Effect of Preparation Process on Microstructure and Mechanical Properties of Metal Hollow Sphere Composites (MHSCs)

Abstract

In order to study the effect of the preparation process on the microstructure and mechanical properties of metal hollow sphere composites (MHSCs), they were fabricated using different preparation processes (different holding times) by a pressure casting method. The physical parameters of the MHSCs were measured using a direct measurement method, and the phases and microstructure of the MHSCs were observed using an X-ray diffractometer (XRD), scanning electron microscope (SEM), energy dispersive spectrometer (EDS), and transmission electron microscope (TEM). Moreover, the mechanical properties of MHSCs were analyzed by a Zwick/Roell Z2.5 microhardness tester and a universal testing machine. The microscopic morphology and composition analysis results showed that the microstructures of MHSCs varied by holding time, while the mechanical properties test results implied that the microhardness of the transition layer and the matrix increased with the extension of the holding time. However, the MHSCs held for 10 min exhibited the best compression performance. These observations indicated that the compressive properties of MHSCs were mainly dependent on the transition layer and the matrix, excluding the influence of metal hollow spheres, and that the transition layer had a significant impact on the compression performance of MHSCs.

Introduction

As a kind of structural and functional material, metal hollow spheres (MHSs) composites (MHSCs) have been widely studied.^1,2,3,4,5 By introducing a large amount of air, the MHSs strengthens the MHSCs, resulting in the low density,^3,4 high specific strength,^5,6,7,8 and high energy absorption^{9,10,11,12,13} of the MHSCs. MHSCs also possess the functional properties of traditional porous materials, such as heat insulation,¹⁴ acoustic insulation and absorption,^15,16 nuclear shielding,^17,18,19 and ballistic performance,^20,21 implying their potential applications in automobiles, aviation, aerospace, and navigation,¹ for example, as energy absorption boxes for cars, helicopter armor, insulation tiles for space probes, and protective materials for nuclear power devices. In addition, the material and matrix of MHSs can be selected according to specific requirements, and the structure of MHSCs can also be designed.²² At present, aluminum alloys,^{23,24,25,26,27} magnesium alloys,²⁸ iron, titanium, and zinc^{29,30,31,32,33} are commonly used as matrix materials. The materials of MHSs with functional characteristics include stainless steel,³⁴ stainless steel/ceramics,³⁵ and Ti-6Al-4V³⁶ and Ni-Ti³⁷ alloys.

The existing research results indicate that the mechanical properties of materials are affected by their microstructures.^{38,39,40,41,42} In order to further expand the application scope of MHSCs, the relationship between the microstructure and the mechanical properties of MHSCs has been studied.^{7,8,9,10,11,12,13} The results showed that the amount and composition of phases present in the microstructure of MHSCs was directly related to the casting temperature and cooling rate. The MHSCs prepared by low-temperature casting and rapid cooling had excellent mechanical properties due to their refined phases. However, further research is needed to understand the influence of process parameters on the mechanical properties of MHSCs, especially to achieve the purpose of precisely controlling the microstructure and mechanical properties through process parameters, which could not be achieved solely by changing the casting temperature and cooling rate. Therefore, if a novel casting process could be proposed to precisely control the microstructure of the MHSCs by changing the process parameters, this would be of important research significance for the application of MHSCs. In addition, the influence of the interface (the transition layer between MHS and matrix) and the matrix on the compressive properties of MHSCs has not been clearly determined. A deeper insight into the influence of the interface and matrix on the mechanical properties of MHSCs will be more conducive to the design of MHSCs with good comprehensive mechanical properties.

Therefore, this study has redesigned the casting process of MHSCs and introduced a new process parameter (holding time). The interface structure and the matrix microstructure in MHSCs can be controlled by controlling the holding time so that MHSCs with different interface structures and matrix microstructures can be fabricated. The physical properties and microstructure of the MHSCs have been characterized. Moreover, the compressive properties of the MHSCs and the A356 alloy fabricated using different holding times have been analyzed by compression experiments, and the influence of the interface and matrix on the mechanical properties of MHSCs has been revealed.

Materials and Methods

Materials

The A356 aluminum alloy was selected as the matrix material, and the MHSs were made of 316L stainless steel. Table I shows the physical parameters and chemical composition of the MHSs.

Table I The physical parameters and chemical composition of the MHSs

Figure 1 shows the initial images of the component materials, in which pores were uniformly distributed on the spherical wall, while the A356 aluminum alloy was composed of a matrix phase and a precipitation phase. In addition, the spherical wall had uniform thickness and a round appearance.

Fig. 1

the initial images of components materials: (a) MHSs, (b) metallographic diagram of spherical wall, (c) SEM of spherical wall, (d) SEM of spherical wall fracture, (e) SEM of the whole MHS, (f) SEM of A356 aluminum alloy.

Fabrication of the MHSCs

The MHSCs were fabricated using a special mold in a resistance furnace. The mold is shown in Fig. 2a, and the assembly diagram of the MHSs and the A356 alloy is shown in Fig. 2b. They were put together into the resistance furnace for casting. The whole casting process is as follows. Firstly, the resistance furnace was heated to 710°C and maintained at this temperature for 45 min. Secondly, pressure was applied to the mold, which was then moved slowly downward until the aluminum liquid flowed out of the riser. Finally, the MHSs and the A356 alloy were taken out of the furnace after it had been maintained at 710°C and cooled to room temperature by air. The MHSCs were prepared using holding times of 0 min, 10 min, and 20 min.

Fig. 2

Fabrication process of the MHSCs: (a) graphite mold, (b) assembly diagram.

Characterization

Physical Properties of MHSCs

The physical parameters of the MHSCs were measured and calculated, including their density (${\rho }_{t}$), the volume fraction of the MHSs (${\varnothing }_{h}$), the volume fraction of the matrix (${\varnothing }_{M}$), the porosity of the matrix (${\varnothing }_{v}$), and the porosity of the MHSCs (${\varnothing }_{n}$). ${\rho }_{t}$ was calculated by measuring the mass ($m$) and volume ($V$) of the test sample, and V was obtained by measuring the length, width, and height of the regular block.

${\varnothing }_{M}$ and ${\varnothing }_{v}$ were calculated by Eq. 1, and ${\varnothing }_{v}$ and ${\varnothing }_{n}$ were, respectively, calculated by Eqs. 2 and 3.

$$ \emptyset_{i} = \frac{{m_{i} }}{{\rho_{i} V}} $$

(1)

$$ \emptyset_{v} = 1 - \left( {\sum \emptyset_{i} } \right) $$

(2)

$$ \emptyset_{n} = \emptyset_{v} + \emptyset_{h} - \frac{{m_{h} }}{{\rho_{S} }} $$

(3)

where ${\rho }_{S}$ is the density of 316L stainless steel, and ${m}_{h}$ is the mass of MHSs in the MHSCs.⁴³

Microstructure Analysis of MHSCs

The phase composition of the MHSCs fabricated under different holding times was analyzed using X-ray diffraction (XRD; X'Pert PRO). The microstructure of the MHSCs was studied using scanning electron microscopy (SU-70), and the transition layer between the MHS and the matrix was investigated by transmission electron microscopy (TEM; Talos F00X; FEI).

Mechanical Properties of MHSCs

The microhardness of the spherical wall, transition layer, and matrix was measured with a Zwick/Roell Z2.5 microhardness tester. Each position was measured ten times, and the average was used as the microhardness value for each position. The load and holding times of the microhardness tester in the measurement process were 100 N and 15 s, respectively. An Instron5500R was used as the universal testing machine for investigating the compressive properties of the MHSCs and the A356 aluminum alloy. Each group of tested samples was compressed ten times, and the stress–strain curve of the MHSCs was plotted by taking the average value.

Results and Discussion

Physical Properties of the MHSCs

Figure 3 shows an image of the MHSCs fabricated under a holding time of 10 min and it can be seen that the MHSs within the same cross-section were unevenly distributed. In addition, due to the difference in the cutting positions of the MHS at the cross-section, the pore size displayed at the cross-section of the MHS was different. At the same time, some MHSs were filled with A356 aluminum alloy.

Fig. 3

Image of the MHSCs fabricated with a holding time of 10 min.

The physical properties of the MHSCs are displayed in Table II, and it can be seen that, when the volume fraction of MHSs was 48.50%, the density of the MHSCs was between 1.99 g/cm³ and 2.02 g/cm³. In addition, the porosity of the matrix was between 5.08% and 6.20%, while that of MHSCs was between 44.08% and 45.02%.

Table II The physical parameters of the MHSCs

Microstructural Analysis of the MHSCs

Figure 4 shows the XRD patterns of the MHSCs. The MHSCs fabricated under the holding time of 0 min were mainly composed of γ-Fe, Al, Si, Fe₂Al₄Si, and Al₈Si₆Mg₃Fe phases, while those fabricated under the holding time of 10 min were mainly composed of γ-Fe, Al, Si, Fe₂Al₇Si, Fe₂Al₄Si, and Al₈Si₆Mg₃Fe phases, and those fabricated under the holding time of 20 min were mainly composed of γ-Fe, Al, Si, Fe₂₅Al₆₀Si₁₅, Fe₂Al₇Si, Fe₂Al₄Si, and Al₈Si₆Mg₃Fe phases.

Fig. 4

XRD patterns of the MHSCs with different holding times.

Figure 5 shows the back-scattering image of the “original interface” between the wall of the MHS and the matrix under different holding times.

Fig. 5

The back-scattering image of the “original interface” between the MHS and the matrix with different holding times: (a) 0 min, (b) 10 min, and (c) 20 min.

It can be seen from Fig. 5a that no transition layer was formed between the wall and the matrix at 0 min, but only physical bonding. However, when the holding times were 10 min and 20 min, a continuous transition layer was formed between the wall and matrix, as shown in Fig. 5b and c, respectively. Also, the structure of the transition layer shifted from single-layer to multi-layer with the extension of the holding time. When the holding time was 10 min, the transition layer was composed of continuous, plate-like phases, with an average thickness of about 25 μm. When the holding time was 20 min, the average thickness of the transition layer increased significantly to 35 μm.

In addition, the shape of the phase in the matrix varied with the extension of holding time. When the holding times were 0 min and 10 min, the phases in the matrix were mainly “needle stick”-shaped and mainly “fish bone”-shaped when the holding time was 20 min. In the compression process, the matrix was very likely to coexist with these “fish bones” to generate stress concentration, which would degrade the compressive properties of the matrix material^24,25 and reduce the compressive properties of the MHSCs.

Figure 6 shows the distribution of alloying elements at the “original interface” in the MHSCs under different holding times, and the energy dispersive spectrometry (EDS) results of the white line from the matrix to the wall.

Fig. 6

The distribution of alloying elements at the “original interface” and EDS results of white lines from the matrix to the wall with different holding times: (a) 0 min, (b) 10 min, and (c) 20 min.

As can be seen from Fig. 6a, when the holding time was 0 min, the alloy elements did not diffuse and pass through the “original interface” except for Mg. As can be seen from Fig. 6b, when the holding time was 10 min, the content of Al gradually decreased from the matrix to the wall, and Si existed in the matrix and spherical wall in the form of a precipitated phase and the matrix. However, Si was evenly distributed in the transition layer. Mg diffused towards the wall through the “original interface” and then was evenly distributed in the wall. In the matrix and the transition layer, Mg mainly existed in the form of a precipitated phase. The distribution form of Fe was just opposite to that of Al, which gradually decreased in the direction of the wall towards the matrix. At the same time, there were precipitates containing Fe in the matrix. After passing through the “original interface”, Cr and Ni were enriched in the transition layer, while Mn and Mo uniformly diffused into the transition layer.

As can be seen from Fig. 6c, when the holding time was 20 min, the content of Fe, Cr, Al, and Ni in the spherical wall gradually decreased in a three-layer ladder shape in the direction from the spherical wall to the transition layer, while the transition layer presented an obvious two-layer structure. Si existed in the matrix and spherical wall in the form of a precipitated phase and the matrix, evenly distributed in the transition layer. The content of Si in the transition layer decreased in a two-layer ladder from the matrix to the transition layer. Mn and Mo uniformly diffused into the transition layer through the “original interface”.

Figure 7 shows the transition layer between the MHS and the matrix under different holding times, and Table III shows the EDS results of dots in Fig. 7. It can be seen from the EDS results in Fig. 7a and Table III that, when holding time was 0 min, the “original interface” was mainly composed of Fe, Al, Si, Cr, Mn, Mo, Mg, and Ni, with no reaction between them.

Fig. 7

The transition layer between the MHS and the matrix with different holding times: (a) 0 min, (b) 10 min, and (c) 20 min (Color figure online).

Table III The EDS results of the dots in Fig. 7

As can be seen from the EDS results of the dots in Fig. 7b, the wall near the transition layer contained Al and Si (as shown at Point D). However, Al and Si did not react with the alloying elements in the sphere wall to form new phases. The transition layer was a new phase composed of Al, Fe, and Si (as shown at Points B and C). In addition, alloy elements in the sphere wall were also found in the matrix (as shown in Point A).

According to the observation results and relevant research results,^2,3,4,11 the alloy elements in the matrix and the sphere wall diffused into each other when the holding time was 10 min, forming the transition layer.

Figure 8 shows the TEM results of the transition layer between the MHS and the matrix when the holding time was 10 min. Based on XRD, EDS, TEM, and relevant research results,^{2,3,4,11,44,45,46} the transition layer could be identified as the Fe₂Al₇Si phase (Color figure online).

Fig. 8

TEM bright-field image and diffraction pattern of the transition layer: (a) bright-field image, (b) diffraction pattern in the red box in (a) (Color figure online).

It can be seen from Fig. 7c that the transition layer was obviously divided into three layers (structural layers I, II, and III), which were, respectively, composed of a single phase, precipitated phase, and matrix phase. The red points in Fig. 7c were analyzed by EDS, and the results are listed in Table III.

It can be inferred from the EDS results that structural layer I was single phase (Fe₂Al₇Si phase) formed by Al, Si, and Fe (as shown at Point A). The structural layer II was the matrix phase (Fe₂Al₇Si phase) and the precipitation phase (b phase, as shown at Points B and C). The structural layer III was the matrix phase (b phase, as shown at Point D) and the precipitation phase (γ-Fe phase, as shown in point E). Similarly, Al and Si were also detected at Point F near the transition layer.

As can be seen from the observation results in Fig. 7c, Table III, and relevant research results,^2,11 Al and Si in the matrix continuously diffused into the spherical wall and formed the matrix phase (b phase) as the holding time was extended to 20 min. At the same time, the alloy elements in the sphere wall diffused to the transition layer, leading to the formation of new precipitates (b phase).

Figure 9 shows the TEM results of the b phase. According to XRD, EDS, TEM, and relevant research results,^{2,11,44,45,46} b phase could be identified as the Fe₂₅Al₆₀Si₁₅ phase.

Fig. 9

TEM bright-field image and diffraction pattern: (a) bright-field image, (b) diffraction pattern of the phase in the red box in (a) (Color figure online).

Mechanical Properties

Microhardness

Table IV shows the microhardness of the MHS, transition layer, and matrix in the MHSCs fabricated under different holding times. The microhardness of all three increased with the extension of the holding time. In addition, the microhardness of the transition layer was much higher than that of the MHS and the matrix. When the holding times were 10 min and 20 min, the microhardness of the transition layer was, respectively, 510.6 HV ± 10.3 HV and 538.5 HV ± 11.6 HV, with the latter significantly higher. This was because the transition layer was a single Fe₂Al₇Si phase when the holding time was 10 min, but was composed of Fe₂₅Al₆₀Si₁₅, Fe₂Al₇Si, and γ-Fe phases when the holding time was 20 min. Also, the microhardness of the Fe₂₅Al₆₀Si₁₅ phase was higher than that of the Fe₂Al₇Si phase,^24,25 indicating that the emergence of the Fe₂₅Al₆₀Si₁₅ phase improved the microhardness of the transition layer.

Table IV The microhardness of the MHS, transition layer, and matrix

In addition, as the holding time was prolonged, the content of Fe, Ni, Cr, Mo, and Mn in the transition layer increased, and the microhardness of the transition layer was further improved under the action of solution strengthening.^24,25

Compression Performance

Figure 10 shows the compressive stress–strain curves of MHSCs fabricated under different holding times (0 min, 10 min, and 20 min). It can be concluded from Fig. 10 and the relevant research result⁴³ that the MHSCs underwent shear deformation. The compression process of the MHSCs could be divided into three stages, namely the elastic stage, shear deformation stage, and densification stage.

Fig. 10

The compressive stress–strain curves of MHSCs with different holding times.

In addition, the MHSCs exhibited the best compressive property when the holding time was 10 min: the yield strength was 60.6 MPa ± 6.3 MPa, the average strength of the stress platform was 21.7 MPa ± 4.8 MPa, and the effective strain was 0.68 ± 0.06.

Since the MHSs used in the experiment were the same, only the influence of the transition layer and the matrix on the compression performance of MHSCs needed to be considered.

Figure 11 shows the compressive stress–strain curves of the A356 aluminum alloy fabricated by the same process as the MHSCs. As shown in Fig. 11, the compressive properties of the alloy samples prepared under holding times of 0 min and 10 min were similar, but higher than those of the alloy samples fabricated under a holding time of 20 min.

Fig. 11

The compressive stress–strain curves of the A356 alloy with different holding times.

Figure 12 shows the scanning images of the fracture surface and deformation diagrams of MHSCs in the deformation process. As can be seen from Fig. 12a, there were obvious shear bands on the surface of the compressed sample, which indicates that the MHSCs experienced shear deformation. In addition, only the MHSs located around the shear band deformed in the formation process of the shear band, though accounting for a minority.

Fig. 12

Images, fracture scanning, and deformation mechanism diagrams of MHSCs during the deformation process: (a) images, (b) fracture scanning, (c) deformation mechanism diagrams (Color figure online).

Figure 12b shows that the MHSs, transition layer, and matrix located on the shear bands had fractured and deformed in the compression deformation process, implying that the compressive deformation resistance of MHSCs mainly came from the deformation and fracture of the MHSs, transition layer, and matrix material on the shear band. Therefore, the more MHSs, transition layer, and matrix materials involved in deformation, the higher the deformation strength of the MHSs.

Based on the above observation results and research literature,⁴³ the compression deformation process of MHSCs could be divided into three stages, as shown in Fig. 12c.

Stage I: The MHSCs only underwent elastic deformation, whose state remained unchanged, and the MHSs, transition layer, and matrix jointly bore the compressive load. In this stage, the stress rapidly increased at the beginning of compression, as shown in the stress–strain curve.

Stage II: The MHSCs underwent shear deformation. At the end of the elastic deformation stage, due to the significantly lower compression performance of MHSs than the A356 Al alloy,³⁴ deformation preferentially occurred where the MHSs were densely distributed. However, due to the higher brittleness of the transition layer, the transition layer fractured or cracked, as shown in the red box at Fig. 12c. Then, under the action of the shear force, cracks rapidly propagated in the direction of the shear force, resulting in a crack band in the same direction as the shear force. The cracks mainly extended along the transition layer in the dense area of MHSs in the oblique box area due to the brittleness of the transition layer, and, in this process, the matrix material located on the expansion path suffered from deformation and fracture.

When the crack band penetrated the sample, the MHSs located on the crack band would deform in the direction of the shear force under the action of the shear force, as shown in the oblique red box in Fig. 12c. After the deformation of the MHSs, the MHSs fractured, resulting in the fracture of the sample along the shear band under the action of the shear force. At this time, the shear deformation was completed.

In the formation process of the shear bands, the resistance to fracture and deformation of MHSs, transition layers, and matrix materials on the shear band was equivalent to that of the formation of shear bands, which is also known as the compressive deformation strength of MHSCs.

Stage III: After the completion of shear deformation, as compression resumed, the fractured MHSCs were insufficient to form new shear bands, and were compressed as a whole with a rapid increase in compressive stress.

In summary, it can be concluded that MHSs, transition layers, and matrix materials were all involved in the entire compression deformation process of the MHSCs. The compression performance of the MHSCs was closely related to the mechanical properties of MHSs, transition layers, and matrix materials.

Due to a lack of the transition layer with the highest hardness, the compression performance of MHSCs fabricated under the holding time of 0 min was the lowest, even when the mechanical properties of the matrix material were the highest, which proves that the transition layer played a decisive role in the compression performance of the MHSCs.

When there was a transition layer between the MHSs and the matrix, the compression performance of the MHSCs was determined by the mechanical properties of the transition layer and the matrix material. Due to the extension of the holding time, the yield strength of the matrix material was reduced by 11.8%, from 204.8 MPa ± 9.6 MPa to 180.6 MPa ± 6.8 MPa, while the microhardness of the transition layer increased by 5.1%, from 510.6 HV ± 10.3 HV to 538.5 HV ± 11.6 HV, resulting in the poorer compression performance of the MHSCs held for 20 min, and further proving that the mechanical properties of the MHSCs could be determined by those of the matrix material when the mechanical properties of the transition layer were basically the same.

In summary, it can be concluded that the compression performance of the MHSCs was influenced by the MHSs, transition layers, and matrix materials. However, the transition layer played a major role in the compression performance of the MHSCs, followed by the matrix material.

Therefore, process parameters for preparing the transition layer and matrix should be successively considered in the design of the casting process of MHSCs. Only by comprehensively considering the influence of the interface and the matrix on the compressive performance of MHSCs, can the reasonable preparation process parameters of MHSCs be determined to fabricate MHSCs with good performance.

Conclusions

MHSCs were prepared by the casting method under different holding times. The physical parameters, microstructures, and mechanical properties of the MHSCs were studied. The conclusions are as follows:

• 1. The casting process of MHSCs was redesigned by introducing a new process parameter (holding time). The MHSCs with different microstructures were successfully prepared by controlling the holding time to achieve the purpose of precisely controlling the microstructure.

• 2. When the holding time was 10 min, the compression performance of MHSCs was the best. At this time, the transition layer was composed of a single Fe₂Al₇Si phase.

• 3. The compression experiment results indicated that the compressive properties of MHSCs were mainly determined by the transition layer and the matrix, excluding the influence of MHSs. The transition layer exerted a greater influence on the compression performance of MHSC than the matrix material.