Introduction

Over the past decades, particulate-reinforced aluminum matrix composites (PRAMCs) have attracted the attention of researchers due to their suitable density, high strength, hardness, and high wear resistance at ambient and high temperatures. Therefore, their industrial applications in aerospace, automotive, railway, electronics, and thermal management have progressed significantly compared to conventional alloys.1,2 So far, extensive research has been conducted on the fabrication approach to investigate the effect of various incorporated ceramic particles on the microstructure and mechanical properties of PRAMCs. Different routes such as powder metallurgy, [3] milling,4 pressure infiltration method,5 ultrasonic cavitation,6 electromagnetic field,7 friction stir welding,8 stir casting,9,10 and compo-casting11 have been used to synthesize PRAMCs. The stir casting process is one of the essential routes, mainly due to the lowest costs and easiest ways of fabrication,12 in which a distributed phase is mixed with a molten metal through mechanical stirring above the liquidus temperature. The particle incorporation is related to the viscosity of the melt, which is processing temperature-dependent. The compo-casting method is similar to stir casting, in which the particles are integrated into the semi-solid slurry with higher viscosity. Shabani et al.13 found that compo-casting significantly improves porosity reduction and strength enhancement.

Understanding the solidification characteristics is vital to assessing the microstructure and the performance of PRAMCs. The kinetics of the phase transformations determines microstructural features during their solidification.14 Depending on the nature of the reinforcement particle and the matrix alloy, there can be effects on the solidification kinetics, i.e., nucleation and growth of the metal matrix.14

Thermal analysis studies heat released or required due to phase transformation during solidification. The generated information can be used to discern the latent heat of solidification, the evolution of the fraction solid, the amounts and types of phases that solidify, and even dendrite coherency.15 Many standardized techniques consist of differential thermal analysis (DTA), differential scanning calorimetry (DSC), thermogravimetry analysis (TGA), and cooling curve analysis (CCA) investigate the solidification behavior of metals and alloys. Although very accurate, DTA, DSC, and TGA techniques require laboratory equipment and are limited to small size and reference samples. The cooling curve analysis (CCA) or cooling curve thermal analysis (CCTA) technique is widely used in monitoring the solidification behavior of monolithic metals and alloys, especially aluminum- and magnesium-based alloys. The obtained curve is called the cooling curve and, together with its derivatives, is employed to characterize the solidification phenomenon. The variation in the nature of the cooling curve always has a significant impact on the microstructure and, consequently, the performance of the material.16

It is imaginable that the presence of an insoluble ceramic reinforcement in a liquid metal will induce changes in the liquid-to-solid transformation.17,18 It is well-known that during solidification, the particles are pushed by the solidification front, engulfed by the solidification front, or/and act as solid-phase nucleation sites.19 Moreover, the reinforcement can be a barrier to heat diffusivity, restrict fluid convection, and induce morphological instabilities in the growth front.20 Formation of the Al5FeSi phase21 and changing the morphology of eutectic Si22 affects the thermal conductivity of aluminum-silicon alloys, which are generally used as base alloys for synthesizing composites.

Any change in manufacturing conditions of aluminum metal matrix composites (AMMCs) affects the solidification transformation. Furthermore, solute segregation around the SiC particles can affect the nucleation and dendritic growth behavior during solidification.23 Braszczyn´ski et al.18 determined the solidification kinetics of each phase and nucleation rate in the Al-SiC composite. In addition, variations in the solidification kinetics of A356-SiC composites were examined by using the thermal analysis method.14 Gonzalez-Rivera et al.24 found that increasing SiC content significantly increases eutectic growth temperature (TEG) and decreases solidification time. Kumar et al.25 reported a slight increase (2–3 °C) in liquidus temperature and TEG for A356-TiB2 composites. Cabrera et al.17 showed that the volume fraction of SiC particles increases the solid fraction at the dendrite coherency point, DCP, in Al-SiC metal matrix composites. Coherency is the instant during solidification when individual dendrites first impinge upon their neighbors. The crucial factor in semi-solid metal processing is the solid fraction at the forming temperature because it affects the microstructure and mechanical properties of the thixoformed components.25 Considering the findings in the literature, it is extraordinary that Jeng and Cheng26 noted that the principal characteristics of the solidification curves of the 6061 and the A356 matrix alloys and their composites are identical.

As pointed out, the metal matrix composites’ solidification analysis and crystallization process are complexes.18 Only a limited and controversial understanding exists regarding the solidification characteristics of AMMCs. This work profoundly explores the influence of ceramic particles with different types, i.e., Al2O3, ZrO2, and SiC, on the solidification characteristics of AMMCs based on the cooling curve thermal analysis technique. Obtained results are paramount as they indicate the behavior of the composite melt during the solidification process and the effect of adding different types of reinforcements on the characteristic temperature of each phase. A good understanding of the solidification features of the ex situ composites is needed to develop new AMMCs and improve casting processes.

Experimental Procedure

Table 1 shows the chemical composition of the A356.2 matrix alloy used in this study, obtained by glow discharge spectrometer (LECO GDS-850A). About 500 g of the A356.2 alloy was put into a ceramic crucible and melted in an electrical resistance furnace. After complete melting, the temperature was decreased and kept at 605 ± 5 °C for stirring in a semi-solid state based on the previous study’s findings.27 Following this, a graphite-coated stainless steel stirrer was inserted into the melt and rotated at 600 rpm28 using a DC motor (Figure 1). Lower speed leads to clustering of reinforcement particles, and higher rotation speed (more than 700 rpm) results in porosity formation in the microstructure. About 5% of SiC, Al2O3, and ZrO2 preheated at 300 °C was added individually to fabricate A356.2/SiC, A356.2/Al2O3, and A356.2/ZrO2 composites. Throughout stirring, the temperature was carefully monitored online by a thermocouple positioned in the melt to be sure that process was performed at the exact temperature range (605 ±5  °C) and solid fraction percentage required for the compo-casting operation. A similar procedure was carried out when 0.5 wt% Bi, in pure metallic granules (99.9%), were introduced into the A356.2 matrix alloy before adding reinforcement particles. Bi was added to decrease the surface tension of the liquid matrix alloy. The molten composite was manually stirred for 30 s and immediately cast and poured at 730± 5 °C in the cylindrical mold (30-mm diameter and 40-mm height), preheated at 500 °C for 15 min for thermal analysis. Two K-type thermocouples were placed near the center and near the wall, 20 mm above the bottom of the mold. To accurately control the radial heat extraction for the Fourier method, the bottom and top of the mold were insulated. The thermocouples were calibrated before all temperature measurements. The temperature variations versus time were recorded using high-speed data acquisition modules (KRYPTON DAQ) linked to a laptop with DEWESoft X3 at a 100 Hz/ch dynamic rate. After smoothing with the moving average model, recorded cooling curves and their first and second derivative curves were plotted and analyzed with FlexPro10 data analysis software. At least three thermal analysis runs were made for each condition. Figure 2 shows the cooling curves recorded for the A356.2 matrix alloy presenting the reproducibility of the applied thermal analysis experiment.

Table 1 Chemical Composition of A356.2 Matrix Alloy
Full size table
Figure 1
figure 1

Schematic set-up for fabrication of composite through compo-casting route.

Full size image
Figure 2
figure 2

(a) Cooling curves of A356.2 matrix alloy after three repetitions of the thermal analysis test. (b) The graphical expression of the method used to determine the characteristic temperature.

Full size image

For a more precise analysis of the cooling curve and drawing the solid fraction curve, it is essential to plot a curve called zero curve or baseline. It is assumed that the metal does not experience any phase transformation on the zero curve during the solidification process.29 The zero curve overlaps the first derivative cooling curve in the single-phase region, before liquidus and after solidus. The zero curve (ZF) was plotted based on the Fourier method, in which it is assumed that heat transfer occurs by conduction. The Fourier technique is more reliable than the Newtonian thermal analysis (NTA) technique.15,30 The Fourier equation with a heat source can be written as follows:

$$ \frac{\partial T}{{\partial t}} = \alpha \nabla^{2} T + \frac{1}{{C_{V} }}\frac{\partial Q}{{\partial t}} $$
(1)

where Cv is the volumetric specific heat and α is the thermal diffusivity. Equation 1 can be written as follows:

$$ \frac{\partial Q}{{\partial t}} = C_{V} (\frac{\partial T}{{\partial t}} - \alpha \nabla^{2} T) $$
(2)

where \(Z_{F} = \alpha \nabla^{2} T\) is the Fourier zero curve. Since a cylindrical mold is used in the current research, the ∇2T can be calculated as follows:

$$ \nabla^{2} T = \frac{1}{r}\frac{\partial }{\partial r}\left( {r\frac{\partial T}{{\partial t}}} \right) $$
(3)

Considering a cylindrical mold with known temperatures at radii R1 and R2 in the test sample,

∇2T can be calculated from Eqn. 3 as follows:

$$ \nabla^{2} T = \frac{{4(T_{2} - T_{1} )}}{{R_{2}^{2} - R_{1}^{2} }} $$
(4)

where T1 and T2 are temperatures at radii R1 and R2, respectively. Samples for metallography were selected near the tip of the thermocouple. Specimens were ground, polished with silica (OPS—0.3 mm), and finally etched with Keller’s reagent. XRD was conducted (Siemens-D500) using Cu Kα line generated at 40 kV and 35 mA to identify reinforcement particles in the matrix alloy. The microstructures were analyzed using an optical microscope (Nikon-MIDROPHOT-FXL).

Results and Discussion

Cooling Curves

Figure 3a shows the assembly of all recorded cooling curves of the A356.2 matrix alloy and synthesized composites collected from the thermocouples at the center.

Figure 3
figure 3

(a) Cooling curves of the A356.2 matrix alloy and synthesized composites and (b) cooling curve of A356.2/Al2O3 composite with corresponding 1st and 2nd derivatives curves and Fourier baseline.

Full size image

It should be noted that for better readability, the curves have been shifted in the time axis. However, the individual cooling curves started at zero. The cooling curves reflect the difference in the solidification behavior of the samples after the addition of SiC, Al2O3, and ZrO2 reinforcements and Bi addition. In general, regarding the slope changes on the cooling curves, three different phase transformations can be seen, indicated by arrows on the cooling curve of the A356.2 matrix alloy. The first phase reaction is related to forming the aluminum phase (α-al phase), and the second is associated with the evolution of the eutectic Al-Si phase. The Mg2Si intermetallic can also be seen on the cooling curve as the last phase. Although, it is more challenging to recognize than the previous two phases. The first and second derivative curves are plotted to accurately identify each phase formation characteristic (Figure 3b). For each phase transformation during solidification, when the temperature drops below nucleation temperature, dT/dt shifts up due to the increase of nucleation rate. After minimum temperature, the evolution of latent heat from primary nuclei causes it to reheat until the melt till starting coarsening step.

Phase Reactions

The segments related to the phase reactions of α-Al, eutectic Al-Si, and Mg2Si phase transformations are plotted separately in Figure 4. Figure 4a shows the temperature changes during the aluminum precipitation in more detail. The cooling rate is measured from the slope of the cooling curve in this area, between the pouring temperature and first phase formation temperature, i.e., nucleation temperature of α-Al or liquidus temperature. The cooling rate in the A356.2 matrix alloy was found to be 0.8 °C/s. However, the cooling rate of composite samples increased to 1.9 °C/s.

Figure 4
figure 4

Three sections for transformations of (a) α-Al, (b) Al-Si, and (c) Mg2Si phases for matrix alloy and fabricated composites.

Full size image

It is clear from Figure 4a that the α-Al transformation characteristics have been changed after adding SiC, Al2O3, and ZrO2 reinforcements with and without Bi addition. Nucleation temperature (TN) of the α-Al and undercooling (ΔT), which is defined as the difference between TN and minimum temperature (TMin), are two critical parameters used to predict microstructural changes, especially grain refinement of α-Al. Variations of TN, TMin, and TG for α-Al precipitation are given in Table 2. The liquidus temperature of composites appears to be lower due to decreased thermal conductivity of the composite melt and increased temperature gradient within the specimen.16 However, the presented results in the literature are not consistent. It can be seen that TN for matrix alloy is 618.3 °C which almost remained constant at 618.4 °C for A356.2/SiC and then slightly increased to 618.9 °C for A356.2+Bi/SiC composites. In addition, this value is 618.5 °C and 618.9 °C for A356.2/Al2O3 and A356.2+Bi/Al2O3 composites, respectively. TN was measured at 618.5 °C and 618.6 °C for A356/ZrO2 and A356+Bi/ZrO2 composites, respectively. As is shown, the maximum change in TN is 0.6°C. Jeng and Cheng26 found only a 2 °C difference in TN even after adding 15% and 20% Al2O3 (3–4 times more than current research). As can be seen, the TN is slightly higher for all fabricated composites than that of A356.2 matrix alloy, although it is negligible. ΔT indicates the barrier for nucleation of the α-Al phase. Calculation of undercooling shows that the magnitude of ΔT for A356.2 alloy is 9.4 °C, decreasing to the 8.9 °C and then increasing to 9.7 °C for A356.2/SiC and A356.2+Bi/SiC composites, respectively. The ΔT was measured at 10.4, 11, 10.2, and 9.2 °C for A356.2/Al2O3, A356.2+Bi/Al2O3, A356.2/ZrO2, and A356.2+Bi/ZrO2 composites, respectively. In addition, contrary to the results presented in,25 the undercooling of most composite samples is even higher than the matrix alloy. It can be related to the absence of nucleation of the α-Al on the SiC particles reported in the previous work.14 Moreover, Bi did not significantly affect the ΔT parameter of α-Al phase.

Table 2 Characteristic Parameters for Each Phase Reaction of the Matrix and the Fabricated Composites During Solidification
Full size table

Figure 4b shows the precipitation segment of the Al-Si eutectic phase. Eutectic growth temperature (TEG) and recalescence magnitude (TEG-TMin) are used to assess eutectic Si morphology. The variation of TEG is shown in Figure 5. It can be seen that the TEG is 568.6 °C for the A356.2 alloy, which increased slightly to 568.9 °C and 569.1 °C for A356.2/SiC and A356.2/Al2O3 composites, respectively. However, TEG remained constant for A356.2/ZrO2 composite at 568.6 °C. Gowri et al.31 and Gonzalez et al.24 observed that the TEG of composites was higher than that of the Sr-modified A356 matrix alloy at all cooling rates. It has been reported that increased TEG could be associated with the nucleation of eutectic Si on SiC particles.14,26 However, according to Figure 5, the variations of the TEG are not adequate to change the solidification kinetic of the eutectic Al-Si phase for all composites. Referring to,26 the principal characteristics of th6061 and A356 composite solidification and their matrix alloy are identical.

Figure 5
figure 5

Variations of growth temperature and recalescence magnitude of eutectic Al-Si reaction in matrix alloy and composites.

Full size image

On the other hand, adding Bi decreased the TEG for all composites regardless of the reinforcing particle type. TEG decreased from 568.9 to 567.1 °C for A356.2+Bi/SiC composite. TEG for A356.2/Al2O3 composite was at 569.1 °C, which fell by 2.2 °C after adding Bi. Moreover, after adding Bi to the A356/ZrO2 composite, TEG was reduced to 567.2 °C. The lowest TEG was obtained for A356.2+Bi/Al2O3. In addition, as can be seen, the recalescence magnitude for composite containing Bi is higher than the A356 matrix and composites without Bi. The recalescence for A356.2+Bi/SiC, A356.2+Bi/ Al2O3, and A356.2+Bi/ ZrO2 is 2.9 °C, 2.5 °C, and 2.5 °C, respectively. It is believed that a depression in TEG and an increase in recalescence are attributed to the modification level of eutectic Si.32,33

Figure 4c shows the formation of the intermetallic Mg2Si segment. The variations in the nucleation temperatures for the Mg2Si phase are shown in Table 2. The TN for Mg2Si was 550.7 °C for the A356.2 alloy. The nucleation temperature is lower for fabricated composites than matrix alloy. The lowest temperature was measured at 542.6 °C for A356.2+Bi/Al2O3. In addition, the variations of solidus temperature, the point where remained liquid was transformed entirely into a solid, are given in Table 2. It can be seen that the solidus temperature of the A356.2 matrix alloy was 532.7 °C, which fluctuated between 528.1 and 531.1 °C after the addition of reinforcement particles.

Solidification Range

The solidification temperature range (ΔTR) and solidification duration (ΔtR) are measured based on the temperature and time in the beginning (liquidus) and the end of solidification (solidus). Both are vital parameters, especially for thixo- and rheo-casting. Figure 6 demonstrates that the solidification range for A356.2 alloy is 85.6 °C. As can be seen, the ΔTR is wider in all fabricated composites than in the matrix. The lowest ΔTR was recorded at 86.9 °C for both A356.2+Bi/SiC and A356.2+Bi/ZrO2 composites. However, the highest ΔTR was measured at 89.4 °C for A356.2+Bi/Al2O3 composite.

Figure 6
figure 6

Solidification temperature range (ΔTR), solidification duration (ΔtR), and solidification rate (SR) for the matrix alloy and synthesized composites.

Full size image

Regarding solidification time, ΔtR, the duration of the matrix alloy was 742.6 s. As shown from Table 2, the solidification of the fabricated composite is completed sooner. The shortest duration time belongs to the A356.2+Bi/Al2O3 composite for 570.8 s. The A356.2/ZrO2 composite took the most extended (658.6 s) but was still 11% less than the A356.2 matrix alloy.

Gonzalez-Rivera et al.34 and Hanumanth et al.35 also reported a shortening of cooling time for the A356 matrix reinforced with SiC particles. The reinforcement particles do not show phase transformations during composite processing. So, there is a decrease in the latent heat released during the solidification of the composite and, consequently, a reduction in the solidification time.14 Furthermore, the quantity of liquid metal per unit volume decreases after adding reinforcement particles, reducing the solidification heat amount. Therefore, composite solidification time is shorter than matrix.18

To summarize this section, the solidification rate, which is affected by both the temperature range and time, was evaluated as a parameter showing the behavior of the composite in the semi-solid region. The solidification rate is vital since the mushy zone is vigorous in the thixoforming process. The solidification rate (SR) can be calculated as follows:

$$ {\text{SR}} = \Delta T_{R} /\Delta t_{R} $$
(5)

Figure 6 demonstrates that the SR for the A356.2 alloy is 0.12 °C/s which increased to 0.16 °C/s for A356.2/SiC and A356.2+Bi/SiC composites. The SR was 0.14 °C/s for both A356.2/Al2O3 and A356.2+Bi/Al2O3. This value was calculated at 0.13 °C/s for A356.2/ZrO2 and A356.2+Bi/ZrO2 composites. It can be attributed to the decrease in the latent heat released during the solidification of composites in which composite melt solidifies faster. Since the duration variation is much more than the solidification range, the role of time is more significant than the temperature.

On the other hand, it is worth noting that the CR and SR show different trends during the solidification of composites. Although the CR is equal for fabricated composites (1.9 °C/s), the SR is dissimilar. This may be related to the particles’ nature, distribution, and thermophysical properties,35 which affect the solidification phenomenon of metal matrix composites. Moreover, it was found that the SR is almost similar to the familiar composite. No significant changes were observed after the addition of Bi.

Coherency Characteristics

Coherency is defined as the point at which dendritic networks touch each other, which is along with hindering the melt flow. Determination of DCP is important as casting defects such as macro-segregation, shrinkage porosity, and hot tearing begin to develop. The first minimum difference between the thermocouple at a nearby inner wall (TW) and the center (TC) after nucleation of α-Al was considered for the dendrite coherency determination due to the higher thermal conductivity of the solid. Figure 7 shows the cooling curves of the A356.2 matrix alloy, A356.2/SiC, A356.2/Al2O3, and A356.2/ZrO2 composites recorded by TC and TW thermocouples and its projection on the TC cooling curve to determine the DCP and the corresponding temperature and time, TDCP and tDCP, respectively. TDCP, tDCP, and solid fraction at DCP are associated with grain size. In the case of dendritic solidification, free dendritic growth occurs until the dendrite does not touch one another, and the dendrite thickens after dendrite coherency. As shown in Figure 7a, the TDCP of matrix alloy is 607.8 °C. Moreover, the coherency of A356.2/SiC, A356.2/Al2O3, and A356.2/ZrO2 occurred at 606 °C, 606.6 °C, and 606.8 °C, respectively (Figure 6b–d). Figure 8 presents the variations of the coherency temperature and time for matrix alloy and synthesized composites. TDCP is lower for all composites than that of the matrix alloy.

Figure 7
figure 7

Cooling curves recorded by the center (TC) and wall (TW) thermocouples and corresponding TW—TC for (a) A356.2 matrix alloy, (b) A356.2/SiC, (c) A356.2/Al2O3, and (d) A356.2/ZrO2 composites.

Full size image
Figure 8
figure 8

Variations of temperature and time at DCP for the A356.2 matrix alloy and synthesized composites.

Full size image

Moreover, adding Bi into the composites slightly increased the TDCP in A356.2+Bi/SiC and decreased it in A356.2/Al2O3 and A356.2/ZrO2. Chai et al.36 emphasized the tDCP and proposed that grain size is inversely proportional to coherency time. As shown in Figure 8, A356.2/SiC and A356.2/Al2O3 composites show lower tDCP, but A356.2/ZrO2 presents higher tDCP than the matrix alloy.

Since there is no specific basis for measuring the temperature and time at the coherency point, and these two criteria can differ based on the time and temperature of the recording, the difference between the beginning of solidification and the coherency point (TN—TDCP and tN—tDCP) is a more accurate criterion for investigation. This difference corresponds to free dendritic growth, while after DCP, it corresponds to dendrite thickening, resulting in a fixed dendritic skeleton37. Figure 9 illustrates the variations of TN—TDCP. It can be seen that this difference is 10.5 °C for matrix alloy, which reached the highest at 12.5 °C for A356.2+Bi/Al2O3 composite. The range of changes is limited to around 2 °C.

Figure 9
figure 9

Variations of TN—TDCP and tN—tDCP for the A356.2 matrix alloy and synthesized composites.

Full size image

On the other hand, the tN—tDCP is 48.7 s for A356.2 alloy, which increased to 52.6 s and 51.7 s for A356.2/SiC and A356.2+Bi/SiC composite, respectively. The longest tN—tDCP was measured at 59.7 s for A356.2+Bi/ZrO2 composite. As shown in Figure 9, the tN—tDCP for all composites is higher than matrix alloy. It means that the contact of dendrites is delayed. Therefore, smaller grain touches each other later.

Solid Fraction at DCP

The solid coherency fraction solid is one of the most critical parameters in the alloy solidification process, which is related to the flow stoppage of molten metal.38 The integrated area between the first derivative curve and baseline was calculated to measure the solid fraction curve. The baseline variation represents phase evolution during solidification and displays as the first derivative curve trend. Figure 10a shows the cooling curve (CC) and corresponding solid fraction curve (fS) for the A356.2 matrix alloy. As can be seen, the fS of A356.2 at DCP is 16%. The fS remained constant at 16% for A356.2/SiC composite.

Figure 10
figure 10

Cooling curves and corresponding solid fraction curve for (a) A356.2 matrix alloy, (b) A356.2/SiC, (c) A356.2/Al2O3, and (d) A356.2/ZrO2 composites.

Full size image

Moreover, compared to the A356.2 alloy, the fS increased slightly to 18% and 17% for A356.2/Al2O3 and A356.2/ZrO2 composite, respectively. Obtained results agree with Cabrera’s et al.17 findings. The reinforcement particles affect the solidification kinetics of the primary α-Al phase. However, this effect is relatively small compared to the grain refinement reagent, i.e., AlTiB and TiC. An increase of fS at coherency can be associated with the longer tN—tDCP while the dendrite touches each other later.

Additionally, this may be caused by reinforcement acting during solidification as a physical barrier to solute diffusion.20 This could decrease the solute gradient and the dendrite tips’ local velocity growing in the particles’ immediate neighborhood.17 On the other hand, adding Bi did not affect the fS at DCP for all composites. The fs remained almost constant after adding Bi into the fabricated composites.

Microstructural Characterization

Composite samples were analyzed by X-ray diffraction (XRD) to evaluate the incorporation of reinforcement particles in the A356.2 matrix. Figure 11 shows the X-ray spectra of the A356.2/SiC, A356.2+Bi/SiC, A356.2/Al2O3, A356.2+Bi/Al2O3, A356.2/ZrO2, and A356.2+Bi/ZrO2 composites. The name of each phase is mentioned next to the matched peak. In addition to the main phase, Al and Si, the peaks corresponding to the SiC, Al2O3, and ZrO2 phases, along with Bi, were detected. In addition, analyses of the results verified the integration of particles in A356.2 alloy before and after adding Bi.

Figure 11
figure 11

X-ray diffraction patterns of investigated composites.

Full size image

Figure 12 shows the microstructure of the A356.2 matrix alloy, A356.2/SiC, A356.2+Bi/SiC, A356.2/Al2O3, A356.2+Bi/Al2O3, A356.2/ZrO2, and A356.2+Bi/ZrO2 composites. Based on the recorded cooling curve presented in Figure 3a, the microstructure of the A356.2 has two main phases: α-Al and eutectic Al-Si (Figure 12a). Evaluation of solidification parameters, i.e., nucleation temperature, undercooling magnitude, coherency temperature, and solid fraction at DCP, showed that the presence of reinforcement particles has no remarkable impact on the primary α-Al phase. Kim and Rohatgi23 noted that the precipitation of the primary phase on the surfaces of ceramic particles in the melt is more favorable in hypereutectic Al-Si alloys than for hypoeutectic alloys. Therefore, Al2O3, ZrO2, and SiC particles do not act as nucleating agents for the primary α-Al dendrites. Moreover, eutectic Si appears as coarse and flake-like morphology. The structure of eutectic Si significantly affects the mechanical properties of Al-Si alloys. Figure 12b shows the microstructure of the A356.2/SiC composite. It can be seen that SiC particles integrated into the matrix and were found to lay in the eutectic region. This is because particulate reinforcements are generally segregated in the last freezing zone during solidification and rejected by the solidifying primary α-Al phase. This effect is known as the particle-pushing phenomenon.34,39 After adding Bi, eutectic Si transforms to a more refined shape in the form of a lamellar structure in A356.2+Bi/SiC composite (Figure 12c). Figure 12d and e shows the microstructure of A356.2/Al2O3 and A356.2+Bi/Al2O3, respectively. Al2O3 particles were seen to be incorporated into the aluminum matrix. Moreover, Si’s flake and sharp nature were transformed into a shorter and more refined structure. The microstructure of A356.2/ZrO2 is shown in Figure 12f. ZrO2 particles are located in the eutectic region, the last freezing zone in hypoeutectic Al-Si alloys.

Figure 12
figure 12

Optical microstructure of (a) A356.2 alloy, (b) A356.2/SiC, (c) A356.2+Bi/SiC, (d) A356.2/Al2O3, (e) A356.2+Bi/Al2O3, (f) A356.2/ZrO2, and (g) A356.2+Bi/ZrO2 composites.

Full size image

Against Samuel’s finding,40 no sign of eutectic Si refinement was observed after adding SiC, Al2O3, and ZrO2 reinforcements into the A356.2 matrix alloy. Perhaps, it is for this reason that Rohatgi and his colleagues41 concluded that further study is needed to quantify the effect of particles on the nucleation and growth of phases during the solidification of composite.

Figure 12g shows the A356.2/ZrO2 composite’s microstructure after adding Bi. It can be seen that the morphology of eutectic Si changed to the refined structure after adding the Bi. It is clear from Figure 12c, e, and f that Bi has a refining effect on the shape and size of eutectic Si. It has been reported that Bi can modify the morphology of eutectic silicon from flake to lamellar crystal.42,43 As shown in Figure 5, Bi addition induces lower eutectic growth temperature (TEG), and the system needs more activation energy to overcome the restriction of Si growth during solidification. These results contrast those previously reported in which the modification effect of Sr addition was lost due to the presence of SiC.14 In addition, the area fraction of eutectic Si increases after refinement. This can lead to the engulfment of reinforcement particles by the growing eutectic Al-Si interfaces in the last pools of the liquid matrix, improving their distribution in the A356.2 matrix.

Conclusions

In the present study, the influence of the addition of SiC, Al2O3, and ZrO2 particles on the microstructural and solidification characteristics of A356.2 alloy with and without Bi addition is studied with the use of the Fourier thermal analysis method. The conclusions can be drawn as follows:

  1. 1.

    Microstructural observation and XRD analysis revealed that SiC, Al2O3, and ZrO2 particles have been successfully embedded in the A356.2 matrix alloy.

  2. 2.

    The cooling rate increased from 0.8 °C/s to 1.9 °C/s after addition of SiC, Al2O3, and ZrO2 particles.

  3. 3.

    The solidification temperature range increased by 5%; however, the solidification duration decreased by 21%.

  4. 4.

    The coherency point characteristics and solid fraction at coherency did not change remarkably after incorporating SiC, Al2O3, and ZrO2 particles.

  5. 5.

    Adding Bi element to composites did not affect the freezing characteristic of the α-Al phase. Still, it resulted in the depression of eutectic growth temperature and the refinement of eutectic Si.

  6. 6.

    The obtained data from thermal analysis and microstructure characterization of A356.2 and fabricated composites depict that reinforcement particles cannot act as nucleation agents for both α-Al and eutectic Si.