In our foundry operations, specializing in the production of critical components such as wind turbine parts and injection molding machine frames using furan resin sand molding lines, we encountered a persistent defect in ductile iron castings. This defect manifested as a white spot or layer at the outer edge of the slant neck riser neck, positioned between the casting surface and the coating layer. The presence of this white spot, which could penetrate up to approximately 2 mm beneath the surface, proved resistant to removal during standard shot blasting. Consequently, additional grinding operations were required, leading to increased labor costs, higher overall casting expenses, and potential delays in meeting customer delivery schedules for our ductile iron components. This issue prompted a comprehensive investigation to understand its root cause and develop effective countermeasures, central to maintaining the quality and reliability of our ductile cast iron products.
The white spot defect was particularly observed on bearing housing castings produced with slant neck risers. Visual inspection revealed that the white material was not composed of sand grains from the molding material but appeared as a distinct, adherent layer. The coating layer in the corresponding area also exhibited similar whitish discoloration. Given that standard production parameters for molding, melting, pouring, and operator actions showed no significant deviations, the analysis focused on the interface between the molten ductile iron and the sand mold. The unique geometry of the slant neck riser was identified as a key factor, as it creates specific thermal conditions at its neck’s outer edge. We hypothesized that within the temperature range of approximately 1200–1400 °C, a reaction occurs at the surface of the ductile cast iron, leading to the formation of a silicon-rich layer that appears white. Since direct sampling from the casting surface was impractical for analysis, we designed and conducted a simulation experiment to replicate the defect under controlled conditions, allowing for detailed material analysis.
The primary objective of our experimental work was to recreate the white spot defect in a controlled setting to facilitate sampling and subsequent compositional analysis. This involved preparing materials to simulate the actual production environment for ductile iron castings. We utilized 20 kg of CQ607 coating from a standard supplier, one 150-size slant neck riser, a test block mold with dimensions of 600 mm × 300 mm × 75 mm, several ceramic tubes with a 40 mm diameter, and appropriate molding flasks. The experimental design positioned the slant neck riser on the side of the test block, connected via ceramic gates to simulate the feeding geometry encountered in the problematic bearing housings. The pouring temperature was strictly controlled between 1340 and 1350 °C to mirror the thermal conditions suspected of causing the defect in production ductile iron castings.
The chemical composition of the base ductile iron was carefully controlled to be representative of standard production grades. The target composition for the molten metal is summarized in the table below, which includes both the base iron and final treated iron specifications crucial for producing high-quality ductile cast iron.
| Element/Parameter | Control Range (Mass %) | Remarks |
|---|---|---|
| Carbon (Base Iron) | 3.40 – 3.45 | Critical for achieving proper graphite nodularity and matrix structure in ductile cast iron. |
| Silicon (Base Iron) | 2.83 – 2.93 | |
| Silicon (Final Casting) | 3.50 – 3.60 | |
| Manganese | < 0.025 | Kept low to minimize carbide formation. |
| Phosphorus | ≤ 0.040 | Limited to reduce embrittlement. |
| Sulfur (Before Treatment) | ≤ 0.025 | Low sulfur is essential for effective nodularization in ductile iron. |
| Sulfur (After Treatment) | 0.005 – 0.015 | |
| Magnesium | 0.035 – 0.055 | Nodularizing element. |
| Rare Earth | < 0.010 | Aids in nodularization and controls trace elements. |
| Carbon Equivalent (CE) | 4.56 – 4.65 | CE = %C + 0.33(%Si). Indicates castability and shrinkage tendency. |
| Antimony (Addition) | 0.005 | Used for pearlite stabilization if required. |
| Nickel (Addition) | 5.0 | For enhancing strength in certain ductile iron grades. |
The melting was conducted using a standard induction furnace, and the treatment involved a sandwich method for nodularization using a blend of Fe-Si-Mg alloys (70% N1 type and 30% N2 type), with a total addition of 0.9–1.2%. Inoculation was performed using BS-1A inoculant, added at the bottom of the pouring ladle with a quantity of 0.1–0.7%. The test block casting was produced, allowed to cool, and then shaken out. The white spot region was successfully reproduced on the test block adjacent to the riser neck. A sample containing the defect layer was sectioned for advanced metallographic and spectroscopic analysis.
Microscopic examination of the sample cross-section using Scanning Electron Microscopy (SEM) revealed the distinct morphology of the white layer. At low magnification, the layer appeared as a continuous band separating the bulk ductile iron matrix from the residual coating/mold material. Energy Dispersive X-ray Spectroscopy (EDS) was employed for elemental analysis. Area mapping clearly showed that the regions corresponding to the white spot were enriched in silicon (Si) and oxygen (O), while iron (Fe) was predominant in the adjacent casting material. Point analysis was conducted at several specific locations within the white layer to quantify the elemental composition. The results from four representative points are consolidated in the following table, highlighting the dominant presence of Si and O.
| Analysis Point | Chemical Element (Mass %) | Atomic % | Primary Constituents |
|---|---|---|---|
| Position 1 | C: 23.46 | 34.07 | Silicon Oxides (SiOx) |
| O: 46.92 | 51.17 | ||
| Si: 17.57 | 10.92 | ||
| S: 0.35 | 0.19 | ||
| Fe: 11.70 | 3.66 | ||
| Others: < 0.5 | < 0.5 | ||
| Position 2 | C: 10.81 | 17.45 | Silicon Oxides (SiOx) |
| O: 45.94 | 55.66 | ||
| Si: 34.40 | 23.74 | ||
| Fe: 8.09 | 2.81 | ||
| Mn, Ti, S: < 0.6 | < 0.3 | ||
| Position 3 | C: 11.27 | 17.73 | Silicon Oxides (SiOx) |
| O: 47.59 | 56.22 | ||
| Si: 36.15 | 24.33 | ||
| Fe: 4.43 | 1.50 | ||
| Ti: 0.55 | 0.22 | ||
| Position 4 | C: 19.56 | 29.69 | Silicon Oxides (SiOx) |
| O: 42.27 | 48.17 | ||
| Si: 29.75 | 19.31 | ||
| Fe: 7.29 | 2.38 | ||
| S, Mn, Ti, In: < 1.2 | < 0.5 | ||
The consistent high concentration of silicon and oxygen across all analysis points confirms that the white spot is primarily a layer of silicon oxide (SiOx). The presence of carbon is likely from residual organic material in the coating or from the atmosphere, and the iron signal indicates some minor penetration or interaction with the underlying ductile cast iron substrate. The thickness of this layer, estimated from cross-sectional observations and corroborated by the depth of the defect on production castings, ranges from several tens of micrometers up to 2 mm. The formation mechanism is intrinsically linked to the high temperature environment sustained by the slant neck riser geometry. Based on prior research and our findings, we propose the following sequence of chemical reactions occurring at the interface between the molten ductile iron and the mold coating at temperatures between 1200°C and 1400°C:
$$ \text{SiO}_2 (\text{from coating}) + (*) \rightarrow \text{SiO}_{(gas)} $$
$$ 2\text{SiO}_{(gas)} + \text{O}_2 \rightarrow 2\text{SiO}_2 $$
$$ 2\text{SiO}_{(gas)} \rightarrow \text{SiO}_2 + \text{Si} $$
Here, the term \( (*) \) represents reducing agents present in the system, which in the case of ductile cast iron include magnesium (from treatment), carbon, silicon, aluminum, and hydrogen. The initial reaction reduces silica (SiO2) from the coating to form gaseous silicon monoxide (SiO). This gaseous species can then re-oxidize in the presence of oxygen to form silica, or disproportionate to form both silica and elemental silicon. The net result is the transport and deposition of silicon-rich oxides onto the relatively cooler surface of the solidifying ductile iron casting, forming the observed white layer. The kinetics of these reactions are highly temperature-dependent, following an Arrhenius-type relationship, which explains why the defect is localized to areas with prolonged elevated temperatures, such as the riser neck edge:
$$ k = A e^{-E_a/(RT)} $$
where \( k \) is the rate constant, \( A \) is the pre-exponential factor, \( E_a \) is the activation energy, \( R \) is the gas constant, and \( T \) is the absolute temperature. The thermal profile around the riser neck can be modeled using heat transfer equations for solidification. For a simplified 1D analysis of heat conduction in the mold medium near the riser neck, Fourier’s law can be applied:
$$ q = -k_m \frac{dT}{dx} $$
where \( q \) is the heat flux, \( k_m \) is the thermal conductivity of the mold material, and \( \frac{dT}{dx} \) is the temperature gradient. The sustained high temperature at the interface, driven by the thermal mass of the riser feeding the ductile iron casting, provides the necessary energy for the aforementioned silica reduction reactions to proceed at a significant rate.
Having identified the root cause as a high-temperature interfacial reaction leading to silica deposition, our mitigation strategy focused on two main fronts: enhancing the coating’s resistance to such reactions and refining the application process to ensure coating integrity. The primary goal was to break the chain of reactions by providing a more stable barrier between the molten ductile iron and the mold sand. For the coating material itself, we implemented a significant reformulation. The key change was increasing the content of zircon flour (zirconium silicate, ZrSiO4) to 30% within the refractory aggregate blend. Zircon has a higher melting point (over 2500°C) and superior thermal stability compared to silica-based aggregates, reducing its tendency to undergo reduction reactions in contact with molten ductile iron. Furthermore, we optimized the particle size distribution of the aggregate to ensure better packing density and a consistent, effective coating layer with a penetration depth of 3–5 mm into the sand mold, creating a robust mechanical interlock. The target dry coating thickness on the mold surface was set between 0.35 mm and 0.50 mm to ensure adequate shielding.
The second aspect involved strict procedural controls for coating application and mold drying. We established a standardized brushing protocol with specific Baume degree (a measure of density) requirements for each consecutive coat to control slurry viscosity and solids content:
| Coating Layer | Baume Degree (°Be) Range | Purpose |
|---|---|---|
| First Coat | 38 – 40 | Ensure good penetration and base layer adhesion. |
| Second Coat | 55 – 60 | Build up coating thickness and density. |
| Third Coat | 45 – 50 | Final smoothing and sealing layer. |
Critically, for the high-risk area around the outer edge of the slant neck riser, a mandatory post-application step was instituted: directed drying using a gas torch (blowlamp) immediately after coating. This ensures rapid and complete drying of the coating in this zone, enhancing its green strength and reducing the likelihood of cracking or erosion when contacted by the molten ductile iron. Additionally, the overall mold drying time using hot air was standardized and monitored using a digital hygrometer to verify that the internal core and mold cavity reached a sufficiently low humidity level (typically below 1%) before pouring. Comprehensive training programs and detailed work instructions were rolled out for all personnel involved in the coating process to ensure consistent and correct execution.
The implementation of these combined measures yielded immediate and significant improvements. The occurrence of the white spot defect at the slant neck riser edges on production ductile iron castings was drastically reduced and effectively brought under control. Post-improvement inspections of bearing housings and similar castings confirmed the absence of the tenacious white layer, eliminating the need for extra grinding operations. This not only reduced direct labor costs associated with finishing but also minimized material waste and improved production flow efficiency. The success of this approach underscores the importance of a holistic view in solving casting defects, considering both material science (coating formulation) and process engineering (application and drying techniques). For ductile iron foundries utilizing slant neck risers in resin sand molds, the synergistic optimization of coating refractory composition towards more stable aggregates like zircon, coupled with stringent process controls for application, is a highly effective strategy for preventing the formation of silicon-oxide-based white spots.
In conclusion, our investigation into the white spot defect prevalent in ductile iron castings with slant neck risers successfully identified the mechanism as high-temperature facilitated reduction and re-deposition of silicon oxides. The slant neck riser geometry creates a localized thermal environment conducive to these reactions between the mold coating and elements within the ductile cast iron. Through systematic experimentation and analysis, we confirmed the composition of the defect layer. The countermeasures, focusing on upgrading the coating with higher zircon flour content for improved thermal inertia and implementing rigorous coating application protocols, have proven entirely effective in eliminating this defect. This resolution enhances the surface quality of the ductile iron castings, reduces post-casting processing costs, and ensures reliable delivery schedules. The principles derived from this study—understanding thermal conditions at metal-mold interfaces, selecting coatings for high-temperature stability, and enforcing precise process controls—are broadly applicable for enhancing quality and productivity in the production of ductile cast iron components across various industrial sectors.