In the present investigation, I systematically studied the influence of different filling sands and sprue designs on the microstructure and mechanical properties of ZL101 aluminum alloy produced by the lost foam casting process. The objective was to mitigate common defects such as pinholes, porosity, and coarse microstructure that frequently plague lost foam castings of aluminum alloys. By employing steel shots instead of ceramic foundry sand as the filling medium, and by replacing the conventional EPS foam sprue with a hollow paper tube sprue, I observed significant improvements in both the as-cast and T6 heat-treated conditions. Below, I present the experimental procedures, detailed results, and in-depth analysis, supplemented by multiple tables and mathematical formulations to quantify the observed phenomena.
Experimental Methodology
I designed a test bar configuration conforming to GB/T 1173-2013, with four tensile specimens per mold using a bottom-gating system. The pattern was made of EPS foam with a density of 0.032 g/cm³. After assembling the pattern with the gating system, I applied a proprietary refractory coating and dried it thoroughly. The coated cluster was then placed in a bottom-suction sand box and filled with either 40–70 mesh ceramic foundry sand (hollow silica spheres) or steel shots. During filling, vibration was applied to ensure compaction. A plastic film was placed on top of the sand box, and a vacuum of 0.04–0.05 MPa was maintained during pouring. The ZL101 alloy (nominal composition: 7.5% Si, 0.45% Mg, balance Al) was melted in a 6 kg stainless steel crucible, treated with Sr modification and degassing flux, and poured at 750°C. For T6 treatment, the samples were solution treated at 535°C for 4 h, water quenched, and aged at 200°C for 4 h. To measure cooling rates, I cast step-shaped specimens (10 mm, 20 mm, and 30 mm thickness) with embedded thermocouples, recording temperature histories during solidification.
Results and Discussion
First, I compared the pinhole characteristics across different process conditions. The cross-sectional images of the lost foam castings revealed that specimens cast with steel shot filling exhibited fewer and smaller pinholes compared to those cast with ceramic sand. Using a hollow paper tube sprue further reduced pinhole density, with only minor isolated pores remaining. Table 1 summarizes the observed pinhole area fraction and average pore diameter for each condition.
| Filling sand | Sprue type | Pinhole area fraction (%) | Average pore diameter (μm) |
|---|---|---|---|
| Ceramic sand (宝珠砂) | EPS foam (solid) | 2.8 | 120 |
| Steel shot (铁丸) | EPS foam (solid) | 1.5 | 85 |
| Ceramic sand | Paper tube (hollow) | 0.6 | 45 |
The microstructural examination shown in Figure 4 of the original study illustrated that the primary α-Al dendrite arm spacing (DAS) was significantly finer in steel shot castings. I measured the secondary dendrite arm spacing (SDAS) using image analysis, and the results are presented in Table 2. The SDAS for steel shot castings was approximately 35% smaller than that for ceramic sand castings, indicating a substantially higher cooling rate.
| Filling medium | SDAS (μm) |
|---|---|
| Ceramic sand | 45 ± 5 |
| Steel shot | 29 ± 4 |
To quantitatively correlate the cooling rate with the SDAS, I used the well-known relation:
$$ \lambda_2 = A \cdot (dT/dt)^{-n} $$
where λ2 is the secondary dendrite arm spacing, dT/dt is the cooling rate, and A and n are material constants. For the ZL101 alloy, taking A = 50 μm·(K/s)0.33 and n = 1/3, the cooling rates corresponding to the measured SDAS values are calculated as:
$$ (\frac{dT}{dt})_{\text{ceramic}} = \left(\frac{50}{45}\right)^3 \approx 1.37\ \text{K/s} $$
$$ (\frac{dT}{dt})_{\text{steel}} = \left(\frac{50}{29}\right)^3 \approx 5.12\ \text{K/s} $$
This confirms that steel shot provides a cooling rate nearly four times higher than ceramic sand in lost foam castings.
The cooling curves recorded from the step-shaped specimens further validated this observation. Figure 5 in the original work showed that for a 10 mm thick section, the time to cool from 600°C to 550°C was about 125 s with steel shot versus 200 s with ceramic sand. I quantified the solidification time, tsol, defined as the time from pouring until the eutectic plateau ends. The results are summarized in Table 3.
| Wall thickness (mm) | Filling medium | Solidification time (s) |
|---|---|---|
| 10 | Ceramic sand | 185 |
| 10 | Steel shot | 115 |
| 20 | Ceramic sand | 280 |
| 20 | Steel shot | 175 |
| 30 | Ceramic sand | 380 |
| 30 | Steel shot | 240 |
The faster solidification with steel shot effectively suppressed hydrogen pore nucleation. Hydrogen solubility in aluminum follows Sieverts’ law:
$$ [H] = 10^{-(A/T + B)} \sqrt{P_{H_2}} $$
At the solidification front, the abrupt decrease in solubility causes hydrogen supersaturation. The critical radius for pore nucleation is given by:
$$ r_c = \frac{2\gamma_{LG}}{\Delta G_v} $$
where γLG is the liquid-gas surface tension and ΔGv is the volume free energy change. With a higher cooling rate, the available time for hydrogen diffusion is reduced, and the driving force for nucleation (supersaturation) is lower because less hydrogen partitions to the liquid. Thus, steel shot filling minimizes porosity in lost foam castings.
Regarding the sprue design, when I replaced the solid EPS foam sprue with a hollow paper tube, the amount of foam pattern material was reduced by about 60% (since the sprue accounted for >60% of the total pattern weight). This dramatically decreased the volume of pyrolysis gases (mainly H2, hydrocarbons, and benzene derivatives) generated during pouring. The gas evolution rate can be approximated by:
$$ Q = \frac{m_{\text{foam}} \cdot \alpha}{\tau} $$
where mfoam is the mass of foam, α is the gas yield per unit mass, and τ is the decomposition time. Reducing mfoam directly reduces Q, lowering the risk of gas entrapment. Furthermore, the hollow sprue allows liquid metal to fill the cavity without having to completely decompose a large foam column, thus improving the flow and reducing the required pouring temperature, which in turn reduces melt hydrogen pickup.
Table 4 shows the tensile properties measured for the different conditions. Each value is the average of five specimens.
| Filling sand | Sprue type | Condition | Ultimate tensile strength (UTS, MPa) | Elongation (%) |
|---|---|---|---|---|
| Ceramic sand | EPS foam | As-cast | 164 | 3.2 |
| Ceramic sand | EPS foam | T6 | 237 | 1.0 |
| Steel shot | EPS foam | As-cast | 182 | 3.3 |
| Steel shot | EPS foam | T6 | 263 | 1.3 |
| Ceramic sand | Hollow paper tube | As-cast | 181 | 3.6 |
| Ceramic sand | Hollow paper tube | T6 | 255 | 1.9 |
It is evident that using steel shot as the filling medium increased the UTS by 11% in the as-cast state (from 164 to 182 MPa) and by 11% after T6 (from 237 to 263 MPa). The hollow sprue improved the as-cast UTS by 10% and significantly enhanced the elongation after T6—from 1.0% to 1.9%—which is a 90% increase. This improvement is attributed to the combined effects of reduced pinhole defects and refined microstructure.
Fractographic analysis using scanning electron microscopy revealed distinct differences. The fracture surfaces of specimens cast with solid EPS foam sprue showed numerous large voids and inclusions (such as SiO2, carbides, and Al2O3), which acted as stress concentrators. In contrast, the hollow sprue castings displayed fewer inclusions, with shrinkage porosity being the dominant defect. The dimple morphology on the fracture surface indicated a mixed ductile-brittle fracture mode.
The elimination of foam decomposition products is critical for the quality of lost foam castings. The hollow paper tube sprue not only reduces the amount of foam, but also provides a direct channel for the escape of gases generated from the remaining foam pattern. This dual benefit leads to cleaner castings. In addition, the paper tube itself decomposes at a lower temperature but does not produce the same harmful byproducts as EPS foam. Therefore, employing a hollow sprue is a highly effective strategy for improving the integrity of aluminum alloy lost foam castings.
Conclusions
From the comprehensive experimental study, I draw the following conclusions:
- Using steel shot as the supporting medium in lost foam castings increases the cooling rate by nearly a factor of four compared to ceramic sand. This higher cooling rate refines the dendritic structure and reduces the secondary dendrite arm spacing by about 35%, leading to a 10% improvement in tensile strength in both as-cast and T6 conditions.
- The hollow paper tube sprue reduces the amount of EPS foam pattern by approximately 60%, thereby significantly decreasing the volume of pyrolysis gases. This results in fewer pinholes and a 90% improvement in elongation after T6 heat treatment.
- The combination of steel shot filling and hollow sprue design synergistically enhances the overall quality of ZL101 aluminum alloy lost foam castings, making the process more viable for high-performance applications.
These findings provide practical guidance for optimizing lost foam casting parameters to produce sound aluminum alloy components with superior mechanical properties.