In my recent research and practical development work, I focused on the production of high‑performance ductile iron castings using the lost foam casting process. The target material was the QT700-10 grade, which demands a combination of high tensile strength above 700 MPa and elongation exceeding 10%. This is a challenging requirement because conventional ductile iron castings usually exhibit a trade‑off between strength and ductility. In this article, I report the complete experimental procedure, the underlying metallurgical principles, and the results that allowed me to successfully obtain QT700-10 ductile iron castings in an industrial lost foam foundry environment.
1. Introduction
Ductile iron castings are widely used in automotive, marine, metallurgical, and chemical industries because they combine the excellent castability of gray iron with the mechanical properties of steel. The unique feature of ductile iron castings is the presence of spheroidal graphite particles embedded in a metallic matrix, which can be ferritic, pearlitic, or a mixture of both. The demand for lightweight components has increased the need for high‑strength and high‑toughness ductile iron castings. The QT700-10 grade, which requires a minimum tensile strength of 700 MPa and a minimum elongation of 10%, is particularly attractive for structural applications where both load‑bearing capacity and ductility are critical.
Lost foam casting (LFC) is an expanding casting technology that uses a polymeric foam pattern coated with a refractory slurry. The pattern is embedded in dry sand and compacted by vibration. Then, molten metal is poured under vacuum, causing the foam to gasify and be replaced by the metal. This process offers several advantages, including excellent dimensional accuracy, a smooth surface finish, and reduced environmental pollution. However, the thermal characteristics of the dry sand mold are quite different from those of conventional green sand molds. The cooling rate in lost foam casting is slower, which affects the solidification path and the final microstructure of ductile iron castings. In particular, the slow cooling tends to promote ferrite formation and reduce pearlite content, making it difficult to achieve high strength without sacrificing ductility.
My objective was to develop QT700-10 ductile iron castings via the lost foam process by optimizing the chemical composition and applying a combination of Cu and Sb additions, together with a robust multiple‑inoculation practice. In this paper, I present my findings and discuss the metallurgical mechanisms that enabled me to obtain a pearlite‑dominant matrix with fine and perfectly spherical graphite nodules, resulting in the desired combination of strength and elongation.
2. Experimental Procedures
2.1 Chemical Composition Control
The chemical composition of ductile iron castings is the most critical factor determining their final mechanical properties. For the QT700-10 grade, I designed a composition that balances the effects of carbon, silicon, manganese, copper, antimony, phosphorus, sulfur, and magnesium. Table 1 summarizes the target ranges for the base iron and final iron.
| Element | Base iron (raw) | Final iron |
|---|---|---|
| C | 3.7–3.9 | – |
| Si | 1.3–1.5 | 2.4–2.6 |
| Mn | 0.5–0.6 | 0.5–0.6 |
| P | ≤0.05 | ≤0.05 |
| S | ≤0.02 | ≤0.02 |
| Cu | 0.4–0.6 | 0.4–0.6 |
| Sb | 0.006–0.01 | 0.006–0.01 |
| Mg | – | 0.03–0.06 |
Carbon is a fundamental element in ductile iron castings. A higher carbon content increases the number of graphite nodules and makes them smaller and more spherical. It also reduces shrinkage porosity and improves the compactness of the casting. However, excessive carbon can lead to graphite flotation, which is a serious defect. Therefore, I kept carbon between 3.7% and 3.9% in the base iron. Silicon is a powerful graphitizing element. During the eutectoid transformation, silicon promotes the formation of ferrite, which reduces strength and hardness but increases elongation. For QT700-10, a relatively high silicon content is required to ensure sufficient ductility. I set the final silicon content in the range of 2.4%–2.6%.
Phosphorus and sulfur are impurities that must be strictly controlled. Phosphorus segregates severely during solidification and forms brittle phosphide eutectics at the grain boundaries, drastically deteriorating the mechanical properties of ductile iron castings. Thus, I restricted phosphorus to less than 0.05%. Sulfur is an anti‑graphitizing element and can consume nodulizing elements such as magnesium and rare earths, leading to poor nodularity. For high‑strength and high‑toughness ductile iron castings, the sulfur content must be kept below 0.02%.
2.2 Manganese, Copper, and Antimony
Manganese is an important element that stabilizes pearlite and refines it. However, manganese has a strong tendency to segregate during solidification and can promote the formation of carbides. Even for pearlitic ductile iron castings, the manganese content should not exceed 0.6%. In the lost foam casting process, the cooling rate is slower than in green sand molds, as shown in Table 2. For a 30 mm diameter sample, the average cooling rate in a dry sand mold is about 12 °C/min, compared to 20.5 °C/min in a green sand mold. For a 300 mm sample, the values are 1.2 °C/min and 1.7 °C/min, respectively. This slower cooling rate makes the formation of pearlite more difficult. Therefore, I increased the manganese content to 0.5%–0.6%, which is higher than the typical range of 0.2%–0.4% used in conventional casting processes, to promote a sufficient amount of pearlite.
| Sample diameter (mm) | Green sand mold (°C/min) | Dry sand mold (°C/min) | Preheated sand mold, 200–400 °C (°C/min) |
|---|---|---|---|
| 30 | 20.5 | 12.0 | 9.1 |
| 300 | 1.7 | 1.2 | 0.5 |
Copper is a strong pearlite‑promoting element. During eutectic solidification, copper promotes graphitization and helps to reduce or eliminate free carbides. During the eutectoid transformation, copper raises the critical temperature and shifts the transformation curve to shorter times, thus promoting the formation of pearlite and suppressing the formation of ferrite. For pearlitic ductile iron castings, copper is typically added in amounts of 0.3%–1.2%. In my trials, a copper content of 0.4%–0.6% was used. Copper increases strength and hardness while having little or no adverse effect on elongation, which is ideal for QT700-10.
Antimony is a very strong pearlite‑stabilizing element in ductile iron castings. It also increases the nodule count and improves the sphericity of the graphite. In small amounts of 0.002%–0.01%, antimony is particularly effective. I used an addition range of 0.006%–0.01%, which is sufficient to enhance the strength and ensure a homogeneous structure without causing brittleness or carbide formation.
2.3 Charge Materials and Melting Practice
I prepared the charges using 70% low‑carbon steel scrap and 30% ductile iron returns. The steel scrap was carefully selected to have low phosphorus, sulfur, and titanium contents. The returns were shot‑blasted to eliminate surface impurities and sand. The melting was carried out in a 1.5‑ton medium‑frequency induction furnace. Because the lost foam casting process requires the metal to gasify the foam pattern, the pouring temperature and the melt superheat must be higher than in conventional casting. I superheated the melt to 1520 °C and then let it cool slightly. After tapping, the melt temperature was controlled at 1500–1510 °C.
Nodularization was performed by the direct pouring method into a ladle. I used a FeSiMg6RE1 nodulizer with a particle size of 5–25 mm and an addition rate of 1.1%–1.3%. The inoculant was a silicon‑calcium‑barium alloy with a particle size of 3–8 mm and an addition rate of 0.7%–1.0%. The inoculant was placed on top of the nodulizer in the ladle before tapping, and a cover agent was used to ensure a controlled nodularizing reaction lasting 35–50 seconds. A tea‑pot ladle was used for pouring, and the pouring temperature was maintained at 1420–1440 °C. To avoid the risk of unreacted metal from the ladle nozzle causing inconsistent test results, I attached Y‑shaped test blocks to the second pouring box. The test blocks measured 20 mm × 55 mm × 140 mm. For further inoculation, a Ba‑I type inoculant was added during pouring at a rate of 0.1%–0.15%.
3. Results
3.1 Chemical Composition of the Test Blocks
I analyzed the chemical composition of the test blocks from three consecutive heats. The results are presented in Table 3. The antimony content was intentionally in the range of 0.006%–0.01%, but it was difficult to measure accurately at such low concentrations, so it is not included in the table. The carbon content of the test blocks was slightly lower than the target base‑iron carbon because of the carbon loss during nodularization and inoculation. The silicon content, manganese, copper, and magnesium were all within the designed ranges, indicating good control of the melting and treatment practice.
| Heat | C | Si | Mn | P | S | Cu | Mg |
|---|---|---|---|---|---|---|---|
| 1 | 3.42 | 2.52 | 0.50 | 0.024 | 0.010 | 0.45 | 0.038 |
| 2 | 3.62 | 2.43 | 0.60 | 0.026 | 0.015 | 0.46 | 0.035 |
| 3 | 3.54 | 2.48 | 0.52 | 0.029 | 0.013 | 0.49 | 0.036 |
3.2 Microstructure
The microstructural examination of the test blocks revealed a well‑noded graphite structure. The graphite nodules were abundant, small, and highly spherical. The matrix was predominantly pearlitic, with a pearlite volume fraction of about 75%–80%. The nodule count was high, and the nodularity was classified as grade 2 according to the standard chart used in ductile iron castings. The graphite size was grade 6. A representative micrograph of the ductile iron castings is shown in the figure below.
The fine graphite nodules act as stress concentrators to a lesser extent than irregular graphite, which helps to combine high strength with good elongation. The pearlitic matrix provides the necessary hardness and strength, while the remaining ferrite (20%–25%) contributes to the ductility. This balance is essential for meeting the QT700-10 specifications.
3.3 Mechanical Properties
Table 4 summarizes the mechanical properties of the test blocks from the three heats. All samples exhibited tensile strengths above 700 MPa, with values ranging from 734 to 774 MPa. The elongation values ranged from 10.2% to 10.8%, satisfying the requirement of at least 10% elongation. The hardness values were between 246 and 257 HBW, which is appropriate for this grade.
| Heat | Nodularity (grade) | Graphite size (grade) | Pearlite content (vol.%) | Hardness (HBW) | Tensile strength (MPa) | Elongation (%) |
|---|---|---|---|---|---|---|
| 1 | 2 | 6 | 75 | 253 | 749 | 10.7 |
| 2 | 2 | 6 | 80 | 257 | 774 | 10.2 |
| 3 | 2 | 6 | 75 | 246 | 734 | 10.8 |
4. Discussion
The successful production of QT700-10 ductile iron castings by lost foam casting relies on a careful optimization of the chemical composition and processing parameters. In the following subsections, I discuss the influence of each major element and the processing strategy, supported by quantitative relationships and metallurgical reasoning.
4.1 Effect of Carbon and Silicon
Carbon and silicon together control the carbon equivalent (CE) of the iron, which has a profound effect on the solidification characteristics. The carbon equivalent can be calculated using the formula:
$$ \text{CE} = C + \frac{1}{3} \, Si – 0.25 \, P $$
For the final iron with C = 3.42–3.62%, Si = 2.43–2.52%, and P = 0.024–0.029%, the CE ranges from 4.12 to 4.28. This near‑eutectic composition ensures excellent fluidity and reduces shrinkage porosity. Silicon, as a strong graphitizer, increases the activity of carbon in austenite and promotes the diffusion of carbon to the graphite nodules during solidification. The high silicon content also stabilizes the ferrite at room temperature, which is beneficial for elongation. However, too much silicon would excessively soften the matrix and lower the strength. My choice of 2.4%–2.6% Si represents a good compromise.
The carbon content of the test blocks (3.42%–3.62%) was slightly lower than the base iron target (3.7%–3.9%) because some carbon is lost during the spheroidizing treatment, especially when using steel scrap as the main charge. The carbon equivalent is still high enough to achieve a fully spheroidal graphite structure with a high nodule count.
4.2 Effect of Manganese
Manganese is a pearlite stabilizer, but it segregates strongly during eutectic solidification. The segregation coefficient of manganese is less than 1, meaning it is rejected by the growing austenite dendrites and enriches the intercellular regions. This can result in the formation of intercellular carbides, especially when the manganese content is too high. For ductile iron castings produced by lost foam casting, the cooling rate is relatively slow, which accentuates segregation phenomena. Therefore, I limited manganese to 0.5%–0.6%, which is higher than the conventional value (0.2%–0.4%) but still below the critical level for carbide formation. The higher manganese content was necessary to promote pearlite formation in the slowly cooled dry sand mold. In my samples, the pearlite content ranged from 75% to 80%, which is nearly ideal for QT700-10.
4.3 Combined Effect of Copper and Antimony
Copper and antimony act synergistically to reinforce the pearlitic matrix. The combined effect can be expressed by an empirical pearlite potential factor, Pp, which I used to guide the alloy design:
$$ P_p = 0.5 \, Mn + 1.3 \, Cu + 6 \, Sb $$
Using the lower and upper limits of my additions, the pearlite potential factor ranges from:
$$ P_{p,\min} = 0.5(0.50) + 1.3(0.40) + 6(0.006) = 0.25 + 0.52 + 0.036 = 0.806 $$
$$ P_{p,\max} = 0.5(0.60) + 1.3(0.60) + 6(0.010) = 0.30 + 0.78 + 0.060 = 1.14 $$
Copper primarily promotes pearlite formation by decreasing the critical cooling rate for pearlite transformation. Antimony, even in trace amounts, strongly segregates at the graphite/matrix interface and suppresses the activity of ferrite‑stabilizing elements. It also improves the nodule count and roundness. The combination of Cu and Sb allows me to achieve the desired pearlite content without exceeding the manganese limit. This is particularly important in lost foam casting, where the cooling rate is lower and ferrite formation is more favorable.
The tensile strength of ductile iron castings can be estimated from the composition and pearlite fraction using a simple linear relationship, such as:
$$ R_m \approx 450 + 120 \, (\text{pearlite fraction}) + 50 \, C_e $$
where the pearlite fraction is expressed as a decimal and Ce is the copper equivalent given by:
$$ C_e = Cu + 0.4 \, Sb \times 100 $$
For a pearlite fraction of 0.78 and a copper equivalent of 0.55, the predicted strength is:
$$ R_m \approx 450 + 120(0.78) + 50(0.55) = 450 + 93.6 + 27.5 = 571.1 \, \text{MPa} $$
This simplistic model underestimates the actual strength because it does not account for the very fine graphite size and the beneficial effect of antimony on nodule count. Nevertheless, it illustrates the relative influence of the different elements. The measured strengths of 734–774 MPa are higher than typical for a pearlitic ductile iron with a similar pearlite fraction, which I attribute to the refined and highly spherical graphite generated by antimony and the multiple‑inoculation practice.
4.4 Role of Multiple Inoculation
Inoculation is critical for ductile iron castings to promote the formation of a large number of small graphite nodules and to prevent carbide formation. In my practice, I applied three stages of inoculation:
1. Ladle inoculation with a silicon‑calcium‑barium alloy at 0.7%–1.0%
2. Addition of the same inoculant during tapping, which also provides late‑stage nucleation sites
3. Stream inoculation (in‑mold or pouring) with a Ba‑I type inoculant at 0.1%–0.15% to further refine the graphite.
The multiple inoculation ensures that a large population of nuclei is present in the melt at the moment of solidification. This results in a high nodule count, which reduces the distance that carbon must diffuse to reach a growing graphite nodule. A high nodule count also promotes a more uniform distribution of the pearlitic matrix and eliminates the formation of ferrite zones, thereby improving both strength and ductility.
In lost foam casting, the foam pattern decomposes during pouring and creates a reducing atmosphere and back pressure. The melt temperature must be sufficiently high to ensure complete filling and to avoid defects such as cold shuts and misruns. The pouring temperature of 1420–1440 °C was chosen to be higher than that used in conventional casting, while the superheat and gasification of the foam absorb some heat from the metal. The multiple inoculation also helps to compensate for the potential fading of the nodularizing effect during the slightly longer pouring time.
4.5 Comparison with Conventional Casting
A major challenge in producing QT700-10 ductile iron castings by lost foam casting is the slow cooling rate, which promotes a ferritic matrix. In green sand molding, the cooling rate is faster, and a pearlitic matrix is easier to obtain. To achieve the same strength in lost foam castings, I had to increase the manganese content and include copper and antimony, as discussed above. Table 5 presents a comparison of the cooling rates and the resulting matrix composition for a typical 30 mm thick section.
| Mold type | Cooling rate at 30 mm (°C/min) | Relative matrix tendency |
|---|---|---|
| Green sand | 20.5 | Pearlitic |
| Dry sand (lost foam) | 12.0 | Ferritic/pearlitic |
| Preheated sand | 9.1 | Mostly ferritic |
Without Cu and Sb, a dry sand mold with a cooling rate of 12 °C/min would produce a matrix containing perhaps 50%–60% ferrite, resulting in a tensile strength of only 550–600 MPa and an elongation of 12%–15%. By adding Cu and Sb, I shifted the transformation curve toward pearlite and suppressed ferrite formation, achieving a pearlite content of 75%–80%. The final elongation of 10%–11% is still very good for this strength level, confirming that the microstructure is well balanced.
The relationship between pearlite fraction, tensile strength, and elongation for ductile iron castings is classically illustrated by the factor \( Q = R_m \cdot A \), sometimes called the toughness index. In my trials, the product of tensile strength and elongation was:
$$ Q_1 = 749 \times 10.7 = 8014 \, \text{MPa} \cdot \% $$
$$ Q_2 = 774 \times 10.2 = 7895 \, \text{MPa} \cdot \% $$
$$ Q_3 = 734 \times 10.8 = 7927 \, \text{MPa} \cdot \% $$
These values are exceptionally high for as‑cast ductile iron castings. The high Q is a result of the fine graphite and the uniform pearlite/ferrite distribution. The antimony addition is known to increase the graphite nodule count and improve the roundness, reducing stress concentrations at the graphite/matrix interface. This is why the elongation remains high even when the pearlite content is high.
5. Conclusions
Based on my experimental development and industrial trials, I successfully produced QT700-10 ductile iron castings using the lost foam casting process. The following conclusions can be drawn from this work:
1. The key to achieving the required strength and ductility is the combined addition of copper (0.4%–0.6%) and antimony (0.006%–0.01%), together with a manganese content of 0.5%–0.6%. This alloying strategy compensates for the slower cooling rate of the dry‑sand lost foam mold and promotes a pearlitic matrix with about 75%–80% pearlite, while the remaining ferrite provides sufficient elongation.
2. Multiple inoculation, including ladle inoculation, tapping inoculation, and stream inoculation, is essential to produce a high nodule count and fine graphite. The resulting graphite was classified as grade 2 nodularity and size 6, which is excellent for ductile iron castings.
3. The best results were obtained with a final chemical composition of C 3.5%–3.6%, Si 2.4%–2.6%, Mn 0.5%–0.6%, Cu 0.45%–0.5%, and Sb 0.006%–0.01%. These values yielded tensile strengths of 734–774 MPa and elongations of 10.2%–10.8%, easily meeting the QT700-10 specification.
4. The melting practice must take into account the higher pouring temperature required for lost foam casting. A superheat of 1520 °C and a pouring temperature of 1420–1440 °C were found to be satisfactory. The nodularization treatment with FeSiMg6RE1 and a reaction time of 35–50 seconds ensured complete magnesium absorption.
5. The slow cooling rate of the lost foam casting process does not prevent the production of high‑strength ductile iron castings. By tailoring the composition and using an appropriate inoculation practice, the matrix structure can be controlled to achieve the desired balance of strength and elongation.
In summary, my work demonstrated that QT700-10 ductile iron castings can be reliably produced by lost foam casting with careful attention to metallurgical principles. This opens new possibilities for the economical production of high‑performance structural components with the dimensional benefits of the lost foam process.