In the production of large and medium-sized ductile iron castings, the use of risers with exothermic insulating covering agents is crucial to prevent shrinkage defects. Ductile iron castings exhibit a mushy solidification pattern, which narrows feeding channels and may require additional measures to counteract solidification shrinkage, despite the self-feeding effect from graphite expansion. This study focuses on developing and testing a novel exothermic insulating covering agent for risers in ductile iron castings, emphasizing its feeding efficiency, environmental friendliness, and cost-effectiveness. We propose a new metric, the relative feeding efficiency, to better characterize riser performance, and compare our product with those from renowned international and domestic companies through practical production trials.
Ductile iron castings are widely used in high-end equipment due to their excellent mechanical properties, but ensuring soundness in large sections remains challenging. Traditional covering agents often contain high levels of fluorides and carbon, which can lead to issues such as coarse graphite formation and difficulties in recycling riser metal. For instance, fluorides like cryolite (Na3AlF6) act as catalysts in exothermic reactions but may cause graphite coarsening at the riser-contact zone in ductile iron castings. Carbonaceous materials, while aiding insulation, can affect the chemical composition of remelted metal. To address these problems, our research aims to create a low-fluoride, low-carbon covering agent that balances ignition temperature, ignition time, heat release rate, and overall cost. The novel agent utilizes fly ash cenospheres and perlite as primary insulating materials, aluminum powder and industrial iron oxide as exothermic components, nitrate-based mixtures as oxidizers, and minimal fluoride salts and carbonized rice hulls as additives. This composition ensures that the fluoride content is below 1%, carbon content under 5.5%, and total heat release exceeds 170 J/g, making it suitable for ductile iron castings.
To evaluate the feeding performance, we introduced the concept of relative feeding efficiency, which provides a more accurate representation than the traditional feeding efficiency. The traditional feeding efficiency, denoted as η, is defined as the ratio of the feeding liquid volume to the total riser volume: $$η = \frac{M_{\text{feed}}}{M_{\text{riser}}} \times 100\% = \frac{V_{\text{feed}}}{V_{\text{riser}}} \times 100\%$$ where \(M_{\text{feed}}\) and \(V_{\text{feed}}\) are the mass and volume of the feeding liquid, and \(M_{\text{riser}}\) and \(V_{\text{riser}}\) are the total mass and volume of the riser. However, this metric only considers the total shrinkage volume and ignores the shape of the shrinkage cavity, particularly the safety height of the riser. In practice, a higher safety height indicates better feeding capability, as it allows for potential reductions in riser size. Therefore, we define the relative feeding efficiency, η_relative, as the ratio of the shrinkage volume to the upper riser volume corresponding to the shrinkage depth: $$η_{\text{relative}} = \frac{V_{\text{feed}}}{V_{\text{upper}}} \times 100\% = \frac{V_{\text{feed}}}{V_{\text{riser}} – V_{\text{lower}}} \times 100\%$$ Here, \(V_{\text{upper}}\) is the volume of the riser portion above the safety height \(H\), and \(V_{\text{lower}}\) is the volume below it. This new metric accounts for both the total shrinkage volume and its shape, offering a comprehensive assessment of riser performance in ductile iron castings.
Our testing methodology involved practical production scenarios where identical risers on the same casting were treated with different covering agents. For each trial, we poured molten iron into molds with multiple risers, applied equal amounts of covering agents at half the riser height, and allowed the castings to cool. After cooling, we cut the risers, measured shrinkage volumes using fine quartz sand, and performed longitudinal sectioning to determine safety heights and upper volumes. This approach enabled direct comparison of the relative feeding efficiency for various covering agents in ductile iron castings.
In the first application, we tested the covering agents on rail castings, which are components for ship lock gates in the Three Gorges Dam, made of QT400-15 ductile iron with a weight of approximately 980 kg per piece. Each casting had two identical elliptical open risers, insulated with cenosphere sleeves on the sides and covered with either the novel agent or a product from a domestic知名 company. The riser dimensions were 230 mm × 330 mm × 400 mm, and 1.0 kg of covering agent was used per riser. Observations showed that the novel covering agent ignited faster, expanded more, and spread evenly, forming a loose and平整 layer without刺激性 odors. In contrast, the domestic product had poorer spreadability and an uneven surface. The combustion intensity and ignition times were similar for both. The feeding efficiency results are summarized in Table 1.
| Covering Agent Type | Total Riser Height (mm) | Safety Height (mm) | Max Shrinkage Depth (mm) | Shrinkage Volume (dm³) | Upper Riser Volume (dm³) | Feeding Efficiency η (%) | Relative Feeding Efficiency η_relative (%) |
|---|---|---|---|---|---|---|---|
| Novel Covering Agent | 270 | 160 | 110 | 3.55 | 7.10 | 20.4 | 50.0 |
| Domestic知名 Company Product | 270 | 175 | 95 | 2.96 | 6.13 | 17.0 | 48.3 |
As shown in Table 1, the novel covering agent achieved a higher traditional feeding efficiency of 20.4% compared to 17.0% for the domestic product. More importantly, the relative feeding efficiency was 50.0% for the novel agent versus 48.3% for the competitor, indicating superior performance in ductile iron castings. The faster ignition and better expansion of the novel agent contributed to its effectiveness, particularly in the early stages of solidification where quick heat supplementation is critical for ductile iron castings.
In the second application, we evaluated pad iron castings made of QT500-7 ductile iron, weighing 730 kg each. These were produced using water glass sand molds and furan resin cores, with three circular open risers of 100 mm diameter and 325–355 mm height. The riser sides were insulated with cenosphere sleeves, and the tops were covered with either the novel agent or a product from an international renowned company. The results revealed that the novel agent produced a larger shrinkage volume with a flat bottom and a safety height of 55 mm, whereas the international product resulted in a smaller shrinkage volume of 45 mm depth and an irregular shape, including hidden shrinkage cavities. Although we could not calculate the exact feeding efficiency due to the hidden cavities, the relative feeding efficiency was clearly higher for the novel agent. Metallographic examination of the riser roots showed that both agents maintained graphite spheroidization rates of 80–90%, with graphite sizes of grade 6 and no coarse graphite or poor spheroidization defects, demonstrating the minimal impact of the low fluoride content in the novel agent on ductile iron castings.
The third application involved pump cover castings of QT600-3 ductile iron, weighing about 1 ton each, with multiple circular open risers of 105 mm diameter. Due to operational constraints, no side insulation was used, and the tops were covered with the novel agent or the international renowned company’s product. Video analysis during the trials indicated that the novel agent ignited about 1 minute earlier, reached peak combustion sooner, and completed combustion earlier, but both had similar total combustion durations of approximately 3.5 minutes. Neither produced刺激性 odors. Post-combustion, the novel agent formed a thicker insulating layer with minimal red glow exposure, whereas the international product had a thinner layer with extensive red areas, indicating poorer insulation. Sectional analysis showed that the novel agent resulted in a flat-bottomed shrinkage with a safety height of 82 mm, compared to an irregular shape and 58 mm safety height for the international product. This further confirmed the higher relative feeding efficiency of the novel covering agent for ductile iron castings.
To understand the thermal behavior differences, we conducted differential scanning calorimetry (DSC) analysis on the novel covering agent and the domestic知名 company’s product. The DSC curves for both agents exhibited three exothermic peaks, but the novel agent’s peaks were more distinct and spread over a wider temperature range, facilitating continuous heat release during different stages of riser solidification. The domestic product’s peaks were less pronounced and concentrated in a narrower range, primarily during the mid to late stages, which is less ideal for early feeding in ductile iron castings. The DSC data for the novel agent are summarized in Table 2.
| Exothermic Stage | Temperature Range (°C) | Max Heat Flow Rate (W/g) | Peak Temperature (°C) | Heat Release (Enthalpy Change) (J/g) |
|---|---|---|---|---|
| Early Stage | 245–385 | 0.14 | 280 | 21.57 |
| Mid Stage | 385–545 | 0.13 | 475 | 25.77 |
| Late Stage | 545–1100 | 0.11 | 690 | 122.72 |
| Total | 245–1100 | – | – | 170.06 |
The total heat release for the novel agent was 170.06 J/g, higher than that of the domestic product, as indicated by the area under the DSC curve. This aligns with the observed better performance in practical trials for ductile iron castings. The novel agent’s formulation, which includes a combination of aluminum-oxygen exothermic reactions and slow-oxidizing carbonaceous materials, ensures rapid ignition and sustained heat release, addressing the quick cooling needs of ductile iron castings. The controlled addition of fluorides and the use of Fe2O3 as an oxidizer and flux aid contribute to reduced environmental impact and minimal interference with graphite morphology in ductile iron castings.
In discussion, the superior performance of the novel covering agent can be attributed to its optimized composition and thermal properties. The relative feeding efficiency metric proves to be a more reliable indicator than traditional efficiency, as it incorporates both volume and shape factors. For ductile iron castings, where riser size is often smaller due to self-feeding from graphite expansion, early and consistent heat supply is vital. The novel agent’s DSC profile, with well-distributed exothermic peaks, supports this requirement, whereas competitors’ products may lag in early-stage heat release. Additionally, the low fluoride and carbon content enhance recyclability and reduce defects, making it suitable for high-quality ductile iron castings in advanced equipment manufacturing.
In conclusion, our novel exothermic insulating covering agent for risers in large and medium-sized ductile iron castings demonstrates excellent feeding efficiency, environmental benefits, and cost-effectiveness. The relative feeding efficiency reaches up to 50%, outperforming products from leading international and domestic companies. The agent’s low fluoride content ensures minimal impact on graphite structure, and its thermal properties provide optimal heat release throughout the solidification process. The introduced relative feeding efficiency metric offers a comprehensive tool for evaluating riser performance in ductile iron castings, advancing the field toward more efficient and sustainable casting practices for ductile iron castings.