2019 Volume 60 Issue 10 Pages 2143-2150
Recently, the realization of the combination of selective laser melting (SLM) and high-strength aluminum alloys (Al alloys) has been discussed in mainly the aerospace industry. However, there have been several problems for the realization, and the one of fundamental problem is the occurrence of microcracks in high-strength Al alloy objects fabricated by SLM. To solve the problem, some researchers have studied addition of alloying elements to high-strength Al alloy. In this study, we focused on the A7075 alloy with additional five% Si (A7075+5Si) and investigated the effects of the Si addition on the processability, microstructure, and mechanical properties. As a result, it was observed that the addition of Si element inhibited the formation of cracks in the A7075 objects. Our result suggests that high density and crack-free A7075+5Si objects had higher mechanical properties than the case of a conventional Al alloy (AlSi10Mg and AlSi12). However, the T6 heat treatment (After solution treatment, water quenching, and artificially ageing) did not increase the mechanical properties of A7075+5Si objects. In order to obtain the A7075+5Si objects with higher strength, it is required that the further studies about suitable heat treatment conditions for A7075+5Si alloy.
This Paper was Originally Published in Japanese in J. Jpn. Soc. Powder Powder Metallurgy 66 (2019) 109–115.
Fig. 16 Tensile test results for A7075+5Si and A7075+5Si-T6 samples in both the horizontal and vertical directions and comparison to conventional A7075-O and A7075-T6 value of standard.
Selective laser melting (SLM) is an additive manufacturing technique, via which three-dimensional parts are created by repeatedly forming powder bed layers with a thickness of several tens of micrometers from metal powder and then melting and solidifying them using selective laser irradiation. SLM enables the formation of parts with complicated shapes that cannot be made using conventional manufacturing techniques such as casting or cutting. Considerable attention has been paid to the application of SLM in manufacturing high-performance and lightweight parts using topology optimization.
In the aerospace industry, SLM has been employed to fabricate lighter structural components using aluminum alloys, which have excellent specific strength.1–3) However, the range of aluminum alloys for SLM has been restricted to alloys based on Al–Si binary systems near the eutectic composition (e.g., AlSi10Mg and AlSi12), and the tensile strength of these SLM parts have remained approximately 460–490 MPa.4–6) In contrast, in 7000 series aluminum alloys, the mechanical properties of which are considered to be the most enhanced among all aluminum alloys, structural defects can easily occur during their solidification because they contain some alloying additions that are used to improve their mechanical properties by precipitation hardening. Therefore, it has been reported that microcracks occurring in 7000 series alloy parts fabricated by SLM, degrade their mechanical properties significantly.7–9)
Modifications of alloying constituents has been studied to suppress these microcracks.10–13) One proposed modified alloy is the A7075+Si alloy, where the addition of silicon prevents the occurrence of microcracks. It was reported that crack-free samples with 98.9% relative density could be obtained by processing A7075+Si alloys by SLM.13) However, appropriate laser conditions for processing A7075+Si alloys have not been investigated, and the microstructure and mechanical properties of parts fabricated using these alloys remain unclear.
In this study, a A7075+5Si alloy (containing 5% additional silicon) was prepared to fabricate SLM samples. The optimal laser condition for this alloy was investigated and the microstructure and mechanical properties of fabricated samples were evaluated. The optimal laser condition was determined by comparing the relative densities of samples that were fabricated under different laser conditions. The microstructure and mechanical properties of A7075+5Si samples fabricated under the optimal laser condition were evaluated and the effectiveness of conventional T6 heat treatment was discussed.
The material powders were A7075 and A7075+5Si pre-alloyed powders (TOYO aluminum K.K., JP). The powders were prepared by gas atomization. Table 1 shows their chemical compositions. Only the Si content of these powders were modified and the composition of other alloying constituents (Cu, Mg, and Zn) remained identical. Figure 1 represents the particle size distributions of the material powders. The average particle diameters of the A7075 and A7075+5Si powders were 26.7 µm and 23.6 µm, respectively. Figure 2 shows the scanning electron microscopy (SEM) images of these powders. The particles are almost spherical and significant differences between the material powders could not be observed. Figure 3 shows the cross-sections of the particles. A net-like microstructure was observed with no coarse precipitates. Figure 4 shows the results of energy-dispersive spectroscopy (EDS) linear scanning analysis conducted across the net-like microstructure in A7075 particles. On the lighter regions of the net-like microstructure, the concentration of aluminum decreased but that of other alloying constituents (Cu, Mg, and Zn) increased. For this reason, the net-like microstructures can be caused by micro-segregation of the alloying constituent and final solidification occurs in the lighter regions.
Particle size distribution of (a) A7075 and (b) A7075+5Si powders.
SEM micrographs of (a) A7075 and (b) A7075+5Si powders.
SEM micrographs of cross-sectional microstructures of (a) A7075 and (b) A7075+5Si particles.
EDS line scan profiles on cross-sectional microstructures of A7075 particle.
SLM samples were fabricated using the ProX DMP 200 (3DSystems, US) SLM machine. The laser unit equipped with the machine irradiates laser at a wavelength of 1070 nm and with a spot diameter of 80 µm. Figure 5 shows a schematic of the laser scanning path. The irradiated areas in each layer were divided into hexagonal regions with circumscribed diameters of 10 mm. The scan directions of laser were rotated by 90 degrees in every layer. The samples were prepared under an Ar atmosphere where the content of oxygen was controlled at 1000 ppm or less. The baseplate or material powders were not pre-heated.
Schematic diagram of laser scanning path.
To assess the laser conditions that could provide sufficiently dense sample, thirty laser conditions were tested by changing the laser power and scan speed in the range shown in Table 2. The relative density of the samples was measured by the Archimedes’ method, which compares the mass of samples in air and water. The samples were coated with paraffin wax to prevent penetration of water.
The microstructure of the samples was investigated by microscopic observations of the cross-sections of samples. A laser optical microscope VK-X150 (KEYENCE, JP) and a scanning electron microscope TM3030Plus (HITACHI high-technologies, JP) were used. The cross-sections of the samples parallel (horizontal plane) and perpendicular (vertical plane) to baseplate surface were polished using emery paper and diamond particles with diameters of 3 µm, and then finished to a mirror-like surface by chemical mechanical polishing using colloidal silica slurry. The microstructure was revealed by etching with certain etching reagents (NaOH: 2 g, NaF: 5 g, H2O: 93 g).
2.4 Tensile testThe mechanical properties of the A7075+5Si samples were determined using tensile testing. Square bars were fabricated under three different laser conditions that could form higher density A7075+5Si samples. The samples were machined to round bar tensile specimens with a diameter of 6.35 mm and gauge length of 25.4 mm. Three tensile specimens were tested for each laser condition. The tensile testing machine AG-XPlus (Shimadzu, JP) was used for the experiment. The temperature during the tensile test was 25°C and the testing rate was 10 MPa/s until the yield strength was determined, and 40%/min after. The optimal laser condition was determined by the results of tensile testing: 0.2% proof strength, ultimate tensile strength, and elongation at break. In addition, vertical tensile specimens where the longitudinal direction was perpendicular to the surface of baseplate were fabricated by optimal laser conditions, and their strength was compared with that of horizontal test specimens to evaluate the anisotropy of the mechanical properties.
To estimate the applicability of conventional heat treatment, T6 heat treatment (post-solution treatment, water quenching, and artificial aging treatment), which offers the highest static strength to 7000 series alloys, was applied to the fabricated samples; the strength of the heat-treated samples was tested using tensile testing. The treatment was conducted at 460–470°C for 2 h for solution treatment and at 115–125°C for 24 h for artificial aging treatment.
Table 3 shows the relative densities of the samples fabricated from A7075 and A7075+5Si powders as a function of laser conditions. The dashes in the table indicate that the samples could not be obtained as they were detached from the baseplate. It was revealed that the relative density of A7075 samples did not exceed 96.0% despite changing the laser conditions. Aluminum alloys commonly exhibit a 3–6% solidification shrinkage;14) thus, a decrease in their volume during cooling causes structural defects instead of deformations. In contrast, A7075+5Si samples with a relative density of nearly 100% were formed except using laser conditions with a high laser power and low scan speed or a low laser power and high scan speed. This result indicates the defects that appeared in the A7075 samples were prevented by the addition of silicon. The highest relative density was obtained at a laser condition of 234 W–1200 mm/s. Therefore, it was confirmed that samples with an almost true density could be fabricated by the addition of 5% silicon and appropriate laser conditions.
Figure 6 shows the relationship between energy density and relative density of the A7075 and A7075+5Si samples. The values of energy density represent the amount of heat supplied to per unit volume of material powder, correlated with the relative density of the SLM parts.15,16) Equation (1) represents the calculation formula of the energy density, where P, v, t, and s, indicate the laser power [W], scan speed [mm/s], layer thickness [mm], and scan pitch [mm], respectively.
| \begin{equation} E = \frac{P}{v\cdot t\cdot s} \end{equation} | (1) |
Relationship between energy density and relative density of (a) A7075 and (b) A7075+5Si specimens.
Figures 7 and 8 show the SEM micrographs of horizontal and vertical planes for A7075 and A7075+5Si samples, respectively. The samples were fabricated under the 234 W–1200 mm/s (E = 65 J/mm3) laser condition. As seen in Fig. 7, in the A7075 samples, microcracks along the building direction can be observed as well as pores over 80 µm in diameter. However, as shown in Fig. 8, microcracks were not observed in the A7075+5Si samples and only small pores were found. Figure 9 shows the SEM micrographs of the pores observed in the A7075+5Si samples. The diameter of pores was less than 30 µm; thus, the negative effects of these pores on the static strength of the samples would not be significant.17)
SEM micrographs of cross-sectional A7075 specimens (a) horizontal and (b) vertical plane.
SEM micrographs of cross-sectional A7075+5Si specimens (a) horizontal and (b) vertical plane.
SEM micrographs of porosities in A7075+5Si specimens.
Figures 10 and 11 show the laser microscopic images of the vertical plane of the samples after etching. The microcracks observed in the A7075 samples were categorized into solidification cracks formed within the melt pool tracks and macrocracks along the crystal grains elongated in the building direction. These macrocracks could be caused by contraction during solid-state cooling, propagating the solidification cracks. This is assumed to be because the length of macrocracks is greater than the layer thickness (30 µm). According to Fig. 11, every crack could not be observed in the A7075+5Si samples. The crystal grains in A7075+5Si samples elongated in the building direction similar to the A7075 samples; however, the grain sizes were smaller.
Optical micrograph of etched cross-sectional A7075 specimens (a) horizontal and (b) vertical plane.
Optical micrograph of etched cross-sectional A7075+5Si specimens (a) horizontal and (b) vertical plane.
Figures 12 and 13 show the SEM micrographs of microstructures observed in the vertical plane of the samples. Granular substructures were observed in the A7075 samples and elongated cellular substructures were observed in the A7075+5Si samples. These substructures were coarsened at the melt pool boundary region shown in the Figs. 12(c) and 13(c). Figure 14 represents the SEM micrographs and EDS elemental distribution focusing on the microstructures of the A7075+5Si samples. This result revealed that microsegregation occurred in the lighter region of the substructure as well as in the net-like microstructures observed in the cross-sections of the particles. Among the alloying constituents, silicon was significantly segregated. The difference in microstructures between the A7075 and A7075+5Si samples can indicate the changes in solidification behavior following laser irradiation. As for the A7075+5Si samples, additional silicon could lead to the retention of liquid metal at solid–liquid transitions because the substructure of these samples exhibited a higher area ratio in the lighter regions than that of A7075 samples; further, the substructure was of a net-like form. The liquid metal retained at the solidification interface provides healing effects for solidification cracks;18) therefore, it could be considered as a mechanism whereby microcracks are suppressed from additional silicon in the A7075 alloys.
SEM micrograph of (a) melt pool geometry, (b) substructure, and (c) substructure of boundary between melt pools for cross-sectional A7075 specimens.
SEM micrograph of (a) melt pool geometry, (b) substructure, and (c) substructure of boundary between melt pools for cross-sectional A7075+5Si specimens.
SEM micrograph and EDS mapping images on cross-sectional microstructures of A7075+5Si specimens.
Figure 15 shows the tensile test results for A7075+5Si samples fabricated under three different laser conditions. The first two laser conditions were 234 W–1200 mm/s and 169 W–1000 mm/s that provide particularly high relative density A7075+5Si samples, and the third condition was 208 W–1200 mm/s with intermediate laser power. The results showed that the ultimate tensile strength and elongation at break were significantly influenced by the laser conditions. An average ultimate tensile strength of 497 MPa was measured when the sample was fabricated under the 169 W–1000 mm/s laser condition; however, this increased to 537 MPa under the 234 W–1200 mm/s condition. The changes in elongation agreed with the trends in ultimate tensile strength; the elongation at break was 5.3% for the 169 W–1000 mm/s condition, but it increased to 9.7% for the 234 W–1200 mm/s condition. The samples fabricated under 208 W–1200 mm/s conditions exhibited intermediate tensile properties between those fabricated under 169 W–1000 mm/s and 234 W–1200 mm/s conditions. The 0.2% proof stress was approximately 360 MPa regardless of the laser conditions. The results indicated that among the laser conditions evaluated in this study, the laser condition of 234 W–1200 mm/s is optimal to process A7075+5Si alloys by SLM.
Comparison of tensile test results for A7075+5Si samples fabricated by three different laser conditions.
Figure 16 displays the anisotropy and changes in tensile strength owing to conventional T6 heat treatment of A7075+5Si samples fabricated by the 234 W–1200 mm/s laser condition. For comparison, the strength of A7075-O (annealed) and conventional A7075-T6 materials from the literature17) are also shown in Fig. 16. The A7075+5Si samples with an as-built condition show a 5% of anisotropy in the ultimate tensile strength and 40% in the elongation at break. This anisotropy arises from the elongated grains and substructures parallel to the building direction because the elongation at break was decreased when the load of the tensile test was applied along the building direction of the samples. Compared to conventional A7075-T6, A7075+5Si samples exhibited a lower 0.2% proof stress; however, the difference in the ultimate tensile strength remained only 10%. In contrast, the T6 heat-treated A7075+5Si samples showed a particularly low tensile strength than that of as-built conditions. The A7075+5Si-T6 samples showed favorable mechanical properties when compared to annealed A7075-O materials; however, they were considerably lower than conventional tempered A7075-T6 materials. The anisotropy of strength could be decreased by T6 heat treatment; thus, solution treatment could only be conducted during T6 heat treatment and the precipitation hardening could not be completed. This is because the appropriate time and temperature to complete precipitation hardening were altered by adding silicon; therefore, optimal heat treatment conditions for the A7075+5Si alloy will need to be investigated.
Tensile test results for A7075+5Si and A7075+5Si-T6 samples in both the horizontal and vertical directions and comparison to conventional A7075-O and A7075-T6 value of standard.
In order to apply SLM to the A7075 aluminum alloy, appropriate laser conditions for the A7075+5Si alloy were evaluated, and the microstructures and mechanical properties of the fabricated samples were investigated. The main results of this study can be summarized as follows;
