In our work on high-performance ZA27 zinc-based alloy, we have focused extensively on the sand casting foundry process. ZA27 is known for its fast melting rate, low energy consumption, and non-polluting characteristics. Its casting temperature is low (around 540°C), requiring little flux and offering excellent fluidity, making it highly suitable for sand casting foundry operations. We have employed various casting methods, but the sand casting foundry remains the most versatile for producing complex geometric parts. The alloy exhibits superior mechanical properties and wear resistance, enabling it to replace bronze, brass, aluminum alloys, cast iron, and ductile iron in applications such as sliding bearings, thrust bearings, washers, flanged bushings, turbine components, and wear-resistant molds. With a density of approximately 5 g/cm³ and a lower cost than tin bronze, ZA27 is an ideal substitute for copper-based wear-resistant materials.
However, due to the wide solidification range of ZA27 (109°C) and the significant density difference between solid and liquid phases, combined with surface tension effects, slow cooling in sand casting foundry processes often leads to bottom shrinkage (also called inverse shrinkage). Controlling such defects, especially in large cross-section castings, is critical to product quality and application scope. The chemical composition of ZA27 is given in Table 1.
| Element | Al | Cu | Mg | Fe | Zn |
|---|---|---|---|---|---|
| Content (wt%) | 25.0–28.0 | 2.0–2.5 | 0.01–0.02 | ≤0.10 | Balance |
1. Melting and Pouring
For industrial batch production, we used crucible furnaces for alloy melting. The charge sequence was: Al-Cu master alloy, aluminum ingots, and zinc ingots, all loaded into the crucible for heating. Magnesium was added using a bell plunger method. After the melt reached the specified temperature, degassing, slag removal, and pouring were performed. Depending on the workpiece cross-section, we employed various sand casting foundry gating systems: single large gates, multiple risers, insulated risers, and pressure gates, along with the occasional use of chills.
2. Mechanical Properties of ZA27
Table 2 compares the mechanical properties of ZA27 with SAE660 bronze. The tensile and yield strengths of ZA27 are nearly twice those of SAE660, and its impact toughness is significantly higher. However, the hardness of ZA27 is much greater, which is advantageous for wear applications.
| Property | ZA27 | SAE660 |
|---|---|---|
| Tensile Strength (MPa) | 400–440 | 205–260 |
| Yield Strength (MPa) | 315–370 | 115–145 |
| Elongation (%) | 3–9 | 12–20 |
| Hardness (HB) | 110–130 | 60–70 |
| Impact Toughness (J) | 50±7 | 8 |
3. Wear Test Results
Wear tests were conducted on an AMSLER machine under sliding friction conditions with lubrication. The mating material was 45# steel. Results are shown in Table 3.
| Specimen | Test No. | Time (h) | Cumulative Wear (mg) | Linear Wear Rate (mg/m × 10⁻³) |
|---|---|---|---|---|
| ZA27 | 2 | 20 | 0.2 (upper) / 7.2 (lower) | 0.007 / 0.272 |
| 4 | 40 | 2.6 / 8.8 | 0.049 / 0.166 | |
| ZQSn6-6-3 | 2 | 10 | 3.2 / 2.2 | 0.242 / 0.166 |
| 3 | 20 | 4.4 / 2.6 | 0.166 / 0.098 |
The data confirm that ZA27 exhibits markedly lower wear loss compared to the bronze alloy, especially under prolonged testing. This makes it highly attractive for sand casting foundry production of wear parts.
4. Casting Process Discussion
Through testing, we determined that the pattern shrinkage of ZA27 is approximately 1.3%. This means thin-walled castings (3–10 mm) generally have good casting performance. In sand casting foundry practice, if chills are used during pouring, thin sections rarely exhibit shrinkage or inverse shrinkage, and they often show excellent pressure tightness. However, for large cross-section castings, inverse shrinkage becomes problematic.
The bottom shrinkage in ZA27 occurs near the end of solidification, when the solid fraction reaches about 75–90%. At that stage, shrinkage manifests as dispersed porosity. This porosity severely degrades tensile strength, impact toughness, and elongation. The mechanism is related to the density difference between the aluminum-rich primary phase and the remaining liquid phase. The density difference can be as high as 0.83 g/cm³. The primary aluminum-rich dendrites float upward, with initial flotation speeds of 5–21 mm/s. When the solid fraction reaches about 50%, dendrite flotation is hindered, and it stops at about 70% solid fraction.
Solidification proceeds from the top (aluminum-rich, higher melting point) to the bottom (zinc-rich, lower melting point). The zinc-rich liquid concentrates at the bottom and solidifies last. During solidification, volume contraction occurs, and the residual zinc-rich liquid is drawn by surface tension into the inter-dendritic spaces of the aluminum-rich phase, causing shrinkage cavities or porosity at the bottom.
Therefore, increasing cooling rate, reducing pouring temperature, and properly placing risers are essential to avoid shrinkage. In our initial attempts, single large risers proved ineffective because the bottom shrinkage occurs after most of the casting has already solidified, leaving no feeding path.
Based on extensive sand casting foundry experiments, we identified the following effective measures for large-section ZA27 castings:
- Orient important surfaces facing upward in the mold.
- Use pressure gates or multiple gates to disperse the hot spot at the gate.
- Apply chills.
- Use insulated risers or riser sleeves.
- Place side risers connected to the ingate, close to the region requiring feeding.
- The modulus (V/A) of the riser should be at least 1.2 times that of the section being fed.
- Adjust gating system ratios: sprue : runner : ingate = 1 : 5 : 3.
- Consider modification treatment of the alloy.
- Keep pouring temperature as low as possible.
- Optimize casting design for uniform wall thickness and avoid sharp corners.
We can quantify the feeding condition using the modulus concept. The modulus M is defined as:
$$M = \frac{V}{A}$$
where V is the volume and A is the cooling surface area. For a riser to feed a section, its modulus must satisfy:
$$M_{\text{riser}} \ge 1.2 \cdot M_{\text{section}}$$
Additionally, the solidification time t is proportional to the square of the modulus:
$$t_f \propto M^2$$
Thus, a riser with a higher modulus solidifies later, allowing it to feed the casting. The critical condition for avoiding inverse shrinkage can be expressed in terms of the thermal gradient G and solidification rate R. Since inverse shrinkage is driven by density inversion, one can define a Rayleigh number:
$$\text{Ra} = \frac{g \beta \Delta T L^3}{\alpha \nu}$$
where g is gravity, β is thermal expansion coefficient (or density inversion coefficient), ΔT is temperature difference, L is characteristic length, α is thermal diffusivity, and ν is kinematic viscosity. In our sand casting foundry, we aim to minimize Ra by increasing cooling rate (reducing L) and lowering ΔT.
The density difference between the primary solid and liquid is approximately:
$$\Delta \rho = \rho_{\text{liquid}} – \rho_{\text{solid}} \approx 0.83 \, \text{g/cm}^3$$
This drives the flotation velocity u, which can be estimated by Stokes’ law:
$$u = \frac{2 g (\Delta \rho) r^2}{9 \mu}$$
where r is the dendrite tip radius and μ is the dynamic viscosity. To prevent inverse shrinkage, we need to ensure that the solidification front moves faster than the flotation velocity, so that dendrites are trapped before they can segregate. This is achieved by increasing the cooling rate, which increases the solidification rate R. The relationship between cooling rate and secondary dendrite arm spacing (SDAS) is:
$$\text{SDAS} = k \cdot \left( \frac{dT}{dt} \right)^{-n}$$
where k and n are material constants. Finer SDAS improves mechanical properties and reduces porosity.
In our sand casting foundry trials, we observed that using a combination of chills and insulated risers reduced bottom shrinkage porosity from over 5% (by area) to less than 0.5% in sections up to 100 mm thick. Table 4 summarizes the effect of different process parameters on shrinkage severity.
| Parameter | Condition | Shrinkage Area (%) | Remarks |
|---|---|---|---|
| Pouring temperature | 540°C | 4.2 | Moderate |
| Pouring temperature | 520°C | 1.8 | Lower temperature reduces shrinkage |
| Gating system | Single large riser | 6.1 | Ineffective feeding |
| Gating system | Multiple gates + insulated riser | 0.9 | Good feeding |
| Chill application | No chills | 5.5 | Slow cooling |
| Chill application | Copper chills on bottom | 0.3 | Rapid directional solidification |
| Mold material | Green sand | 3.2 | Standard |
| Mold material | CO₂ sand with chill | 0.5 | Enhanced cooling |
The key to successful sand casting foundry production of large ZA27 components is to control the solidification direction. We want the top to solidify first, so that the zinc-rich liquid is not trapped at the bottom. This can be achieved by placing chills at the top or using insulating materials at the bottom. Another approach is to modify the alloy composition within the specified range to reduce the density inversion. The density of solid aluminum-rich phase is about 3.2 g/cm³, while the liquid zinc-rich phase is about 4.0 g/cm³. The difference can be expressed as:
$$\Delta \rho = \rho_{\text{liq}} – \rho_{\text{sol}} = 4.0 – 3.2 = 0.8 \, \text{g/cm}^3$$
If we increase the aluminum content slightly (e.g., to 27.5%), the primary phase density decreases further, worsening the problem. Conversely, increasing zinc content reduces the density difference but may affect mechanical properties. We found that maintaining Al at 25–26% and Cu at 2.0–2.2% gave the best balance.
Thermal analysis of the solidification process reveals that the fraction solid f_s as a function of temperature can be approximated by the Scheil equation:
$$f_s = 1 – \left( \frac{T_m – T}{T_m – T_l} \right)^{\frac{1}{k-1}}$$
where T_m is the melting point of pure zinc (or liquidus), T_l is the liquidus temperature of the alloy (~540°C), and k is the partition coefficient. For ZA27, the solidification range is 109°C, so T_s ≈ 431°C. The inverse shrinkage occurs when f_s is between 0.75 and 0.90. At this stage, the remaining liquid is highly enriched in zinc and has low viscosity. The feeding distance d_f can be estimated by:
$$d_f = \frac{\Delta P \cdot r_p^2}{8 \mu v}$$
where ΔP is the pressure gradient, r_p is the pore radius, μ is liquid viscosity, and v is the solidification velocity. In practice, ensuring that the riser is within the feeding distance is critical.
Our sand casting foundry experiments also examined the effect of modification. Adding a small amount of grain refiner (e.g., Al-Ti-B master alloy) reduced the dendrite arm spacing and improved feeding characteristics. The modified structure showed a 30% reduction in porosity. The typical SDAS without modification was around 80 μm, while with modification it dropped to 50 μm, as shown in Table 5.
| Condition | SDAS (μm) | Porosity (%) |
|---|---|---|
| Unmodified | 80 | 4.5 |
| Modified (0.2% Al-Ti-B) | 50 | 3.1 |
Furthermore, we developed a feeding rule for sand casting foundry of ZA27. The riser volume V_r should satisfy:
$$V_r \ge \frac{\beta}{1 – \beta} V_c$$
where β is the solidification shrinkage of the alloy (about 3–4% for ZA27), and V_c is the volume of the casting section being fed. For large castings, we often use multiple risers with volumes exceeding this minimum.
One of the most successful approaches was using a combination of a pressure gate and a top riser with an insulating sleeve. The pressure gate provides additional pressure to push liquid into the inter-dendritic spaces, while the insulated riser ensures that liquid is available until the very end of solidification. Figure 1 illustrates a typical sand casting foundry mold setup for a ZA27 bushing.
In summary, through systematic study of the sand casting foundry process for ZA27, we have established robust guidelines to produce high-quality castings. The key factors are: control of pouring temperature (optimal 520–530°C), use of chills to promote directional solidification, proper riser design with modulus ratio >1.2, modification treatment to refine microstructure, and optimized gating system ratios. By implementing these measures, we successfully reduced bottom shrinkage defects from unacceptable levels to below 1% area, achieving mechanical properties comparable to premium bronze alloys at significantly lower cost.
5. Conclusions
- High-performance ZA27 zinc alloy can serve as a low-cost, ideal substitute for wear-resistant copper and aluminum alloys when processed under appropriate sand casting foundry conditions.
- Thin-walled ZA27 castings (3–10 mm) are relatively straightforward to produce with high quality using sand casting foundry methods.
- For large cross-section ZA27 castings, inverse shrinkage (bottom porosity) is a major defect that must be controlled through composition design, melting practice, mold design, and pouring control in the sand casting foundry environment.
- The effective measures include: use of chills, insulated risers, multiple gates, low pouring temperature (520–530°C), modification treatment, and proper riser modulus (>1.2 times that of the section).
- Quantitative relationships such as the modulus criterion, Rayleigh number minimization, and Scheil equation provide a theoretical basis for process optimization in sand casting foundry.