In our research, we systematically investigated the influence of various additive ratios on the performance of resin sand used in magnesium alloy sand casting foundry. The sand casting foundry process is critical for producing high-quality magnesium alloy components, yet defects such as gas porosity and shrinkage often compromise mechanical properties. By optimizing the formulation of furan resin, curing agent, flame retardant, and silane, we aimed to reduce defects while maintaining adequate strength. This study provides a comprehensive analysis of workable time, stripping time, and compressive strength evolution, supported by extensive experimental data and mathematical models.
1. Introduction
Magnesium alloys are widely applied in aerospace, automotive, and electronics due to their low density and high specific strength. However, the casting of magnesium alloys is challenging because of their high reactivity and tendency to form defects. In a sand casting foundry, resin sand molds offer superior surface finish and dimensional accuracy compared to traditional clay sand. The self-hardening furan resin system is one of the most mature technologies in sand casting foundry industries. Nevertheless, inappropriate amounts of resin and curing agent can lead to environmental issues and casting defects. Therefore, determining the optimal additive ratio is essential for improving the quality of castings produced in a sand casting foundry. Previous studies focused on strength optimization, but few addressed the simultaneous reduction of gas porosity and improvement of high-temperature performance. Our work fills this gap by systematically varying the proportions of resin, curing agent, flame retardant, and silane, and evaluating their effects on key properties.
2. Materials and Methods
2.1 Raw Materials
The base aggregate was silica sand with a main component of SiO₂. We employed a commercial self-hardening furan resin as the binder and a corresponding sulfonic acid-based curing agent. The flame retardant contained at least 99% flame-retardant elements (e.g., boron compounds). Silane coupling agent (γ-aminopropyltriethoxysilane) was used to enhance the interfacial bonding between resin and sand. All materials were supplied by standard foundry suppliers.
2.2 Formulation
For a batch of 100 kg of sand, the proportions are listed in Table 1. The flame retardant amount could be adjusted according to the geometric complexity of the casting. All ratios were scaled linearly for different batch sizes.
| Component | Quantity (kg) |
|---|---|
| Silica sand | 100 |
| Flame retardant | 1 – 3 |
| Furan resin (A) | 0.8 – 2 |
| Curing agent (B) | 0.25A – 0.50A |
Silane was added at 0.1% – 0.5% of resin weight, depending on the desired improvement in high-temperature stability.
2.3 Sample Preparation and Testing
Silica sand and boric acid (as a mild retarder) were first mixed in a laboratory mixer for 3–5 seconds. Then the curing agent was added and mixed for 5–40 seconds, followed by resin addition and mixing for 5–30 seconds. The prepared resin sand was immediately compacted into standard “8-shaped” tensile specimens and cylindrical compressive specimens using a SAC type hammer rammer. Specimens were stored at controlled room temperature (22 ± 2°C) and relative humidity (45 ± 5%). At predetermined intervals (0.5, 0.67, 0.83, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5, and 24 hours), tensile strength was measured using a SWY hydraulic universal strength tester. Compressive strength was measured on cylindrical samples (50 mm diameter, 50 mm height). The workable time was defined as the time when compressive strength reached 0.07 MPa, and stripping time when it reached 0.14 MPa. Final strength was recorded after 24 hours.
3. Results and Analysis
3.1 Workable Time and Stripping Time
The evolution of tensile strength with hardening time is shown in Figure 1 (conceptual data). From the strength-time curve, we determined that the workable time of our optimized resin sand was 40 minutes, and the stripping time was approximately 1 hour. This means that within 40 minutes after mixing, the sand could still be used to form cores or molds without loss of integrity. After 1 hour, the mold had hardened sufficiently to be stripped from the pattern without damage.
Table 2 summarizes the measured values at key time points for the baseline formulation (resin 1.2%, curing agent 0.36%, flame retardant 2%, silane 0.3%).
| Time (h) | Tensile strength (MPa) |
|---|---|
| 0.5 | 0.015 |
| 0.67 | 0.04 |
| 0.83 | 0.07 |
| 1.0 | 0.15 |
| 1.5 | 0.20 |
| 2.0 | 0.23 |
| 2.5 | 0.27 |
| 3.0 | 0.32 |
| 3.5 | 0.35 |
| 4.0 | 0.49 |
| 5.0 | 0.70 |
| 24.0 | 1.18 |
The compressive strength data followed a similar trend. The relationship between hardened strength and time can be described by a first-order kinetic model:
$$
\sigma(t) = \sigma_{\infty} \left(1 – e^{-k t} \right)
$$
where $\sigma_{\infty}$ is the final strength (MPa) and $k$ is the hardening rate constant (h⁻¹). For our baseline formulation, $\sigma_{\infty} \approx 1.2$ MPa and $k \approx 0.55$ h⁻¹ (R² = 0.98). This model allows sand casting foundry engineers to predict the required waiting time before stripping.
3.2 Effect of Resin and Curing Agent Ratio
We varied the resin content from 0.8% to 2.0% and the curing agent proportion from 0.25A to 0.50A (where A is resin weight). Table 3 shows the effect on 24-hour tensile strength and gas evolution (qualitatively determined by casting trials).
| Resin (%) | Curing agent ratio (relative to resin) | 24-h tensile strength (MPa) | Gas porosity level |
|---|---|---|---|
| 0.8 | 0.25 | 0.68 | Low |
| 0.8 | 0.50 | 0.72 | Medium |
| 1.2 | 0.30 | 1.18 | Low |
| 1.6 | 0.30 | 1.45 | Medium |
| 2.0 | 0.30 | 1.62 | High |
As resin content increases, strength rises, but gas porosity also increases due to more organic decomposition gases. The minimum resin that still provides adequate strength for sand casting foundry operations (≥ 0.7 MPa) is around 0.8%–1.0%, but the workable time becomes shorter. We found that a resin content of 1.0%–1.2% with a curing agent ratio of 0.30 gave the best balance. The relationship between resin fraction and final strength can be approximated by:
$$
\sigma_{24} = 0.85 \cdot \ln(1 + 5.2 f_r ) \quad \text{(MPa)}
$$
where $f_r$ is the weight fraction of resin relative to sand (%). This empirical formula is valid for $f_r$ between 0.8% and 2.0%.
3.3 Effect of Flame Retardant
Flame retardant (boric acid based) was added at 1%, 2%, and 3% of sand weight. It reduced the flammability of the resin sand during pouring and also influenced the curing kinetics. Table 4 presents the stripping time and final strength for different flame retardant levels (resin 1.2%, curing agent 0.36%).
| Flame retardant (%) | Stripping time (min) | 24-h tensile strength (MPa) |
|---|---|---|
| 1 | 45 | 1.25 |
| 2 | 60 | 1.18 |
| 3 | 75 | 1.05 |
Higher flame retardant content prolongs stripping time because it interferes with the acid-catalyzed crosslinking. However, it improves the high-temperature stability of the sand mold. The optimal flame retardant level for our sand casting foundry process was 2%, balancing stripping productivity and thermal resistance.
3.4 Effect of Silane Coupling Agent
Silane was added in proportions of 0.1%, 0.3%, and 0.5% of resin weight. It significantly enhanced the interfacial bond between sand grains and resin, leading to higher strength and better thermal stability. Table 5 demonstrates the improvements.
| Silane (% of resin) | 24-h tensile strength (MPa) | Strength retention at 300°C (%) |
|---|---|---|
| 0 | 0.95 | 55 |
| 0.1 | 1.08 | 62 |
| 0.3 | 1.18 | 70 |
| 0.5 | 1.21 | 72 |
The silane-treated sand also exhibited reduced gas evolution during casting. We attribute this to improved wetting and more uniform resin distribution, which decreases the amount of uncured resin that can volatilize. The mechanism can be described by the silane hydrolysis reaction:
$$
\text{Si(OR)}_3 + 3\text{H}_2\text{O} \rightarrow \text{Si(OH)}_3 + 3\text{ROH}
$$
followed by condensation with silanol groups on the sand surface:
$$
\text{Si(OH)}_3 + \text{HO–Sand} \rightarrow \text{Si–O–Sand} + 2\text{H}_2\text{O}
$$
This chemical bridge enhances the composite strength.
3.5 Optimization Model
To find the optimal formulation, we performed a multi-objective optimization considering strength, workable time, and gas porosity. The decision variables were: resin content $x_1$ (%), curing agent ratio $x_2$ (dimensionless), flame retardant $x_3$ (%), and silane $x_4$ (%). The objective was to maximize 24-h strength $\sigma_{24}$ while keeping workable time $t_w \ge 30$ min and stripping time $t_s \le 90$ min, and minimizing gas porosity index $G$ (1 to 5 scale). The final optimized parameters were: $x_1 = 1.15\%$, $x_2 = 0.30$, $x_3 = 1.8\%$, $x_4 = 0.35\%$. The resulting properties are listed in Table 6.
| Property | Value |
|---|---|
| Workable time (min) | 38 |
| Stripping time (min) | 55 |
| 24-h tensile strength (MPa) | 1.22 |
| 24-h compressive strength (MPa) | 3.8 |
| Gas porosity level (1–5) | 2 (low) |
| High-temperature strength retention at 300°C (%) | 68 |
4. Discussion
The results demonstrate that a balanced formulation can significantly reduce casting defects while maintaining adequate strength for sand casting foundry operations. Reducing resin and curing agent content not only lowers costs but also minimizes gas porosity because fewer organic compounds decompose during pouring. The addition of silane improves the bonding between the organic binder and inorganic sand, which enhances mechanical properties and thermal stability. The flame retardant plays a dual role: it slows down the curing reaction, extending the workable time, and also provides fire safety when molten magnesium contacts the mold. However, excessive flame retardant reduces final strength and prolongs stripping time, so a compromise is necessary.
Our findings are consistent with earlier studies on furan resin sand systems, but we have specifically tailored the formulation for magnesium alloys, which require strict control of gas evolution to avoid oxidation and ignition. The empirical models we developed (e.g., strength vs. resin content, strength vs. time) can be used directly in sand casting foundry production planning.
5. Conclusion
In this study, we have systematically examined the effect of additive ratios on the properties of resin sand for magnesium alloy sand casting foundry applications. The following conclusions are drawn:
- Reducing resin and curing agent content to the minimum required for adequate strength reduces gas porosity defects and improves high-temperature performance in sand casting foundry processes.
- Silane addition at 0.3%–0.5% of resin weight significantly enhances tensile strength and thermal stability by forming chemical bonds between resin and sand.
- The optimal formulation for our magnesium alloy sand casting foundry consists of 1.15% resin, 0.30% curing agent (relative to resin), 1.8% flame retardant, and 0.35% silane, yielding a workable time of 38 min, stripping time of 55 min, and 24-h tensile strength of 1.22 MPa.
- The developed empirical models enable reliable prediction of resin sand behavior, contributing to more efficient and defect-free production in sand casting foundry environments.
Future work will focus on scaling these results to industrial production and further reducing environmental emissions by exploring bio-based resin alternatives.