Additive manufacturing of ceramic particle-reinforced aluminum‐based metal matrix composites: a review

Abstract

Laser powder bed fusion (LPBF) is one of the most widely used additive manufacturing methods for fabricating metal components. It is possible to produce multi-material structures and functionally graded materials with LPBF. The usability of powder mixtures provides a great potential for the production of metal matrix composites (MMCs) with advanced mechanical properties. Among the wide variety of MMCs, aluminum matrix composites are highly potential candidates for aerospace, automotive and biomedical applications due to their outstanding properties including high wear resistance, better chemical inertness and excellent mechanical properties at elevated temperature. Therefore, in this study, ceramic particle-reinforced Al-based MMCs produced by LPBF method are reviewed for the recent developments. Feedstock preparation methods for MMCs are emphasized. The effects of reinforcement particle properties and LPBF process parameters on the microstructure, densification behavior, hardness and tensile properties are discussed comprehensively. The strengthening mechanisms that occur with the addition of ceramic reinforcement are examined. Summary of the findings from this review and trends for future research in the development of Al-based MMCs by LPBF are addressed in the final section.

Introduction

In recent years, laser powder bed fusion (LPBF) has become an emerging additive manufacturing (AM) technique that allows the production of complex parts. In the LPBF methods, parts are fabricated layer by layer based on computer-aided design (CAD) models [1]. This layered structure enables for much greater design freedom with almost no geometric restrictions. Therefore, AM is highly promising for many high-value industries such as aerospace, automotive, defense and biomedical that require high-performance products [2,3,4,5].

LPBF methods have a lot of advantages over conventional manufacturing techniques. AM technologies offer the advantages of direct production, design and processing flexibility, near-zero material wastage and net-shaped fabrication without requiring expensive molds and tools [6, 7]. Due to the complex geometry of the components used in the aerospace and automotive industries, LPBF methods can provide an extraordinary opportunity to reduce assembly cost, especially for these industries. LPBF methods can be rapidly used to investigate the manufacture of components used in specific applications with new materials. Components redesigned by designers with topology optimization can be produced easily and quickly with LPBF methods [8]. It is possible to produce multi-material structures and functionally graded materials with LPBF [9,10,11]. As part production in LPBF methods relies on rapid melting and solidification processes, parts have substantially refined microstructures and therefore high hardness and strength [12, 13]. The usability of powder mixtures provides a great potential for the production of metal matrix composites (MMCs) with advanced mechanical properties. However, there exist several significant challenges that adversely affect the properties of additively manufactured parts such as high porosity and surface roughness [14], high residual stress [15] and non-uniform microstructure [16]. Several post-processing operations have been adopted to enhance the surface and mechanical properties [17]. Among these post-processes, especially heat treatment and hot isostatic pressing (HIP) are utilized effectively to reduce the porosity and residual stresses in the material structure [18, 19].

Composites are consisting of two or more components that show the characteristics of a single material, but are chemically different from each other. By producing composites, it is possible to obtain some properties that cannot be obtained when the materials used alone. MMCs consist of a mixture of metal matrix and ceramic reinforcements and are also known as ceramic-reinforced metal matrix composites. MMCs have many superior properties compared to traditional metals such as better thermo-mechanical properties, excellent chemical inertness, high wear resistance, low coefficient of thermal expansion and better fatigue resistance [20,21,22,23]. Because of these superior properties, they are used in extreme operating conditions such as high load, high temperature and extensive wear operations [24,25,26,27]. MMCs are widely used in aerospace, biomedical, electronics and various engineering applications [28,29,30]. Various methods such as powder metallurgy [31], squeeze casting [32] and stir casting [33, 34] are used to fabricate particle-reinforced MMCs. However, these techniques, which require expensive and special tools, are insufficient, especially in the production of complex-shaped components. In addition, due to the high hardness and melting point of the manufactured components, it is very difficult to process with traditional machining techniques [35, 36]. LPBF methods have opened up new opportunities in the production and development of MMCs, thanks to their advantages over traditional manufacturing methods.

This study is a comprehensive literature review on the production of particle-reinforced MMCs by LPBF methods. Within the scope of the study, research studies in which Al alloys were used as matrix and different ceramics as reinforcing particles are examined. The microstructural properties, densification behavior and mechanical properties of ceramic particle-reinforced MMCs are compiled and discussed in detail.

Laser powder bed fusion for fabricating MMCs

Metal additive manufacturing methods such as LPBF, directed energy deposition (DED) and electron beam melting (EBM) are potential promising alternatives in the production of metal matrix composites. Among these methods, it is possible to produce highly dense AMCs with minimal defects by overcoming the problems encountered in conventional processing with the LPBF method [37]. LPBF offers significant variety for the processing of various powder materials. During the laser melting process, new constituents can be formed between the melted powders, allowing different materials to be obtained from the initial powder mixtures. The laser used to fuse the powders also triggers chemical reactions and as a result, secondary phases are formed in the part. In this complex process, it is possible to produce complex-shaped composites with a uniform distribution of secondary phases. Compared to the LPBF method, the high layer thickness and low internal cooling rate in the DED method cause the formation of coarse microstructure and high porosity in the built part [38, 39]. In the LPBF method, rapid melting and solidification rates provide fine microstructures and thus high mechanical properties. Moreover, higher accuracy and better surface quality are obtained in the LPBF method compared to other metal AM techniques [40].

LPBF methods are classified into three basic categories: selective laser sintering (SLS), selective laser melting (SLM) and direct metal laser sintering (DMLS) [41]. DMLS is a trademark of EOS and is often used to describe SLM [42]. In SLS, metal powders are sintered with a focused laser at a temperature below the melting temperature of the powder [43]. Since the melting temperature does not have to be reached, alloys containing materials with different melting points can be easily produced. On the other hand, in the SLM method, metal powders are completely melted, resulting in a homogeneous part [44]. The schematic representation of the LPBF method is presented in Fig. 1.

Figure 1

Schematic representation of the LPBF method

Part production in the LPBF method includes the following steps:

• A powder layer of a certain thickness (typically 0.3–0.6 µm) is spread over the build platform.

• A laser fuses the first powder layer of the model.

• A new powder layer is spread over the melted powder layer using a roller.

• Further layers are fused and added.

• The process is repeated until the entire model is built.

There are several parameters that control the processing and directly affect the part quality in the LPBF. Various parameters such as process parameters (laser power, scanning speed, layer thickness, hatch distance, scanning strategy, etc.), material parameters (powder morphology, powder size, powder shape, etc.) and machine characteristics (laser type, atmosphere, etc.) should be chosen carefully. Appropriate selection of these parameters is essential to reduce internal defects inside parts [45].

LPBF has become an alternative method in composite production as it allows the use of powder mixtures as feedstock. In recent years, many research works have been carried out examining the composite production process by using LPBF. There are very few studies on the fiber-reinforced MMCs in the literature [46, 47], and these studies have generally focused on particle-reinforced MMCs. Therefore, this paper covers the emerging research on ceramic particle-reinforced Al-based MMCs.

Preparation of composite feedstock

Composite powder feedstock is required as starting material in the production of MMCs with LPBF. Size, morphology, weight/volume ratio and dispersion pattern of composite powders play significant role in determining part quality. Feedstock preparation with matrix material and particle reinforcements can be carried out with different methods such as regular mixing [48], ball milling [49], flux-assisted synthesis [50], agent-assisted deposition [51] and electroless plating [52], which of these methods chosen is of vital importance for the composite powder properties and the resulting part quality. Regular mixing and ball milling methods have attracted a great deal of attention in recent years due to their higher practicality, time and cost efficiency, and applicability to a wide range of materials, and are therefore known as the most applicable processes for the preparation of composite powders. Regular mixing and ball milling methods are schematically presented in Fig. 2.

Figure 2

Schematic representation of a ball milling and b regular mixing processes

The part quality is closely related to the powder morphology in LPBF. Spherical powders are spread more uniformly on the build platform with the recoater blade during production. For this reason, spherical powders produced by gas atomization [53, 54] and plasma atomization [55,56,57] are mostly used in the LPBF method. Distortion of the spherical shape of the powders in the preparation of composite feedstock adversely affects the part quality. Figure 3 shows the spherical deformity of the TiN/AlSi10Mg composite feedstock prepared by wet ball milling followed by ultrasonic mixing [19]. In all feedstock preparation methods, the spherical shape of the powders is generally distorted, but mechanical mixing methods such as regular mixing and ball milling attract attention and are frequently used due to their time and cost efficiency and applicability to various materials [58]. A summary of the studies using ball milling and regular mixing methods is presented in Table 1. Han et al. [59] investigated the effect of milling and pause duration on the fabrication of ball-milled Al/Al₂O₃ composite feedstock. They found that a combination of the short milling (10 min) and long pause (15 min) duration is suitable for preparation of composite feedstock.

Figure 3

a Pure AlSi10Mg powders and b nTiC/AlSi10Mg composite powders after wet ball milling followed by ultrasonic mixing [19]

Table 1 Summary of the literature review of the studies using ball milling and regular mixing methods to prepare composite feedstock

Whether the prepared composite feedstock is suitable for LPBF is determined using some metrological and analytical techniques. Composite powder flowability plays an important role in the uniform laying of powders on the powder bed in the LPBF method. Good-flowing powders generate powder layers with continuous and uniform thickness, reducing the detrimental dimensional effects and providing better mechanical properties [60]. The Carr index and Carney or Hall flow testing methods can be used to evaluate the flow behavior of composite feedstock. Powders with a Carr index of < 15% are considered to have good flowability.

The suitability of prepared composite feedstock for LPBF also directly depends on the morphology and size distribution of the powders, and the homogeneous mixing of the reinforcement and the matrix [61]. Powders that retain their spherical shape are more suitable for LPBF because they increase powder flowability and small particles provide uniform layer thickness. Powders morphology can be determined by analyzing powders under scanning electron microscope (SEM) and optical microscope. Particle size analyzers can be used to evaluate powder size distribution, and whether the reinforcements are uniformly dispersed in the matrix can be determined by SEM and transmission electron microscope (TEM) analyses.

Matrix materials and reinforcement particles

With LPBF method, MMCs are produced with a wide variety of matrices including Al [62], Ti [63], Fe [64, 65] Cu [66, 67] and Ni [68] and reinforcing elements including carbides (WC, SiC, B₄C, TiC) [69], oxides (Fe₃O₄, ZrO₂, Al₂O₃) [62, 70], nitrides (ZrN, Si₃N₄, TiN) [71, 72], borides (TiB, ZrB₂, TiB₂, WB) [73] and different forms of carbon (graphite, carbon nanotubes, graphene) [74]. Among these, Al as a matrix material and ceramic particles such as SiC, TiC, TiN, TiB₂ and Al₂O₃ as particle reinforcements are the most frequently used [75, 76].

In composite materials, the relationship between the matrix and the reinforcing particles is one of the most significant factors determining the composite properties. The chemical composition, type, shape and size of the reinforcing particles determine the functionality of the composites. Therefore, the selection of reinforcing particles is very crucial. In the following sections, research works using Al and Al alloys as matrix material and ceramic particles as reinforcements are extensively studied.

Reinforced by SiC particles

SiC is one of the most widely used ceramic particle reinforcements in the production of MMCs by the LPBF method. SiC is preferred due to its compatibility with Al-based matrix material and its superior properties [77,78,79,80]. The basic physical properties of SiC are listed in Table 2 [81, 82]. SiC particle-reinforced MMCs are typical candidate materials in the aerospace, automotive and military industries due to their high strength, high hardness and modulus, and high corrosion and wear resistance [83].

Table 2 Basic physical properties of SiC [81, 82]

Microstructural characterization

When the composite feedstock prepared with SiC and Al powders is irradiated by laser beam, Al powders melt completely, while SiC particles partially melt or do not melt due to the high melting temperature. SiC particles come into contact with liquid Al and intense chemical reactions occur between SiC and liquid Al, and new phases are formed as a result of these reactions. When the powder bed temperature exceeds 1670 K, the following reaction occurs:

$${\text{4Al }} + {\text{ 4SiC }} \to {\text{ Al}}_{{4}} {\text{SiC}}_{{4}} + {\text{ 3Si}}$$

(1)

Another reaction that occurs when the temperature exceeds 940 K [84]:

$${\text{4Al }} + {\text{ 3SiC }} \to {\text{ Al}}_{{4}} {\text{C}}_{{3}} + {\text{ 3Si}}$$

(2)

In order to evaluate the microstructure of SiC-reinforced Al matrix composites produced by LPBF, it is necessary to analyze the reactions occurring at the interface well. There are two theories to explain the reaction mechanisms between SiC and Al matrix: melting theory and reaction theory [85, 86]. The melting theory proposes that SiC and Al melt simultaneously and phases such as Al₄SiC₄ and Si precipitate during solidification. The reaction theory suggests that SiC particles react with molten Al at the interface and different phases are formed on the particle surface. Cons, the melting theory occurs at high temperatures, while the reaction theory occurs at low temperatures. It has been reported that the interface products formed with the combination of both theories in the LPBF method [87].

Xue et al. [88] used the LPBF method at different scanning speeds to produce SiC-reinforced AlSi10Mg composite. They found that the SiC density in the structure decreased with the decrease in the scanning speed. As the scanning speed decreases, the interaction time of the laser beams with the powders increases. The chemical reactions between SiC and liquid Al also increase, resulting in a decrease in SiC density. They also determined that the Al₄SiC₄ phase was formed at the interface and that there were no micropores and microcracks at the interface as shown in Fig. 4 [88]. The absence of micropores and microcracks at the interface indicates that the bonding between the matrix and the particle reinforcement is strong, which positively affects the mechanical properties of the composite. However, the difference in the coefficient of thermal expansion between the particle reinforcement and the matrix can sometimes cause cracks and porosities at the interfaces [89]. These cracks and porosities reduce the load transfer capacity from the matrix to the reinforcement. The increase in laser energy density due to the increase in laser power also increased the chemical reactions at the interface and caused a decrease in the SiC density in the microstructure and the formation of reaction products of Al₄C₃ [87]. The increase in laser energy density caused more decomposition of SiC in the Al matrix, which increased the reaction products [90].

Figure 4

Optical Microscope a, b and Scanning Electron Microscope c, d, e images of the SiC-reinforced AlSi10Mg composite [88]

Chemical reactions at the interface can also vary depending on the SiC particle size. As the SiC particle size decreases, the interfacial reactions increase and the amount of newly formed phase (Al₄SiC₄) is observed in a larger volume fraction [91, 92]. This is due to the relatively higher specific surface area of the small-sized SiC particles, thus forming a stronger bonding with the Al matrix.

With the addition of SiC reinforcement, it was observed that the average grain diameter decreased by 30.99% and SiC agglomerations were observed in some regions of microstructure [93]. SiC particles in molten pools provide a large number of nucleation sites, enabling the formation of fine recrystallized grains and restricting the growth of Al grains. In addition, as a result of XRD analyses, it was determined that Al, Si, Mg₂Si and SiC phases were formed in the microstructure, as well as nanosized reaction products Al₄C₃ phase. The different phases that occur in the microstructure of the specimen produced by the LPBF method by adding nano-SiC particles to the AlSi₇Mg matrix are schematically presented in Fig. 5 [93].

Figure 5

Schematic representation of the formation mechanism of different phases during LPBF method of AlSi₇Mg/nano-SiC composite [93]

Densification behavior

The densification behavior of particle-reinforced MMCs produced by LPBF method depends on many factors. These factors can be reinforcement size, fraction and shape, as well as LPBF processing parameters. In addition, the temperature in the powder bed during production is very crucial on the densification behavior. The laser absorption of the composite feedstock has a notable importance on the densification behavior of the composites, as it has a significant effect on the powder bed temperature. The porosity values of the SiC-reinforced and non-reinforced Al powders were measured as 0.25 + 0.11% and 0.59 + 0.79%, respectively [93]. After the addition of SiC particles, the laser reflectivity decreased, that is, the laser absorption increased [94, 95]. The laser reflectivity of SiC-reinforced and non-reinforced Al alloys was determined as 31.71% and 23.05% at 1064 nm wavelength, respectively [91]. The increase in laser absorbance helped to improve the liquid–solid wettability by increasing the molten pool temperature. Moreover, increased laser absorption provides a more stable melt pool formation.

The decrease in SiC particle size significantly increases the relative density. When coarse SiC particles are used, the particles cannot interact with the laser for a sufficient time and remain in the structure almost unmelted, which limits the microstructural homogeneity. Coarse SiC particles increase the melt viscosity and cause the melting flow to be irregular. As a result, large-sized pores are formed in the structure due to the instability of the molten pool and the balling effect [92]. With the decrease in the particle size, the laser absorption of the powder mixture and thus temperature in the molten pool increase, leading to the increase in the Marangoni flow [96]. In addition, high laser absorption causes an increase in the molten pool size and a decrease in the viscosity of the blended melt [95]. Consequently, the increase in fluidity in the molten pool provides a more homogeneous distribution of SiC particles within the structure [91]. Reducing SiC particle size also increases chemical reactions at the interface. Thus, the wettability of liquid Al and SiC particles increases, preventing the formation of porosities and cracks at the particle–matrix interface.

Figure 6 shows OM and SEM micrographs and relative density of the as-built AlSi10Mg/SiC composites fabricated under different laser powers [97]. The densification rate of Al-based SiC-reinforced composites increases as the applied laser energy density increases [87, 90]. Increasing the laser power or decreasing the scanning speed elevates the molten pool temperature. With the increase in the molten pool temperature, SiC particles melt at a higher rate, reducing the viscosity. The increase in laser intensity also allows the molten pool to remain molten for a longer time, which increases the fluidity and facilitates the liquid metal flow in the molten pool. Thus, the liquid melt can easily fill the pores [98]. At high laser intensities, the molten pool temperature rise excessively, resulting in a decrease in the densification rate. Extremely high temperatures, which may exceed the boiling point, can cause vaporization of alloying elements [99]. Vaporization of volatile elements causes the formation of keyhole-shaped pores in the molten pool [97]. As discussed above, proper selection of LPBF process parameters is critical for fabricating nearly fully dense parts.

Figure 6

Optical micrographs and SEM images of as-built AlSi10Mg/SiC composites at different laser power of 120 W (a, a₁), 150 W (b, b₁), 180 W (c, c₁), 210 W (d, d₁), 240 W (e, e₁) and the variation of relative densities f [97]

Hardness and tensile properties

The hardness and tensile properties of the Al-based MMCs produced by the LPBF method are significantly associated with the density level and microstructure. The addition of SiC particles increases the composite hardness [93]. As a result of the interfacial reactions with SiC, the precipitate phases and interfacial products are formed, which leads to the precipitation strengthening effect. In addition, the microstructure refined with SiC reinforcement is another reason for the hardness increase. According to the Hall–Patch formula, the microhardness is improved due to the grain refinement [100].

Figure 7 shows the effects of coarse, medium and fine SiC particles on microhardness and its distribution [92]. The use of coarse SiC particles enables the formation of large-sized pores in the microstructure. Moreover, the SiC particles remain nearly unmelted in the microstructure and the interfacial bonding is weak [92]. As the SiC particle size decreases, the grain size decreases and almost full density is achieved, stronger interfacial bonding is formed between SiC and Al matrix and the amount of interfacial product increases [91]. The hardness difference between the SiC-reinforced AlSi10Mg composite and the unreinforced AlSi10Mg can vary up to 40% [91, 101]. Consequently, hardness increases with the decrease in particle size can be attributed to three strengthening mechanism: precipitation strengthening, grain refinement strengthening and strong interfacial bonding. Increasing the SiC volume fraction also increases the hardness [89].

Figure 7

Effect of SiC particle size on microhardness and its distribution [92]

The increase in the laser intensity, which depends on the laser power and scanning speed, increases the hardness [90]. Increasing the laser intensity also strengthens the interfacial bonding by increasing the chemical reactions between the SiC particles and the Al matrix. However, increasing the laser intensity too much can increase the keyhole porosity as discussed in the densification behavior section and these keyhole porosities may cause a decrease in the hardness [97]. However, in general, increasing laser intensity (increasing laser power or decreasing scanning speed) tends to increase in the hardness [88].

The addition of SiC particles increases the tensile strength and strain as shown in Fig. 8 [93]. Dislocations at the cell boundaries prevent dislocation motions and allow more dislocations to be stored [93]. Si and Mg₂Si particles formed at the cell boundaries by interfacial reactions prevent dislocation motions through the Orowan mechanism [102], thus contributing to the further development of mechanical properties [97]. Nanosized hard Al₄C₃ precipitates also increase tensile properties. Grain refinement is another factor that increases strength and strain simultaneously. However, it has also been observed that SiC reinforcement may have a negative effect on the tensile stress. The decrease in tensile strength is attributed to the low density as a result of the pores in the composite with the addition of SiC. It has been emphasized that the pores in the structure can lead to nucleation of microcracks, which may cause premature fracture [88].

Figure 8

Effect of addition of SiC particle on the tensile stress and strain (S₀ is unreinforced and, S₂ is 2 wt.% nano-SiC-reinforced) [93]

When the fractographs of SiC-reinforced composites were examined, it was observed that the fracture mode was brittle fracture as shown in Fig. 9 [91]. The bonding between adjacent tracks and adjacent layers are metallurgically weak zones [103]; thus, the fracture started in adjacent tracks and adjacent layers. Under tensile loading, cracks start at the overlapping layer boundaries in parts produced by the LPBF method, which leads to premature fracture [104, 105]. The high laser energy density contributes to high stress and strain by increasing the bonding between adjacent tracks and adjacent layers [97]. Moreover, reducing the SiC particle size in the matrix provides an increase in the geometrically necessary dislocation density, which increases the strength by impeding dislocation motions [91]. As the particle size decreases, the load transfer of the matrix to the reinforcing increases, which leads to the initiation of deformation under higher load [106].

Figure 9

Fracture surface of LPBFed SiC-reinforced composites with different particle size: a–c 300 mesh, d–f 600 mesh and, g–i 1200 mesh [91]

From the literature survey, it was found that SiC particles are applicable reinforcement to increase the mechanical properties of the Al-based composites produced by LPBF method. The homogeneous dispersion of the SiC reinforcement and the matrix material significantly affects the microstructure, porosity and mechanical properties. SiC fraction, size, morphology and LPBF process parameters appear as other factors affecting part quality. The summary of the literature review is shown in Table 3.

Table 3 Summary of the literature review for SiC-reinforced Al-based matrix produced by LPBF

Reinforced by TiC particles

TiC ceramic particles are frequently used as a reinforcement elements in Al-based MMCs due to their superior properties such as high elastic modulus, high hardness and good wettability [107]. The basic physical properties of TiC are listed in Table 4 [108]. Studies show that TiC particles are mostly nanosized, because the refined reinforcing particles effectively improve the mechanical properties of the composite [109, 110]. Composites produced with nanoparticles are known as nanocomposites, and studies on the production of nanocomposites with LPBF have increased significantly in recent years.

Table 4 Basic physical properties of TiC [108]

Microstructural characterization

The reinforcement size, shape, fraction and LPBF process parameters have different effects on the microstructure in Al-based MMCs with nano- and microsized TiC. It was determined that the microstructure became coarser with the increase in the fraction of reinforcing particles [111]. The addition of TiC to the Al matrix tended to increase the viscosity of the melt, making the flow of the melt difficult and reducing its rheological performance [112]. In addition, the high laser energy intensity caused the formation of high temperature gradients in the molten pool. The temperature gradients in the molten pool increase the surface tension gradients and resultant Marangoni convection [111]. Therefore, the undercooling degree of molten pool decreases and causes a low nucleation rate and hence coarsens the microstructure [113].

It was observed that the grain size decreased with the addition of TiC [19]. The thermal conductivities of AlSi10Mg and TiC particles are 103 W/m/K and 330 W/m/K, respectively [114, 115]. With the addition of TiC particles, the thermal conductivity of the composite increases, leading to a high solidification rate and thus refined grains. In addition to thermal conductivity, the laser absorptivity of the pure AlSi10Mg powder and TiC particles is 0.09 and 0.82, respectively [116, 117]. The increase in laser absorptivity increases the temperature gradients in the molten pool, resulting in a more homogenous distribution of TiC particles [118,119,120].

Besides reinforcing particle properties, LPBF process parameters have a significant impact on the microstructure. In order to examine the effects of laser power and scanning speed on the microstructure, microstructure characterizations were made by producing parts with different laser energy densities (η), which was defined by [121]:

$$n = \frac{P}{v}$$

(3)

where P is the laser power and v is the scanning speed.

Figure 10 shows the microstructures of TiC/AlSi10Mg nanocomposite with different η [122]. At the low η it was observed that, TiC particles agglomerates in the matrix. Increasing the η provided a more homogeneous distribution of TiC particles and almost no agglomeration [123]. In addition, the average size of TiC particles increased with the increase of η. Especially at very high η, the TiC particles lost their favorable initial morphology [90].

Figure 10

Microstructure of LPBFed TiC/AlSi10Mg nanocomposites at various laser energy density: a 250 J/m, b 500 J/m, c 700 J/m and d 1000 J/m [122]

The melt pool temperature, which improves significantly when high laser energy density is applied, leads to a decrease in viscosity and consequently intensification of the Marangoni flow [124]. With the decrease in viscosity, the rearrangement rate of nanosized TiC particles increases, thus avoiding agglomeration. At low laser energy density, the solidification rate of the molten pool increases. Grain growth of TiC particles is inhibited by the rapid solidification rate and the initial morphology is preserved [122, 125].

Densification behavior

In the LPBF method, the densification behavior is related to the melt pool viscosity, wettability and liquid/solid rheological properties. The dynamic viscosity of melt pool containing Al–Si–Mg liquid can be defined by [43]:

$$\mu = \mu_{0} \left( {1 - \frac{{1 - \varphi_{1} }}{{\varphi_{{\text{m}}} }}} \right)$$

(4)

where µ₀ is the base viscosity that includes temperature terms, ${\varphi }_{1}$ is the volume fraction of liquid phase and ${\varphi }_{m}$ is a critical volume fraction of solids above which the mixture has essentially infinite viscosity.

The densification level is controlled by the base viscosity, which is directly related to the temperature [126]. The base viscosity decreases with increasing temperature and leads to better wettability, thereby improving the densification level. At low laser energy density, viscosity increases due to the low temperature in the molten pool. The high viscosity results in poor wettability characteristic and consequently low densification level [127, 128]. Meanwhile, the addition of TiC increases the densification level due to the increase in laser absorption, while it decreases the densification level due to its tendency to increase the viscosity. To summarize, limited wettability and high melt viscosity are the main factors to determine the densification behavior of TiC-reinforced composites.

LPBF process parameters affecting the melt pool temperature, especially laser power and scanning speed, have significant effects on the densification behavior. At low laser energy density, irregular-shaped interlayer porosities are formed in the composite structure. With the increase in the laser energy density, it is revealed that the porosities in the composite get smaller and almost no porosity occurs at a certain level (Fig. 11) [122]. However, it has also been reported that there are decreases in the density of the composite at excessive laser energy densities. Excessive laser energy density may cause the turbulence in the molten pool or evaporation, resulting in a decrease in density [119].

Figure 11

Microstructures of SLM-processed TiC/AlSi10Mg nanocomposites at different laser energy density, a 250 J/m, b 500 J/m, c 700 J/m and d 1000 J/m [122]

Besides the process parameters, HIP as a post-process also has important effects on the densification characteristics. With the HIP process, it was determined that the porosity of pure AlSi10Mg decreased from 1.32% to 0.38%, and the porosity of TiC-reinforced AlSi10Mg composite decreased from 1.82% to 0.42%.

Hardness and tensile properties

The hardness of the composites increases with the addition of TiC particles as a reinforcement. The hardness of TiC/AlSi10Mg nanocomposite can reach 185 HV and this value is 21% higher than the hardness of unreinforced AlSi10Mg [5, 129]. In a study using Al-15Si as the matrix material, it was determined that the hardness decreased slightly with the addition of 1% and 2.5% TiC, while the hardness increased with the addition of 10% TiC (Fig. 12) [111]. This increase can be attributed to the presence of high fraction of hard TiC particles and primary Si precipitation. TiC particle size is another factor that affects hardness. It was observed that the hardness of the composite increased with the decrease in the TiC particle size [119]. The high hardness largely depends on the uniform and homogeneous distribution of TiC particles in the matrix. With the reduction in the TiC particle size, TiC particles in the powder mixture are mixed more homogeneously with the matrix powders, leading to the production of parts at almost full density, a more uniform TiC distribution and therefore an increase in hardness. In addition, uniformly distributed TiC particles improve the strength of the interfacial bonding and thus contribute to the increase in hardness.

Figure 12

Effect of different TiC contents on microhardness of TiC/Al-15Si composite [111]

The variation in microhardness at different laser energy densities is presented in Fig. 13 [122]. The increase in laser energy density contributes to the improvement of hardness. The increase in hardness with the increase in the laser energy density can be attributed to the uniform distribution of TiC particles and the improvement of part density [128]. However, further increase in laser energy density (> 800 J/m) has a negative effect on hardness [112]. The hardness decreases as a result of grain growth, more defects in microstructure and reduction in the density at excessive laser energy density.

Figure 13

Variation in microhardness at different laser energy densities [122]

With the addition of TiC, the tensile strength of composite increased from 400 to 482 MPa, while the elongation was consistent with the unreinforced part [129]. The increase in tensile strength without a decrease in ductility can be attributed to the increase in dislocation density at reinforcement/matrix interfaces by causing dislocation pileup of TiC particles [130, 131]. In addition, TiC particles act as a barrier against dislocation movements, increasing the tensile strength. The weight/volume ratio of TiC particles in the composite also has important influences on the tensile properties. It was determined that both tensile stress and ductility increased with the addition of 1% and 2.5% TiC to the Al-15Si matrix, while both values decreased with the addition of 10% TiC. The reason for the reduction in the tensile stress at the addition of 10% TiC is the rapid precipitation of Si atoms trapped in the Al matrix onto the existing eutectic network Si [111]. In the case of using 1 µm TiC particles, tensile stress increased from 410 to 470 MPa, while elongation decreased from 8.9% to 4%. However, when 30 µm TiC particles were used, tensile stress and elongation were simultaneously reduced compared to the unreinforced material [119]. The reduction in tensile properties can be attributed to the balling effect, which tends to break up into several spherical agglomerates to reach equilibrium due to the high viscosity of the liquid in the molten pool [132].

Compared to the unreinforced part, the tensile stress of the TiC-reinforced composite increased from 400 to 452 MPa at an energy density of 160 J/mm and increased to 486 MPa at an energy density of 240 J/mm. Ductility values decreased from 10.7 to 9.8% at 160 J/mm and increased to 10.9% at 240 J/mm. The homogeneous distribution of TiC particles at an energy density of 240 J/mm and the ring structures formed in the microstructure increased both tensile stress and ductility [125]. The increase in part density with the increase in the laser energy density is another reason for the increase in tensile properties.

In Fig. 14, the tensile properties of TiC-reinforced and unreinforced parts before and after the HIP process are presented [19]. It is seen that the tensile stress increases and elongation decreases with the addition of TiC. The reason for the decrease in elongation was attributed to the increase in porosity and dislocation density with the addition of TiC [133]. After the HIP process, the tensile strength of the reinforced and unreinforced parts decreased, but the elongation increased drastically. The work hardening effect on reinforced and unreinforced parts has been vanished by the HIP process, and the material has been homogenized. In addition, the elimination of dislocation walls is another reason for the increase in ductility [19]. To sum up, the HIP process is an important post-process that increases the ductility of reinforced and unreinforced parts produced with LPBF.

Figure 14

Tensile properties of pure AlSi10Mg and nTiC/AlSi10Mg before and after HIP process, a engineering stress–strain curves, b comparison of UTS, YS and elongation [19]

From the literature survey, it was found that TiC particles are another applicable reinforcement to increase the mechanical properties of the Al-based composites produced by LPBF method. Homogeneous mixture of TiC reinforcement and matrix material, TiC fraction and size, as well as LPBF process parameters are significant factors affecting part quality. As a post-process, HIP has significant effects on the mechanical properties of the composite. With the HIP application, the internal pores have decreased, and thus, the ductility has increased significantly. The summary of the literature review is shown in Table 5.

Table 5 Summary of the literature review for TiC-reinforced Al-based matrix produced by LPBF

Reinforced other ceramic particles

In addition to SiC and TiC reinforcements, research works are conducted in which TiB₂, TiN and Al₂O₃ ceramics are used as reinforcement elements. These ceramics are used extensively in Al-based composites due to their superior mechanical properties. In Fig. 15, SEM images of composite feedstocks prepared with micro- and nanosized TiB₂ reinforcement [134] and microsized Al₂O₃ reinforcement are presented [135]. TiB₂/AlSi10Mg feedstock was prepared with ball milling in 48 h without any grinding medium, and Al₂O₃/AlSi10Mg feedstock was prepared in 10:1 ball-to-powder weight ratio with 70 rpm rotation speed in 4 h. It was determined that dispersion state and sphericities of the mixed powders do not change markedly in both composite feedstock.

Figure 15

Scanning electron micrographs of powders: a AlSi10Mg, b 10 wt.% TiB₂(µm)/AlSi10Mg, c 1 wt.% TiB₂(nm)/AlSi10Mg [134] and d AlSi10Mg, e Al₂O₃, f Al₂O₃/AlSi10Mg [135]

Microstructural characterization

In Fig. 16 a, the microstructural development of TiN/AlSi10Mg composite is shown schematically [136]. In the LPBF process, Marangoni flow is formed by a high temperature gradient from the edge to the center of the molten pool [71]. Due to the high melting temperature, the unmelted ceramic reinforcements flow and rearrange with the Marangoni flow. As a result, most of the reinforcing elements are pushed and distributed along the grain boundaries during the solidification process. Only limited number of particles act as heterogeneous nucleation sites [137]. In Fig. 16 b, the behavior of Al₂O₃ particles in the molten pool is shown schematically [138]. The distribution state of the Al₂O₃ particles in the molten pool mainly depends on the interaction between the reinforcement particles and the advancing solid–liquid interface.

Figure 16

a Schematic representation of behaviors of Al₂O₃ reinforcing particles in melt pools [136], b schematic representation of the microstructural development for the TiN/AlSi10Mg composites during LPBF process [138]

The microstructures of TiN/AlSi10Mg composite and AlSi10Mg are presented in Fig. 17 [139]. It is seen that the grain size of the composite is more refined than the unreinforced AlSi10Mg part. Since the interaction between the laser beam and the powders is extremely short in the LPBF process, the unmelted ceramic reinforcements and the molten matrix are simultaneously present in the molten pool. Solid-state reinforcing elements act as heterogeneous nucleation sites, improving the nucleation rate and therefore causing grain refinement [140]. Gao et al. [71] determined that the grain size depends on the laser process parameters. They stated that the grain size increased from 0.284 to 0.388 µm as a result of the increase in thermal gradients and cooling rate with the decrease in the scanning speed from 600 to 200 mm/s.

Figure 17

SEM images of: a TiN/AlSi10Mg nanocomposite and b AlSi10Mg microstructure [139]

In Fig. 18, EBSD inverse pole figures of composites produced at different TiB₂ reinforcement content are presented [141]. The grain size of the composite became more refined as the reinforcement content increased. TiB₂ particles act as grain growth inhibitor, thus limiting the growth of grains along the heat flux direction, similar to other ceramic particles [142]. The grain sizes of the composites produced with 1% and 5% reinforcement content were measured as 6.32 ± 0.07 µm and 1.55 ± 0.14 µm, respectively. In summary, the fact that composites produced with ceramic reinforcements have more refined grains is attributed to a combination of increased nucleation sites, limitations on grain growth and rapid solidification rate during LPBF.

Figure 18

EBSD inverse pole figure maps of TiB₂/AlSi10Mg composites with different reinforcement fractions: a 1 wt.%, b 2 wt.% and c 5 wt.% TiB₂ [141]

Densification behavior

The densification behavior of ceramic-reinforced composites produced with LPBF is significantly affected by the physical and chemical properties of the feedstock powders as well as the LPBF process parameters. Micro-CT images of TiN/AlSi10Mg composite produced at different reinforcement weight ratios and porosity values as a function of TiN content are presented in Fig. 19 [138]. It is seen that the porosity decreases with the increase in the reinforcement content from 0 to 4%, but the porosity increases with its further increase. With the addition of TiN as a reinforcing element, the reduction in the laser reflectivity increases the melt pool temperature and decreases the melt viscosity [139]. In addition, nanosized reinforcements provide lower melt pool viscosity than microsized reinforcements [143]. With the decrease in viscosity, sufficient melt flow is ensured, and the melt fills the voids more easily, thus increasing the composite density. Moreover, the higher temperature in the molten pool provides better wettability [144]. In this way, a stronger bonding is formed between the reinforcement and the matrix, and the formation of pores and cracks at the interface is reduced.

Figure 19

a Micro-CT images of porosity distribution of different reinforcement weight ratios TiN/AlSi10Mg composites, b the porosity as a function of TiN content [138]

However, if the reinforcement content exceeds a certain threshold, it negatively affects the composite density. The use of a high content of reinforcement makes it difficult to prepare a homogeneous feedstock. Moreover, reinforcement particles, especially at the nanoscale, tend to agglomerate due to their high surface energy and specific surface area. The agglomeration of the unmelted reinforcements in the molten pool increase the porosity [138]. Solid-state reinforcements in the molten pool also increase the melt viscosity, causing the formation of more defects and pores.

Micro-CT images of Al2O3/AlSi10Mg composite produced at different volumetric energy densities are presented in Fig. 20 [144]. In general, it has been stated that increasing the laser power and decreasing the scanning speed increase the composite density due to elevating the melt pool temperature and decreasing the melt viscosity. At low laser power and high scanning speeds, high viscosity causes poor wettability, resulting in the formation of pores and defects in the molten pool [141]. Research works in the literature have suggested different optimum process parameters for the production of nearly fully dense composites with different reinforcement. Xi et al. [145] produced 1 wt%, 2 wt% and 5 wt% composites with TiB₂ reinforcement and stated that the maximum relative density was obtained at 1800 mm/s scanning speed. Li et al. [146] investigated the densities of TiB₂/AlSi10Mg composite produced in different parameter sets and obtained higher relative density at 400 W laser power and 2000 mm/s scanning speed. Gao et al. [71] stated that the relative density decreased with increasing scanning speed for nano-TiN/AlSi1OMg composite, and the highest relative density was obtained at the scanning speed of 200 mm/s. Han et al. [70] determined that the optimum laser energy density for Al/Al₂O₃ composite was 317.5 J/mm³, and the relative density decreased above and below this critical threshold. As a result, it is revealed that the LPBF process parameters have significant effects on the composite density, and different process parameters should be used depending on the reinforcement for the production of nearly fully dense composites.

Figure 20

Micro-CT images of porosity distribution of 2 wt.% Al₂O₃/AlSi10Mg composite at different volumetric energy densities: a 23.33 J/mm³, b 109.38 J/mm³ [144]

Hardness and tensile properties

In Fig. 21 (a), tensile stress–strain curves of TiB₂/AlSi10Mg composite produced with different reinforcement contents are presented [141] and Fig. 21 b shows the tensile stress–strain curves of pure Al and composite samples under different scanning speed [70]. With the use of ceramic particles as reinforcement, the hardness and tensile properties of composites improve. The yield strengths of the pure Al-12Si alloy and the reinforced with 2% TiB₂ composite were measured as 211 ± 4 and 225 ± 4 MPa, respectively [147]. The hardness and tensile properties of the ceramic-reinforced composite materials are largely dependent on the reinforcement size and content, homogeneous dispersion of the reinforcements and the matrix and interfacial bonding between matrix and reinforcement.

Figure 21

Tensile stress–strain curves of: a TiB₂/AlSi10Mg composite different reinforcement fractions [141], b pure Al and Al₂O₃/Al composites [70]

There are different strengthening mechanisms in ceramic-reinforced composites, including grain refinement effect, dislocation density strengthening, Orowan strengthening and load-bearing strengthening through strong interfacial bonding [73, 94, 148]. However, especially if the reinforcement content exceeds a certain threshold, it can cause negative effects on the hardness and tensile properties, since it causes the formation of pores by increasing the melt viscosity. The ultimate tensile stresses of the composites produced with 2% and 5% TiB₂ content were determined as 444 MPa and 422 MPa, respectively [141] and the yield strength of the composites produced with 2%, 4% and 6% nano-TiN content was measured as 295.9 ± 4.6, 315.4 ± 5.2 and 261.6 ± 3.5 MPa, respectively [138].

As a result of the literature review, it has been revealed that TiB₂, TiN and Al₂O₃ ceramic reinforcements can be used in the production of nearly fully dense composites by LPBF method. In order for the composite quality to be at the desired values, it is necessary to mix the reinforcement and matrix powders homogeneously, and to use the appropriate reinforcement content and laser process parameters. Too long or too short the interaction of the laser beam and the powder negatively affects the rheology of the molten pool and causes the formation of pores in the molten pool. The summary of the literature review for the TiB₂, TiN and Al₂O₃ reinforcements is shown in Table 6.

Table 6 Summary of the literature review for other ceramic particle-reinforced Al-based matrix produced by LPBF

Conclusions and future research

This study presents a comprehensive review of the production of ceramic particle-reinforced Al-based composites by LPBF method. The manufactured composites were characterized in terms of microstructure, densification behavior and mechanical properties (hardness and tensile properties) which have the most significant effects in the widespread application of MMCs. The effects of reinforcing particle properties such as type, morphology, size and content as well as LPBF process parameters on the composite properties were evaluated.

From the literature survey, it is revealed that the LPBF is a promising method for producing high-quality, low-cost, repeatable and nearly fully dense ceramic particle-reinforced Al-based MMCs. In addition to the LPBF method, composite production processes are investigated by using various reinforcement elements and matrix materials in different AM techniques such as direct ink writing (DIW) [149], stereolithography (SLA) [150] and digital light processing (DLP) [151]. Particularly, carbon fiber-reinforced ceramic and metal matrix composites are intensively studied [152, 153]. These studies show that these AM techniques are promising methods for producing complex-shaped and high-quality ceramic and metal matrix composites at a lower cost than conventional methods.

The flexibility that LPBF offers in the use of different feedstocks indicates that it would be increasingly used in MMC production. However, it is necessary to pay attention to numerous issues in composite production in LPBF, which is a complex process. The selection of reinforcement is the primary issue as it will affect the chemical reactions and the evolution of the internal defects. The chemical reactions that occur between the matrix and the reinforcement cause the formation of new phases at the interface. In order to evaluate the microstructure and internal defects of ceramic-reinforced composite, it is necessary to analyze the chemical reactions and reaction products occurring at the reinforcement–matrix interface. Reinforcements such as TiB₂, CNT and TiC, which react less with the Al matrix, can be considered primarily in the production of Al-based MMCs, as they reduce the formation of undesirable reaction products at the interface. Moreover, different reinforcements exhibit different wetting, solidification and agglomeration behavior during the LPBF process. The agglomeration tendency of nanosized reinforcements have detrimental effects on composite quality. To sum up, find the right reinforcement is the key factor for producing high-quality composites.

One of the other significant issue is to ensure the homogeneous distribution of the particles with a suitable mixing method. Ball milling and regular mixing methods have attracted great attention in recent years due to their practicality, time and cost efficiency, and applicability to a wide variety of materials. However, in both methods, long-term collisions of the powders have negative effects on the powder surface quality. Repeated deformation, fracture and cold welding inevitably degrade the sphericity of the initial powders, which adversely affects the powder flowability and causes a decrease in composite quality. Another undesirable effect of these mixing methods is the inability to reduce the agglomeration effects, especially of nanosized reinforcements. Therefore, the development of a novel and advanced mixing technique which cannot affect the powders initial morphology and surface quality and can provide a homogeneous mixture is considered very important in the production of high-performance composites with LPBF.

A comprehensive understanding of the strengthening mechanisms of composites produced with LPBF would be important in producing parts with high mechanical properties. The addition of ceramic reinforcements increases the mechanical properties of the composites with the grain refinement effect, dislocation density strengthening, Orowan strengthening and load-bearing strengthening through strong interfacial bonding. The most significant factor determining the tensile properties is the bonding between adjacent tracks and adjacent layers. Under tensile loading, cracks start from adjacent tracks and adjacent layers. Better tensile properties can be obtained with stronger bonding. Improving the mechanical properties of composites depends on the elimination of internal defects like porosity and residual stresses. As a post-treatment, HIP and heat treatment would be an effective process to reduce the porosity, release residual stresses and increase the mechanical properties of composites.

When the effects on the microstructure, density and tensile properties of ceramic reinforcements are examined, it is seen that the most important factor is the rheological behavior of the molten pool. Proper selection of reinforcing particle properties and LPBF process parameters which have the effect on rheological behavior of the molten pool is necessary to achieve high-quality composites. As a result of the findings obtained from the literature survey, it is revealed that there is a gap on process optimization, modeling simulation and post-treatment operations such as heat treatments and HIP. In particular, the findings to be obtained by modeling the molten pool can make significant contributions to a better understanding of microstructure formation, solidification behavior and internal defects. By understanding the underlying mechanisms with reliable predictive models, the LPBF process can be optimized.

Declarations

Conflict of interests

The authors declare that they have no conflict of interest.