Preparation and Compressive Performance of an A356 Matrix Syntactic Foam
2018 Volume 59 Issue 5 Pages 699-705
Details
2018 Volume 59 Issue 5 Pages 699-705
The uniformity of the cell size of aluminum foams prepared by traditional melt foaming and gas injection methods is difficult to control. In order to improve the controllability of cell size and mechanical performance of aluminum foam, an A356 matrix syntactic foam was prepared by the improved stir casting process. Moreover, the compressive property of the syntactic foam was studied and compared with other metal foams. Al2O3 hollow spheres with the diameters of 3–5 mm were added in the molten A356 alloy, then the syntactic foam can be obtained after stirring and compaction. The hollow spheres in the syntactic foam are random loose packed. The volume fraction of 3–3.5 mm hollow spheres in the syntactic foam could reach to 56%. The total densities of the syntactic foams are around 1.79 g/cm3. Through the analysis both in macroscopic and microscopic, the interfacial bonding between aluminum matrix and hollow spheres was proved to be good even after mechanical cutting. The quasi-static compression result showed that the plateau stress of the syntactic foam with 3–3.5 mm hollow spheres could reach up to 46 MPa, and the corresponding densification strain energy is 20 MJ/m3. Moreover, the plateau region is flat, long and relatively high. The syntactic foam with the hollow spheres of 3–4 mm diameter has the highest overall porosity and the best compressive performance. Compared with other traditional aluminum foams, this syntactic foam has advantages in energy absorption performance and the adjustability of cell size.
Fig. 8 (a) Quasi-static stress-strain curves of the syntactic foams with different sphere diameters, (b) the energy absorbed-strain curves of the syntactic foams with different sphere diameters.
Closed cell aluminum foams have been getting a wider application in automotive, aerospace and decoration fields for their light mass, high specific strength, good energy absorption performance and noise reduction.1) Melt foaming route and gas injection route are the two main methods to fabricate closed cell aluminum foams.2–4) The foam cells are not uniform in size and irregular in shape because the cell size and foam structure are difficult to control and adjust, which to a certain extent, limit the further application of aluminum foams prepared by these two methods. Composite metal foams are usually fabricated by adding the hollow spheres into metal matrix,5) the regular spherical pores of which help to avoid stress concentration during deformation. Moreover, the composite aluminum foams have the advantage of designability. The desired porous structure can be achieved by designing the pore size, stacking mode and stacking density of the composite foam, which could greatly extend the application of porous metals. The syntactic foams have a broad application prospect in aerospace and military fields6) due to their high plateau stress and energy absorption capacity. So it is meaningful to explore a simple and effective method to prepare aluminum matrix syntactic foams.
Usually the aluminum alloys are used as the matrix in syntactic foams for their lightweight. Different types and sizes of hollow spheres can be used to synthesize the syntactic foams. Currently the materials of hollow spheres typically included alumina,7) SiC,8) low carbon steels9) and stainless steels.9) Besides, expanded perlite particles,10) pumice particles11) which contain the hollow structure and fly ash cenospheres12) were also employed as fillers. The preparation methods of hollow sphere syntactic foams mainly include stir-casting,13) gravity casting,9) pressureless infiltration,8) counter-gravity pressure infiltration,14) pressure infiltration15,16) and vertical centrifugal casting.17) Moreover, powder metallurgy method was usually adopted when the matrix and hollow spheres were both steels.18)
Among these preparation methods, casting method is most widely used, and various pressure infiltration methods are adopted to improve the binding strength between hollow spheres and matrix. Stir-casting is simpler and more economical compared to other methods, but the volume fraction of hollow spheres is low because spheres are easy break during mechanical stirring.19,20) Ceramic material is often used as fillers in the military applications, such as ceramic armors, which is mainly due to the excellent energy absorption in the collapse process of brittle ceramic material. Moreover, this energy absorbing effect is more obvious in the dynamic impact. Ceramic hollow spheres are usually used for refractories and have extensive sources. Compared with composite foams prepared with steel hollow spheres which get more research at present,5,9) the foams prepared with ceramic hollow spheres have lower cost and greater value of potential application. But the poor wettability between spheres and aluminum matrix should be considered when using ceramic materials. Moreover, in order to increase the porosity and energy absorption of the syntactic foam, the diameter and volume fraction of hollow spheres in the aluminum matrix should be increased as much as possible. In the present work, the preparation process has been improved based on stir-casting, the ceramic hollow spheres with diameters of 3–5 mm were used and the interfacial bonding conditions were studied. The compressive property of the syntactic foam was also investigated.
Commercial A356 alloy provided by Qinhuangdao Anye Aluminium Industry Co., Ltd. was used as the matrix. The Al2O3 hollow spheres were provided by Zhengzhou Yu Li Industrial Co., Ltd. The Al2O3 hollow spheres with 3–5 mm outer diameter were used in order to achieve the lightweight property, and the hollow spheres were further sieved to 3–3.5, 3.5–4, 4–4.5 and 4.5–5 mm for the uniformity of cells in the syntactic foam samples. The average densities of these hollow spheres are given in Table 1. The density roughly decreases with the increasing of outer diameter. Figure 1(a) shows the macro morphology of the hollow spheres after sieved. It can be seen that the shape of the spheres is relatively round. The variation in wall thickness with the outer diameter of hollow spheres is shown in Fig. 1(b), and the wall thickness increases with the increasing of sphere size. When the outer diameter of sphere is greater than 4.0 mm, the wall thickness increases quickly in order to keep the strength of sphere. Due to the good match of sphere size and wall thickness, the spheres with 4–4.5 mm diameter have the lowest density. Equation (1) describes the relationship among porosity, outer diameter and wall thickness of the hollow sphere.
| \begin{equation} p_{\text{s}} = \frac{(d_{\text{s}}-2t)^{3}}{d_{\text{s}}{}^{3}} \end{equation} | (1) |
(a) The macroscopic morphology of the hollow spheres used in this paper, (b) the relationship between outer diameter and wall thickness of the Al2O3 hollow spheres.
Then the overall porosity of the syntactic foam could be obtained by the volume fraction of hollow spheres in the foam and the porosity of the hollow spheres, as described in eq. (2).
| \begin{equation} p = \varphi_{\text{s}}\cdot p_{\text{s}} \end{equation} | (2) |
Figure 2 shows the schematic diagram of the improved stir-casting process. First, the aluminum was melted and held at 690°C, and the hollow spheres were preheated at 620°C for 60 min. The volume fraction of hollow spheres was preliminary designed to about 60% according to the random loose packing.9,19,21) Next, the preheated hollow spheres were added into the aluminum melt and stirred using an alumina rod at a speed of 1 cycle per second, the diameter of the rod was 8 mm. It is noteworthy to make every sphere separated by a liquid aluminum layer before strongly stirring, which could ensure the minimum gaps between spheres. The gas among hollow spheres could escape from the slot of middle-divided crucible and the molten aluminum alloy surface. Then, external force was applied through a stainless steel plate to make the spheres stacked more closely. The liquid aluminum can be considered sufficiently filled the space among hollow spheres when a little bit of molten aluminum escaped from the gap between the platen and crucible. Finally, the sample was held at 690°C for about 60 min before furnace cooling. The temperature of the melt may be decreased a little during the addition of spheres, so the holding process could make the temperature in the crucible more uniform and high enough to ensure the fluidity of the melt after stirring, and then it facilitates a more rational distribution of the spheres in the liquid melt. There are two highlights in this improved stir-casting process. The guarantee of the minimum gap between spheres is conducive to the bonding strength between the spheres and matrix. The compaction after stirring could avoid the floating of the hollow spheres and help to obtain the syntactic foam with high volume fraction of hollow spheres.
Schematic of the improved stir-casting process.
A SYJ-200 precision cutting machine was used to obtain the cross-section of the syntactic foam sample. The surface of syntactic foam after cutting was ground and polished, and an optical microscope of Carl Zeiss Axio Scope.A1 was used to obtain the interface combination condition between hollow spheres and A356 matrix.
2.2 Compressive testThe specimens for compressive test was 30 mm in diameter, and about 50 mm in height. The quasi-static compressive tests were conducted on a WDW-100E electronic universal testing machine at the strain rate of about 1 × 10−3 s−1. The samples with each size of hollow spheres were tested three times, and the mechanical property of the most representative one was chosen to be studied. The load-displacement curves were obtained from the experiments, then the nominal stress and strain can be calculated by the following formulas (3) and (4).
| \begin{equation} \sigma = \cfrac{F}{\pi\cdot\biggl(\cfrac{d}{2}\biggr)^{2}} \end{equation} | (3) |
| \begin{equation} \varepsilon = \frac{\Delta L}{L_{0}} \end{equation} | (4) |
Energy absorption capacity of the composite foam is characterized by the densification strain energy, namely the absorbed energy before densification, as shown in the gray area of Fig. 3.22) Figure 3 also shows the quasi-static compression performance of the syntactic foam with 3.5–4 mm sphere diameter, which will be analyzed later. The energy absorption efficiency of the syntactic foam η(ε) is defined as eq. (5), and the η(ε)-ε curve was also drawn in Fig. 3.
| \begin{equation} \eta(\varepsilon) = \frac{\displaystyle\int_{0}^{\varepsilon}\sigma(\varepsilon)\mathrm{d}\varepsilon}{\sigma(\varepsilon)} \end{equation} | (5) |
| \begin{equation} \frac{d\eta(\varepsilon)}{d\varepsilon}\bigg|_{\varepsilon = \varepsilon_{\text{D}}} = 0 \end{equation} | (6) |
| \begin{equation} W = \int_{0}^{\varepsilon}\sigma(\varepsilon)\mathrm{d}\varepsilon \end{equation} | (7) |
| \begin{equation} W_{\text{D}} = \int_{0}^{\varepsilon_{\text{D}}}\sigma(\varepsilon)\mathrm{d}\varepsilon \end{equation} | (8) |
| \begin{equation} \sigma_{\text{P}} = \frac{\displaystyle\int_{\varepsilon_{0}}^{\varepsilon_{\text{D}}}\sigma(\varepsilon)\mathrm{d}\varepsilon}{\varepsilon_{\text{D}}-\varepsilon_{0}} \end{equation} | (9) |
Densification strain energy calculation of an A356 matrix syntactic foam with 3.5–4 mm sphere diameter.
The specific plateau stress could be obtained by dividing the plateau stress by the overall density of the syntactic foam.
Figure 4 shows the cross-sectional morphologies of the specimens prepared by the improved stir-casting process. It can be seen that the hollow spheres arranged uniformly in the matrix. The final compaction process helps to avoid the density stratification of ceramic spheres and aluminum matrix. The balls were barely broken and filled with aluminum. Most spheres remain connection with matrix even after mechanical cutting. So the matrix and hollow spheres have a good bonding. The preheated hollow spheres had less heat exchange with liquid aluminum, which ensures the fluidity of aluminum melt. The wettability between ceramic spheres and A356 melt could be improved by the elevated temperature of ceramic spheres. Moreover, the wettability could also be greatly improved by the Mg and Cu elements in A356 melt.23) The good fluidity of aluminum melt and improved wettability are responsible for the good bonding between ceramic spheres and liquid aluminum.
Cross-sectional morphologies of the samples with hollow sphere diameters (a) 3.5–4 mm, (b) 4.5–5 mm.
Figure 5 describes the microscopic interface between hollow spheres and aluminum matrix. The two spheres joints show that the spheres were separated by a layer of aluminum alloy, even though the layer is very thin at some place (Fig. 5(b)). The coating of liquid aluminum before stirring during the preparation process guaranteed the minimum gap between spheres. Figure 5(c) shows that the liquid aluminum could wrap the sphere tightly even if the surface of the sphere is rough. Figure 5 also shows that cell wall of the Al2O3 hollow sphere is not very dense, which is also beneficial for the lower density of the syntactic foam. The A356 matrix syntactic foam prepared by the improved stir-casting method not only has the tightly coating of aluminum melt on the hollow spheres, but also the thin matrix walls between spheres, which helps to obtain the high porosity of the syntactic foam.
Microscopic interfaces between hollow spheres and A356 matrix (a) and (b) two spheres joint, (c) a connection characteristic.
Figure 6 describes the compressive process of A356 matrix syntactic foam. Figure 6(a)–(g) show that the compression failure started from wrinkles on the lower part, then the outer layer of the foam gradually peeled and collapsed, and finally the holes were compacted. A layer by layer failure characteristic was also exhibited in Fig. 6. Figure 7 shows the compressed morphology of the A356 matrix syntactic foam. The ceramic cell walls have changed into powders and distribute in the gap of compressed cells. The compaction of powders can effectively buffer the energy at the end of compressive process, which could further improve the energy absorption performance of the syntactic foam. Figure 7 also shows the collapse of the outer layer after deformation, which is a characteristic of brittle fracture. The combination of Al2O3 ceramic sphere wall and plastic aluminum alloy matrix affect the deformation mode of the syntactic foam. It can be foreseen that brittle sphere walls would cause some fluctuations in stress-strain curve. But the fluctuation is much less than the aluminum foam with brittle particles in cell walls,24) as shown in Fig. 8(a). Precisely because the suitable sphere diameters and compactness of the syntactic foam, the relatively flat, long and high platform region were obtained in the quasi-static stress-strain curves shown in Fig. 8(a).
(a)–(g) Deformation of the A356 matrix syntactic foam (3.5–4 mm hollow spheres) with gradual compression, the compressive strains from (a) to (g) were 0%, 12%, 24%, 33%, 40%, 48%, 57%, respectively.
Macroscopic morphology of the A356 matrix syntactic foam (the diameter of the hollow spheres is 3–3.5 mm) after compression.
(a) Quasi-static stress-strain curves of the syntactic foams with different sphere diameters, (b) the energy absorbed-strain curves of the syntactic foams with different sphere diameters.
Table 2 shows the performance parameters of syntactic foams prepared in this paper with different sphere diameters. The volume fraction of the 3–3.5 mm hollow spheres could reach to 56%. The volume fraction and the overall porosity both decrease with the increase of sphere diameter. Although the density of the hollow spheres decreases with the increasing of outer diameter, the volume fraction of the spheres can be filled also decreases, so the densities of the samples with different sphere sizes are close, they are all around 1.79 g/cm3. The compressive stress is mainly related to the overall density of porous materials,25) so the compressive stress and plateau stress of syntactic foams with different cell sizes are close due to the similar overall density. Although Fig. 8(a) shows that the differences of stress-strain curves with different spheres diameters are not obvious, the specific plateau stress is roughly decreased with the increasing of sphere diameter. It can be seen from Fig. 8(b) that the energy absorption performances of syntactic foams with 3–3.5 mm and 3.5–4 mm sphere diameters are obviously better than the syntactic foams with larger spheres. The syntactic foam with the sphere diameter of 3–3.5 mm has the largest porosity and the highest plateau stress, the plateau stress of which can reach to 46 MPa. Due to the large densification strain (which can reach to 0.55) and high energy absorption efficiency (the maximum of which can reach to 0.50), the foam with the sphere diameter of 3.5–4 mm has the best energy absorption property (the value of the densification strain energy can reach to 23 MJ/m3). So the hollow spheres with the diameter of 3–4 mm are appropriate to prepare the A356 matrix syntactic foam by using this stir-casting method, then the syntactic foam with the optimum of volume fraction of hollow spheres and compressive performance can be obtained. One can conclude from Fig. 8(a) that compressive process of the A356 matrix syntactic foams conform perfectly to the typical compressive characteristics of porous materials,24,25) the stress-strain curve of which can be divided into linear elastic stage, stress plateau stage and densification stage. Moreover, it had the relatively flat platform region and large densification strain on the basis of high compressive strength. Therefore, this syntactic foam is suitable for using as energy absorption material, and the force applied by the protected object would maintain at around the plateau stress of syntactic foam in a large deformation range.
Figure 9 shows the comparisons of plateau stress and energy absorption capacity of different metal syntactic foams. In the case of traditional aluminum foams prepared by melt foaming and gas injection methods, the compressive strength and cell uniformity are far less than the syntactic foams prepared in the present work. Even though for the gas injection aluminum foams with the best mechanical properties as far as is known, their densification strain is low due to the high density of aluminum foams.33) The aluminum foam in ref. 17) was prepared with expanded polystyrene, and energy absorption performance of which is also not that satisfactory. In the case of hollow spheres packed structure, the mechanical properties of which are relatively low due to the absence of matrix.
(a) Comparison of plateau stress of different metal syntactic foams, (b) comparison of energy absorption capacity of different metal syntactic foams.
The syntactic foam in the present work was also compared with other aluminum matrix syntactic foams. When the hollow particle diameters are no more than 1 mm, many stress-strain curves rise evidently in the platform region.8,13) In the case of Al matrix foams with larger hollow structure (the diameters are in the range of 1.92–5.6 mm), the sphere materials include steel,9,32) iron,15) alumina,7) expanded perlite,10,14) pumice particle,11) etc. Densities of foams in the upper right area in Fig. 9(a) and (b) are high. The upper left area is the most concern, because syntactic foams in this area have the better compressive performance with lower density. Although some foams are in the upper left of the syntactic foam obtained by the present work, the advantage of the A356 matrix syntactic foam prepared in this paper is that the platform region is both flat and long,7,10,11,15) which could ensure that the stress born by the protected object be almost constant in a large deformation. The good match of the amounts of matrix and spheres also contributes to the mechanical properties of the syntactic foams. Through comparing the stress-strain curves of different metal foams, the syntactic foams in the present work is found to have great advantages in lightweight, plateau stress and the length and flatness of platform region. The syntactic foam prepared in this paper is suitable for developing as a promising energy absorption material.
