. Al7075 also demonstrated high susceptibility, similar to Al6061. Despite being grounded in the theoretical assumptions of the Scheil equation—such as equilibrium partitioning, perfect mixing in the liquid phase, and no diffusion in the solid phase—this approach consistently aligns well with the practical outcomes observed in L-PBF processes for aluminum alloys, affirming its applicability in predicting actual cracking behavior [56].

4. Conclusions

This study explored the microstructural characteristics and mechanical properties of Al6061 alloy produced via L-PBF. Crack-free fabrication of Al6061 has been challenging; therefore, optimization of process parameters is crucial, to achieve high-quality parts. Our findings indicate that controlling the energy density is key to optimizing both the microstructure and mechanical properties. Specifically, an optimal energy density range of 44–50 ${\mathrm{J}/\mathrm{mm}}^3$ was identified, which significantly mitigated defects while enhancing part density.

The successful fabrication of crack-free and pure Al6061 using L-PBF was demonstrated by optimizing energy density alone, without the need for nucleation or other alloy-modifying aids. This was achieved by fine-tuning laser power, scanning speed, and hatch spacing to identify the optimal process window.

In conclusion, this work demonstrates that adjusting processing conditions—specifically, laser power, scanning speed, and hatch spacing—can effectively eliminate defects in L-PBF-fabricated parts. This paper also delved into the underlying physics of porosity and solidification cracking, exploring how these phenomena relate to process parameters and their impact on part characteristics and overall quality. Future work will focus on fine-tuning these processing parameters to further optimize the L-PBF fabrication of Al6061. We aim to examine these refined parameters for their potential in industrial applications, ensuring the successful processing of Al6061 for a wide range of practical uses. In future work, we will explore adding critical minerals and rare-earth elements as grain refiners and micro-straightening alloying agents to eliminate cracks in Al6061, enhancing its application in extreme environments, such as energy storage systems.