In modern sand casting foundry operations, large thin-wall gray iron castings are widely used in industries such as machinery manufacturing and automotive engineering. However, due to uneven solidification rates at geometric hot spots, defects like shrinkage porosity and cavities frequently occur. To address this, external chills (metallic inserts) are strategically placed in the mold to accelerate local cooling and improve feeding conditions. The effectiveness of this chill technique heavily depends on the performance of the coating applied to both the sand mold surface and the chill surface. In this study, we aimed to develop a novel alcohol-based coating specifically for sand molds with external chills, targeting the common issues of sand adhesion, metal penetration, and difficulty in cleaning encountered in conventional coatings. Through systematic experiments, we investigated the effects of a composite modified suspending agent, refractory aggregate gradation, and binder composition on coating properties.
The coating was formulated using a bauxite-white corundum composite refractory aggregate. A lithium-based bentonite was first modified via a two-step “intercalation-grafting” method to prepare a high-performance suspending agent. A composite binder system consisting of phenolic resin and polyvinyl butyral (PVB) was employed, with industrial ethanol as the solvent. Orthogonal experiments were designed to optimize the formulation. The results showed that the optimal coating achieved a 24 h suspension rate of 95%, a gas evolution of 12.6 mL/g, a brushing performance grade of I, and excellent wear resistance. Production trials on a 30-ton machine tool bed casting demonstrated that the coating significantly reduced surface roughness (Ra 12.5 μm), sand adhesion area (<2%), eliminated visible gas pores (>2 mm), and greatly improved chill recovery. This study provides an effective solution for coating optimization in external chill quenching processes in sand casting foundry.
In the sand casting foundry industry, the demand for high-quality castings with complex geometries and thin walls is ever-increasing. The use of external chills is a common practice to control solidification sequence and prevent defects. However, the interface between the chill and the sand mold is a critical area where molten metal can penetrate, leading to sand inclusion and surface defects. Traditional water-based coatings suffer from slow drying and moisture absorption, increasing gas porosity. Alcohol-based coatings dry quickly but often exhibit poor adhesion on both sand and metal surfaces. Therefore, we focused on developing a coating that balances adhesion, thermal stability, and easy removal. Our approach involved modifying the bentonite suspending agent to enhance its dispersion in alcohol, optimizing the particle size distribution of refractory aggregates to achieve a dense packing structure, and using a phenolic resin-PVB composite binder to provide sufficient strength and low gas evolution.
The base materials used in this study included bauxite (250 mesh and 400 mesh, Al₂O₃ 86%), white corundum (250 mesh, Al₂O₃ 57%), calcium bentonite (200 mesh, montmorillonite content ≥70%, CEC=92 mmol/100g), thermosetting phenolic resin (carbon residue 55%, free phenol <1%), PVB (coating viscosity 210 s), and industrial ethanol (95%). Modification reagents included Li₂CO₃, CTAB, OTAC, MNTA, CMC, KH-550, and BTFA.
The composite modification of bentonite was carried out as follows: first, calcium bentonite was lithium-exchanged using Li₂CO₃. Then, intercalation was performed with MNTA, CMC, and CTAB to expand the interlayer spacing. Finally, grafting was achieved using KH-550 and BTFA. The modified bentonite was dried and ground. X-ray diffraction analysis showed that the interlayer spacing increased from 1.26 nm (natural) to 3.085 nm after modification, confirming successful intercalation. Scanning electron microscopy revealed a rougher surface with increased porosity, indicating better dispersibility. The swelling index increased from 5.8 mL/g to 20.1 mL/g, demonstrating enhanced swelling capacity in alcohol mediums.
The coating preparation involved mixing the refractory aggregates (bauxite and white corundum in a fixed mass ratio of 4:1) with the modified bentonite slurry, binder solution, and ethanol in a roller mixer for 2.5–3 hours. The final density was adjusted to (1.60±0.02) g/cm³. To systematically optimize the formulation, we selected three factors: the mass ratio of 250-mesh to 400-mesh bauxite (A), the amount of modified suspending agent (B, based on refractory aggregate mass), and the amount of composite binder (C, phenolic resin:PVB = 2.5:1). An L9(3³) orthogonal array was employed. The factor levels are shown in Table 1.
| Factor | Level 1 | Level 2 | Level 3 |
|---|---|---|---|
| A: m(250 mesh bauxite):m(400 mesh bauxite) | 60:20 | 53:27 | 40:40 |
| B: Suspending agent /% | 3 | 4 | 5 |
| C: Binder /% | 2 | 4 | 6 |
The nine experimental runs were conducted, and the properties of each coating, including conditional viscosity (by #6 cup), 24 h suspension rate, gas evolution, coating wear resistance, and brushing performance, were measured according to JB/T 9226–2008. Brushing performance was graded as I (uniform, no brush marks, thickness 0.5–1.2 mm on both sand and metal), II (slightly non-uniform, fine brush marks), or III (uneven thickness or obvious brush marks). For quantitative analysis, grades were assigned scores: I=1, II=2, III=3. The orthogonal test results and range analysis are presented in Tables 2 and 3.
| Run | Conditional Viscosity / s | 24 h Suspension / % | Gas Evolution / (mL/g) | Wear Resistance / g | Brushing Grade |
|---|---|---|---|---|---|
| A1B1C1 | 5.0 | 88 | 11.0 | 0.067 | III |
| A1B2C2 | 6.5 | 94 | 13.4 | 0.056 | I |
| A1B3C3 | 7.3 | 91 | 15.9 | 0.052 | II |
| A2B1C2 | 6.2 | 90 | 13.8 | 0.063 | I |
| A2B2C3 | 8.3 | 93 | 14.2 | 0.049 | III |
| A2B3C1 | 5.9 | 95 | 12.6 | 0.069 | I |
| A3B1C3 | 8.0 | 89 | 13.5 | 0.050 | III |
| A3B2C1 | 6.0 | 94 | 12.0 | 0.070 | II |
| A3B3C2 | 7.1 | 93 | 13.2 | 0.058 | II |
| Performance Indicator | Mean | A | B | C | Order of Influence |
|---|---|---|---|---|---|
| Conditional Viscosity / s | k1 | 6.27 | 6.40 | 5.63 | C > B > A |
| k2 | 6.80 | 6.93 | 6.60 | ||
| k3 | 7.03 | 6.77 | 7.87 | ||
| Range R | 0.76 | 0.53 | 2.24 | ||
| 24 h Suspension / % | k1 | 91.0 | 89.0 | 92.3 | B > A > C |
| k2 | 92.7 | 93.7 | 92.3 | ||
| k3 | 92.0 | 93.0 | 91.0 | ||
| Range R | 1.7 | 4.7 | 1.3 | ||
| Gas Evolution / (mL/g) | k1 | 13.43 | 12.77 | 11.87 | C > B > A |
| k2 | 13.53 | 13.20 | 13.47 | ||
| k3 | 12.90 | 13.90 | 14.53 | ||
| Range R | 0.63 | 1.13 | 2.66 | ||
| Wear Resistance / g | k1 | 0.0583 | 0.0600 | 0.0687 | C > A > B |
| k2 | 0.0603 | 0.0583 | 0.0590 | ||
| k3 | 0.0593 | 0.0597 | 0.0503 | ||
| Range R | 0.0020 | 0.0017 | 0.0184 | ||
| Brushing Grade (score) | k1 | 2.00 | 2.33 | 2.00 | C > A ≈ B |
| k2 | 1.67 | 2.00 | 1.33 | ||
| k3 | 2.33 | 1.67 | 2.67 | ||
| Range R | 0.66 | 0.66 | 1.34 |
To determine the optimal formulation considering all performance indicators simultaneously, we applied a weight normalization method. The weights were assigned as: conditional viscosity 10%, 24 h suspension 25%, gas evolution 20%, wear resistance 10%, and brushing performance 35%. After normalizing the data (with suspension as positive indicator and others as negative indicators), the composite scores for each run were calculated. The results are shown in Table 4. The highest score (20.28) was obtained for run A2B3C1 (A: 53:27 bauxite ratio, B: 5% suspending agent, C: 2% binder). This formulation achieved low viscosity (5.9 s), high suspension (95%), acceptable gas evolution (12.6 mL/g), good wear resistance (0.069 g), and excellent brushing performance (Grade I).
| Run | Normalized Score |
|---|---|
| A1B1C1 | 18.24 |
| A1B2C2 | 19.81 |
| A1B3C3 | 18.13 |
| A2B1C2 | 18.76 |
| A2B2C3 | 18.52 |
| A2B3C1 | 20.28 |
| A3B1C3 | 17.70 |
| A3B2C1 | 19.79 |
| A3B3C2 | 19.44 |
The optimal coating was then prepared and tested in a sand casting foundry production environment. The mold was a furan resin-bonded sand mold with external chills (single chill area 25 cm × 15 cm, total chill surface area about 1.2 m²). The casting was an HT300 machine tool bed weighing 30 tons. The coating was brushed twice to a total thickness of 0.7 mm. It exhibited excellent brushing performance and ignited within 5 s after application. The pouring temperature was 1420 °C. After cooling and shakeout, the coated surface was clean, with easy separation of the chills from the casting. Table 5 compares the casting quality before and after applying the optimized coating. The surface roughness (Ra) decreased from 25.0 μm to 12.5 μm, sand adhesion area dropped from 8.5% to less than 2%, cleaning time reduced from 45 min to 25 min, chill recovery rate increased from 65% to over 90%, and no gas pores larger than 2 mm were observed.
| Property | Before Optimization | After Optimization |
|---|---|---|
| Surface Roughness Ra / μm | 25.0 | 12.5 |
| Sand Adhesion Area / % | 8.5 | <2 |
| Cleaning Time / min | 45 | 25 |
| Chill Recovery Rate / % | 65 | >90 |
| Gas Pore Defect Rate (Φ>2 mm) | Present | None |
In conclusion, we have successfully developed an alcohol-based casting coating for sand molds with external chills that addresses the key challenges in sand casting foundry operations. By using a composite modified bentonite suspending agent (via intercalation-grafting), optimizing the bauxite-white corundum aggregate gradation (53:27 ratio of 250 mesh to 400 mesh), and employing a low dosage (2%) of phenolic resin-PVB composite binder, the coating achieved excellent suspension (95%), low gas evolution (12.6 mL/g), good wear resistance, and superior brushing performance on both sand and metal surfaces. Production validation in a sand casting foundry demonstrated significant improvements in casting surface quality, reduced cleaning time, and enhanced chill reusability. This work provides a practical solution for optimizing coatings in the external chill process, contributing to higher efficiency and quality in sand casting foundry industry.