Effect of Nano-Si₃N₄ Reinforcement on the Microstructure and Mechanical Properties of Laser-Powder-Bed-Fusioned AlSi10Mg Composites

Abstract

Laser powder bed fusion (LPBF) technology is of great significance to the rapid manufacturing of high-performance metal parts. To improve the mechanical behavior of an LPBFed AlSi10Mg alloy, the influence of nano-Si₃N₄ reinforcement on densification behavior, microstructure, and tensile property of AlSi10Mg was studied. The experimental results show that 97% relative density of the 3 vol.% nano-Si₃N₄/AlSi10Mg composite was achieved via optimization of the LPBF process. With the increase in the nano-Si₃N₄ content, the tensile strength and the yield strength of the composite steadily increase as per the Orowan strengthening mechanism while the elongation decreases. In addition, nano-Si₃N₄ reinforcement reduces the width of the coarse cell structure region and the thermal influence region of the AlSi10Mg matrix. After annealing, the tensile strength of the nano-Si₃N₄/AlSi10Mg composite decreases and the elongation increases significantly.

1. Introduction

Additive manufacturing (AM) technology is revolutionizing the way that traditional products are manufactured, via its unique layer-by-layer superposition fabrication manner based on a 3D model [1] that can help create complex-shaped products with good physical and mechanical properties [2]. Laser powder bed fusion (LPBF) is one of the advanced AM technologies that provide outstanding advantages in combining shape forming and sintering/melting/casting in a single step. At present, it has wide application prospects in aerospace, biomedicine, automobile engineering, and other fields [3,4,5,6]. However, the bottleneck in LPBF is the relatively limited number of metals suitable for this technology, such as some aluminum, titanium, and nickel alloys and stainless steel [7]. Therefore, expanding the material range is becoming more and more important for the development of LPBF.

Due to their high specific strength, corrosion resistance, and preparation convenience, Al–Si casting alloys are now the main aluminum alloy for LPBF and have been widely investigated for applications in aerospace and automotive fields. Among all the Al–Si casting alloys, AlSi10Mg attracts the most attention [8,9]. AlSi10Mg has been proven to be a typical material suitable for LPBF due to its good weldability, admirable hardenability, high thermal conductivity, and favorable corrosion resistance. The reasons for these excellent properties are that the near-eutectic composition of Al and Si reduces the solidification range and a small amount of Mg makes AlSi10Mg suitable for heat treatment to harden and strengthen the matrix [10]. Some scholars have studied the LPBF process of AlSi10Mg, often focusing on the laser parameters, porosity, microstructure, mechanical properties, and heat treatment [11,12,13,14,15]. For example, Cauwenbergh [16] studied the threefold inter-relationship between the process (LPBF and heat treatment), the microstructure at different scales (macro-, meso-, micro-, and nano-scale), and the macroscopic material properties of AlSi10Mg. Zhou [17] investigated the effects of curvatures on the geometrical performance, defects, microstructure, and mechanical properties of as-built AlSi10Mg parts with curved surfaces.

With the development of the automobile industry and other fields, the mechanical performance requirements for aluminum alloy are getting more stringent, which makes the development of ceramic-particle-reinforced aluminum alloy matrix composites extremely urgent. Powder metallurgy was once the essential technology to make particle-enhanced metal matrix composites (MMCs), but LPBF has become an important alternative way to produce MMCs at present [18,19]. Recently, high-density ceramic-particle-reinforced aluminum composites have been successfully fabricated by LPBF. Li et al. [7] prepared an 11.6 wt.% nano-TiB2-decorated AlSi10Mg composite (NTD–Al) with a crack-free microstructure, high tensile strength, excellent plasticity, and high microhardness. Gu et al. [20] prepared nano-TiC/AlSi10Mg composites that are suitable for SLM. Wang et al. [21] also prepared 3 wt.% TiC/AlSi10Mg composites by LPBF. Their experimental results showed that nano-TiC particles can significantly improve tensile properties but increase the porosity of nano-TiC/AlSi10Mg composites. Jiang et al. [22] prepared 1 wt.% CNT AlSi10Mg via SLM and found a positive relationship between laser scanning speed and composite mechanical properties. Gao et al. [23] prepared nano-TiN-enhanced AlSi10Mg. Their tests showed that the evenly distributed particles brought about excellent microhardness and wear properties to the composite. Chang et al. [24] used micro-SiC to produce the AlSi10Mg composite by SLM, with improved wear resistance. With the help of good crystal atomic matching at the Al–LaB6 interface, Tan et al. [25] achieved high-plasticity LaB6/AlSi10Mg composites. The above research manifests that suitable ceramic particles can make metal-based composites achieve better mechanical properties through LPBF. However, among these state-of-art LPBF studies, the nano-Si₃N₄-particle-reinforced Al matrix composite has rarely been investigated and little attention has been paid to the effect of heat treatment on the performance of nano-scale-particle-enhanced metal-based composites.

In this work, nano-silicon nitride, due to its high strength, low thermal expansion coefficient, superior corrosion resistance, and good thermal stability [26,27], was designed as an enhancement phase for AlSi10Mg. This study was focused on the LPBF process parameters, microstructure, and mechanical properties of nano-Si₃N₄-enhanced AlSi10Mg composites. Moreover, the effect of heat treatment on the mechanical properties of composites was explored.

2. Materials and Methods

Gas-atomized AlSi10Mg powder with a median particle size of 37.43 μm, as shown in Figure 1a,c, provided by AVIC Maite Powder Metallurgy Technology (Gu’an) Co., Ltd., (Langfang, China), and nano-Si₃N₄ powder with a median particle size of 50 nm, provided by Beijing Guoxin Hongyu Curtain Wall Material Co., Ltd.,(Beijing, China), were used as the feedstock powders in this study.

Table 1. Chemical composition of AlSi10Mg powders (in wt.%).

Element	Al	Si	Mg	Fe	Cu	Mn	Pb	Zn	Ni	Be
Content (wt.%)	Bal.	9.5	0.39	<0.1	<0.1	<0.45	<0.1	<0.1	<0.1	<0.1

shows the composition of AlSi10Mg powder. The contents of Si₃N₄ powders in nano-Si₃N₄/AlSi10Mg composites were set as 1, 3, and 5 vol.%, respectively. The Si₃N₄ and AlSi10Mg powders were mixed by ball milling, using a ball-to-powder weight ratio of 1:1, a rotation speed of 200 rpm, and a mixing time of 10 h. Figure 2 shows the SEM of mixed powders. AlSi10Mg powder maintained its original spherical shape after ball milling, while nano-Si₃N₄ particles were uniformly distributed on the surface of AlSi10Mg particles.

Pure AlSi10Mg and nano-Si₃N₄/AlSi10Mg composites were produced by LPBF via SLM equipment (SLM-100C, Guangdong Hanbang 3d Tech Co., Ltd., Zhongshan, China). The substrate was sandblasted and preheated to 150 °C before LPBF. During the LPBF process, argon was continuously filled in the chamber to ensure that the oxygen content was less than 50 ppm. Various combinations of laser powers (180–300 W), scanning speeds (300–800 mm/s), hatching spaces (0.03–0.07 mm), and layer thicknesses (0.03 mm) were explored to obtain the optimal LPBF parameters in order to fabricate samples with the highest possible relative density. The chess scanning strategy was used to fabricate block samples (6 mm × 6 mm × 5 mm) and tensile bars.

The relative densities of LPBF specimens were measured using the Archimedes methods (in accordance with ASTM B962), and 4 samples were measured at a time. The microstructure was observed and analyzed by a scanning electron microscope (SEM, Gemini500). In metallographic preparation, the polished surface was etched with Keller’s reagent (95% deionized water + 1.5% hydrochloric acid + 2.5% nitric acid + 1% HF) for 10–30 s. LPBF specimens were characterized using X-ray diffraction (XRD, Bruker, D8 ADVANCE A2) with a scanning range of 20–80°. Heat treatment at 180, 230, and 280 °C annealing temperatures was applied to the LPBFed specimens, respectively. The samples were heated in the furnace for 1 h and then kept in the corresponding annealing temperature for 6 h, followed by air cooling. The micro-Vickers hardness was tested using a microhardness tester (Taiming, HXD-1000TMC/LCD) with a load of 200 g and a dwell time of 20 s. The tensile properties were evaluated using a universal testing machine (Letry, PLD-5) with a displacement speed of 0.5 mm/min.

3. Results and Discussion

3.1. Densification Behavior

Figure 3 shows the relative density of nano-Si₃N₄/AlSi10Mg composites as the function of laser power (at laser scanning speed 500 mm/s and hatching space 0.05 mm), laser scanning speed (at laser power 200 W and hatching space 0.05 mm), and hatching space (at laser power 200 W and laser scanning speed 500 mm/s). It can be seen that as the content of Si₃N₄ nanoparticles increases, the relative density of the samples decreases and the optimum process parameter window shrinks greatly. Due to the strong agglomeration tendency caused by their high specific surface area, the nanoparticles may lead to an increase in the composite porosity. With the increase in the nanoparticle content, the fluidity of the powders becomes worse, which leads to more gas trapping between nanoparticles.

For all the nano-Si₃N₄/AlSi10Mg composites with different nano-Si₃N₄ contents, laser power, laser scanning speed, as well as hatching space have optimum ranges corresponding to the peak relative density. Although the energy input during LPBF can be monotonically controlled by laser power, laser scanning speed, hatching space, and layer thickness, the relative density of the nano-Si₃N₄/AlSi10Mg composites does not simply increase or decrease with these LPBF parameters. If the laser power is too low, porosity and incomplete melting will lead to a low relative density. If the laser power is too high, keyhole, spheroidization, crack, and other defects can also result in a low relative density. Other LPBF parameters have a similar effect on relative density. Based on the LPBF parameter evaluation in this study, the optimum LPBF parameters of each nano-Si₃N₄/AlSi10Mg composite are shown in

Table 2. The optimum LPBF parameters for the maximum relative densities of AlSi10Mg and nano-Si₃N₄/AlSi10Mg composites.

Samples	Laser Power/W	Scanning Speed/(mm·s−1)	Hatching Space/mm
AlSi10Mg	200	500	0.05
1 vol.% Si3N4	200	400	0.05
3 vol.% Si3N4	200	400	0.04
5 vol.% Si3N4	200	500	0.04

.

3.2. Microstructure

As shown in Figure 4, the microstructure of LPBFed nano-Si₃N₄/AlSi10Mg composites is mainly divided into three regions: (1) the inner part of the molten pool, which is a fine-cell-structure area with an average cell size of about 1 μm, (2) the boundary of the molten pool, which is a coarse-cell zone, and (3) the heat-affected zone. With the increase in the Si₃N₄ content, the cell structure becomes smaller. At the same time, the widths of its coarse-cell zone and heat-affected zone also decrease. Figure 4e shows that complete, undecomposed, Si₃N₄ particles still exist in the composite sample. From a theoretical point of view, the addition of Si₃N₄ particles affects the energy fluctuation and heat transfer in the molten pool and it also changes the heat transfer in the solidified region to a certain extent. Thus, it reduces the widths of the coarse-cell zone and the heat-affected zone. Furthermore, as Si₃N₄ particles are dispersed in the molten pool, the solid–liquid interface encounters and engulfs them during solidification at the boundary of the molten pool (see Figure 4e). Si₃N₄ particles hinder the movement of solid–liquid interface and eventually refine the cell structure. With the increase in the Si₃N₄ content, the distance between particles further decreases and the influence of refinement also increases.

3.3. Mechanical Properties

The tensile stress–strain curves of AlSi10Mg and nano-Si₃N₄/AlSi10Mg composites are shown in Figure 5, and the tensile test results are shown in Figure 6. The alloys tested in Figure 5 and Figure 6 were prepared using the optimized process parameters in Table 2. For example, a 1 vol.% Si₃N₄ sample was manufactured under the conditions of laser power = 200 W, scanning speed = 400 mm·s⁻¹, and hatching speed = 0.05 mm. As compared with pure AlSi10Mg, 1 vol.% Si₃N₄ does not result in noticeable change in elastic modulus and elongation. When the Si₃N₄ content is increased to 3 vol.%, the strength and elastic modulus increase slightly, while the elongation decreases slightly. Compared with the test results of the nano-Si₃N₄/AlSi10Mg composite with 3 vol.% Si₃N₄, the strength and elastic modulus of the 5 vol.% Si₃N₄ composites increased steadily, but the relative elongation decreased. In Figure 7, we used the same batch of powder as the previous experiment. This figure shows the tensile fractograph of the composite samples. The fractographs of the composites with different Si₃N₄ contents are similar to that of AlSi10Mg. In addition, there are a few small tearing edges and cleavage steps, which are typical features of ductile and brittle fracture, respectively. Figure 8 shows the XRD patterns of AlSi10Mg and nano-Si₃N₄/AlSi10Mg samples. Because the Si₃N₄ content is low, the diffraction peak of the Si₃N₄ phase is not detected in XRD patterns. The peak intensity and angle of Al and Si phase diffraction peaks are not influenced by Si₃N₄ particles.

As for the composite’s strengthening mechanism, the second phase strengthening and the fine grain strengthening are included. On the one hand, the hard/brittle Si₃N₄ phase uniformly dispersed in the matrix and blocked dislocation movement during deformation, increasing strength and reducing elongation. That is to say, the second-phase strengthening was the main strengthening factor. Moreover, with the increase in the Si₃N₄ content, the more obvious the increase in strength, the more obvious the decrease in elongation. On the other hand, as shown in Figure 4, the grain size decreased to a certain extent after the addition of Si₃N₄ and fine grain strengthening could also improve the strength of the material. In addition, the porosities of different alloys were tested (

Table 3. The relative density of Si₃N₄/AlSi10Mg samples.

Samples	1 vol.% Si3N4	3 vol.% Si3N4	5 vol.% Si3N4
Relative density	97.94 ± 0.31	97.68 ± 0.22	96.36 ± 0.65

), which indicated that the high Si₃N₄ content corresponds to the high porosity. The defects caused the alloy mechanical properties to deteriorate. Considering all the above factors, the strength of the alloy did not increase significantly. Therefore, the nano-Si₃N₄/AlSi10Mg composite with a 5 vol.% Si₃N₄ content shows the highest tensile strength and yield strength.

3.4. Heating Treatment

The 3 vol.% nano-Si₃N₄/AlSi10Mg composite was used for heat treatment tests because it had a high relative density (>97%) and better mechanical properties compared with the other nano-Si₃N₄/AlSi10Mg composites. Figure 9 demonstrates the XRD patterns of 3 vol.% nano-Si₃N₄/AlSi10Mg composite samples upon heat treatment under different annealing temperatures. The microstructure of a 3 vol.% nano-Si₃N₄/AlSi10Mg composite after annealing is shown in Figure 10. Annealing at 180 or 230 °C does not change the microstructure noticeably, while annealing at 280 °C seriously decomposes the eutectic structure and spheroidizes the Si phase because of the significant diffusion of the Si phase. When the annealing temperature is low, the diffusion ability of Si is not strong enough to change the eutectic structure. In fact, the meshed eutectic structure distributed along the cell boundary is not uniform and the eutectic Si phase at these locations is more likely to diffuse at a higher annealing temperature, resulting in the decomposition of the eutectic structure. In addition, Si in the solid solution begins to precipitate during annealing, contributing to the Si phase aggregation and granulation.

Figure 11 shows the tensile test results of an LPBFed 3 vol.% nano-Si₃N₄/AlSi10Mg composite at different annealing temperatures. To ensure the reliability of the data, twelve 3 vol.% Si₃N₄/AlSi10Mg samples were divided into four groups. One group was taken as the control group, and the rest were heat-treated at different temperatures. As the relevant changes in the mechanical properties of the sample could be clearly seen from the original experimental data, Figure 11 was drawn to show these changes: with the increase in the annealing temperature, the strength of the samples decreases while the elongation increases. The tensile fractograph is shown in Figure 12. We can see that, as the annealing temperature increases, the size and depth of the dimple increase, indicating the ductility of the composite increases. At the same time, the microhardness after annealing decreases slightly. The microhardness values of the composite after annealing at 180, 230, and 280 °C are 130 ± 1 HV, 113 ± 2 HV, and 108 ± 2 HV, respectively.

4. Conclusions

In this study, laser powder bed fusion (LPBF) was used to fabricate nano-Si₃N₄/AlSi10Mg composites with different Si₃N₄ contents. Firstly, the LPBF process parameters were optimized to achieve a higher density. Then, the effects of nano-Si₃N₄ content and heat treatment on the microstructure and mechanical properties of LPBFed composites were investigated. The main conclusions can be summarized as follows:

(1)

The relative density of nano-Si₃N₄/AlSi10Mg composites decreases with the increase in the Si₃N₄ content. LPBF parameters have been optimized to achieve a high relative density (above 97%) for 1–3 vol.% nano-Si₃N₄/AlSi10Mg composites.

(2)

Nano-Si₃N₄ particles noticeably refine the microstructure of nano-Si₃N₄/AlSi10Mg composites and improve the tensile strength as per the Orowan strengthening mechanism. The tensile strength, yield strength, and elongation of 5 vol.% nano-Si₃N₄/AlSi10Mg reach 448 ± 10 MPa, 311 ± 12 MPa, and 3.85 ± 0.16%, respectively.

(3)

After annealing, the strength of the 3 vol.% nano-Si₃N₄/AlSi10Mg composite decreases while the elongation greatly increases. Furthermore, annealing decreases the microhardness of the 3 vol.% nano-Si₃N₄/AlSi10Mg composite and a higher annealing temperature results in lower microhardness.