Manufacturing of Al/SiC p composite foams using calcium carbonate as foaming agent

Materials

Commercial A356 cast aluminium alloy was used as a base material. The reinforcement phase consisted of SiC particles with purity of 98·0 wt- and mean mass particle size of 10 μm. Heating SiC particles for 1 h at 950°C and then for 2 h at 650°C in a conventional air furnace was carried out to improve the wettability between SiC_p and Al melt by removing adsorbed gases from the surface of particles. CaCO₃ powders with purity of 99·5 wt- and 5 μm average size were used as blowing agent. CaCO₃ powders were also heat treated at 200°C for 2 h to remove humidity and adsorbed gases from the surface and improve wetting properties and dispersion of CaCO₃ powders in molten metal. Scanning Electron Microscope (SEM) micrograph of heat treated SiC particles and CaCO₃ powders are respectively shown in Fig. 1.

OPEN IN VIEWER

Figure 1

Image (SEM) of heat treated a SiC particles and b CaCO₃ powders

Processing methodology

SiC_p reinforced aluminium matrix composite slurry was prepared by conventional stir casting techniques at 650–680°C, and then it was poured into a steel mould to obtain a composite ingot. This ingot was melted again at 650–700°C and then it was stirred at 650°C. The rotational speed of the stir equipment was 1400 rev min⁻¹. A content of 1 wt- of magnesium was subsequently introduced into the melt. After CaCO₃ powders were added into the melt swirl, the melt was stirred unceasingly for <1 min and then held at 710–730°C for several minutes to allow CaCO₃ blowing agent to decompose and release CO₂ gas. The foamed melt was removed quickly from the furnace and cooled immediately in air. Different amounts of CaCO₃ (1, 3 and 5 wt-) and SiC particles (5, 10 and 15 vol.-) was selected to produce composite foams with different relative densities and mechanical properties.

Oxides of aluminium and magnesium (assuming that Mg was introduced into the melt) form during decomposition of the CaCO₃ and associated reaction with the melt. These oxides form on the inner surfaces of the cells created during foaming and they apparently help to stabilise the cells against coarsening and coalescence. They also inhibit the migration of ceramic particles to these surfaces.6

The porosities of composite foams were calculated using equation (1) (1) where P is the porosity of composite foams, ρ and ρ_s are the densities of composite foams and the cell wall material respectively, and ρ/ρ_s, which is called the relative density of composite foams, indicates the ratio of the density of composite foams to the density of cell wall material. Produced composite foams and the amounts of relative density (ρ/ρ_s) and porosity P of each product are listed in Table 1.

OPEN IN VIEWER

Table 1

List of produced composite foams and amounts of relative densities, porosities and plateau stresses

Name of produced composite foams	AS5C3	AS10C5	AS10C3	AS10C1	AS15C3
Volume fraction of SiC particles, vol.-	5	10	10	10	15
Weight percentage of CaCO3 Powder, wt-	3	5	3	1	3
Relative density (ρ/ρs)	0·16	0·20	0·25	0·27	0·28
Porosity,	84·0	80·0	75·0	73·0	72·0
Plateau stress, MPa	2·25	3·33	8·59	12·37	16·4

Materials characterisations

The decomposition reaction of CaCO₃ blowing agent can be written as equation (2)9 (2) In order to determine reasonable foaming temperature of Al alloy melt, the decomposition temperature of CaCO₃ blowing agent has to be measured. Therefore, CaCO₃ blowing agent was studied with thermogravimetric analysis (TGA) using DTA/TG Rigaku-8150 system under a protection of high purity argon. The heating rate and flowrate of argon were respectively 20°C min⁻¹ and 40 mL min⁻¹. The amount of analyzed particulate was 16 mg.

Scanning electron microscopy by means of a LEO 1450VP (35 kV) system and optical microscopy with an Olympus PM3 microscope were carried out to observing the cell structure and evaluate the distribution of SiC particles in Al/SiC_p composite foams.

Specimens with the dimensions of 30×30×40 mm were prepared with a wire cutting machine for compression test. The size of compressive specimens was chosen to guarantee at least six cells in each direction because samples with smaller dimension can cause a significant loss of mechanical properties.3,13 All tests were performed using a Zwick Z250 universal test machine controlled by computer at a cross-head speed of 0·5 mm min⁻¹. Three compression tests were done for each run to show the reproducibility of the results.

Thermogravimetric analysis

The TGA curve of CaCO₃ powders is shown in Fig. 2. The results show that synthetic calcium carbonate powders (CaCO₃) undergo thermal decomposition above ∼620°C and that decomposition ends ∼850°C. Mass loss of CaCO₃ powders was ∼27·48. The experimental data were in agreement with other TGA studies carried out by Gergely et al.6

OPEN IN VIEWER

Figure 2

Thermogravimetric analysis curve of CaCO₃ powders

Microstructural features

The microstructure of produced composite foam with relative density of 0·20 is shown in Fig. 3. It can be found that the structure of cells is uniform (Fig. 3a) and SiC particles uniformly distribute in the cell wall of composite foams (Fig. 3b).

OPEN IN VIEWER

Figure 3

a optical microscope and b SEM microstructure of produced composite foam with relative density of 0·20

Compressive properties

The compressive stress–strain curves of composite foams with different relative densities are shown in Fig. 4. It is seen that the stress is a strong function of relative density (Fig. 5). SiC particles in cell wall can support the stress until the stress becomes sufficiently large and causes the damage by the failure either at or near the interface, or by the fracture of particles (Fig. 4a). The results show that with decreasing weight percentage of CaCO₃ powders, higher relative density and compressive strength was achieved (Fig. 4b). Just like other closed cell metallic foams, composite foams have characteristic compressive stress–strain curves, i.e. they involve three distinct stages: linear elastic deformation region, collapse plateau region and densification region.14–17

OPEN IN VIEWER

Figure 4

Compressive stress–strain curves of composite foams with different a volume fractions of SiC particles and b weight percentage of CaCO₃ powders

OPEN IN VIEWER

Figure 5

Effect of relative density on plateau stress of Al/SiC_p composite foams

The linear elastic deformation region of composite foams appears only at low strain (<0·03) (Fig. 4). In the collapse plateau regions, the strain is about from 0·03 to 0·60. The collapse plateau stresses of composite foams are almost constants. When the relative density is 0·25, the mean plateau stress of composite foams is ∼8·6 MPa. The stress in the collapse region increases with increasing relative density.

During the compression loading of closed cell foams, the strain tends to localise into a thin band, which causes buckling of cell walls.15 A consequence of the topology of the cell structures is that the strain distribution tends, for all of foams, to be locally non-uniform. The onset site of local plastic deformation depends on the cell structure. Deformation initiates in a single band, which is in contact with the loading surface, and proceeds to other layer one after the other.

The plateau region of the compressive stress–strain curve of composite foams is not very smooth and exhibits some serrations. The main reason is the addition of SiCp to Al alloy. According to the mechanical properties of matrix materials, metallic foams include three types, i.e. elastic, plastic and brittle.18 The brittleness of Al/SiCp composites is generally more than that of Al alloy,19 therefore, Al/SiCp composite foams belongs to brittle foams. When the stress of Al/SiCp composite foams reached the maximum, it came into the collapse plateau region. With the increasing the compressive strain, parts of cell walls produced cracks and suddenly brittle rupture, the space inside these cells decreased, and the stress was also reduced suddenly. Then the stress rose again with the increasing strain, and the next decrease in stress, which results from the sudden rupture of other cell walls, occurred at a larger strain. In addition, during straining, some cell wall fragments were ejected from the boundary cells, which also indicated the local brittle fracture of the Al/SiCp composite foams. Consequently, the typical serration plateau was formed in the compressive stress–strain curve.8 This is expected in view of the higher ductility associated with the cell wall microstructure of Al/SiCp composite foams.

In the densification regions of the compressive stress–strain curves of foams, the internal surfaces of more and more cells touched each other, foams became more and more compact, and the stress rose when the complete compaction of forms commenced. The rates of densification are higher for the denser samples of composite foams.

The plateau stress of Al/SiC_p composite foams are listed in Table 1. The effects of volume fraction of CaCO₃ and weight percentage of SiC_p on the plateau stress are respectively shown in Fig. 6. It can be found that when the weight percentage of CaCO₃ is a constant, the plateau stress of the composite foams increases with increasing volume fraction of SiC_p. This indicates that SiC_p has a strengthening effect on the compressive properties of composite foams. The more the content of SiC_p, the more evident the strengthening effect is. On the other hand, when volume fraction of SiC_p is a constant, the plateau stress of the composite foams decreases with increasing weight percentage of CaCO₃. It is the result of increasing volume fraction of porosities in the structure.

OPEN IN VIEWER

Figure 6

Effect of a volume fraction of SiC_p and b weight percentage of CaCO₃ powders on plateau stress of Al/SiC_p composite foams