In modern industrial manufacturing, the properties of ferritic ductile iron castings are fundamentally governed by the successful nodularization of graphite. The choice of nodulizing agent is a critical process parameter, directly impacting the final microstructure and, consequently, the mechanical performance of the cast components. While magnesium-based agents are prevalent, the use of rare earth elements as supplements or primary nodulizers has become increasingly important, particularly for challenging applications. This study investigates the comparative effects of a yttrium-based heavy rare earth nodulizer and a cerium-based medium rare earth nodulizer on the mechanical properties and microstructure of ductile iron castings with varying section thicknesses.
The primary function of any nodulizer is to facilitate the formation of spherical graphite nodules during solidification. Elements like magnesium (Mg), cerium (Ce), lanthanum (La), yttrium (Y), and calcium (Ca) are commonly employed. Their effectiveness stems from a dual action: first, deoxidizing the molten iron, and subsequently, desulfurizing it. This sequence is thermodynamically dictated by the relative free energy of formation of their respective oxides and sulfides. For instance, among rare earths, cerium has a high boiling point (3440 °C) and offers stable chemical reactions with good resistance to interference from anti-nodularizing elements. However, its high cost and tendency to produce less perfectly rounded graphite spheres often limit its application. Yttrium-based heavy rare earth nodulizers are frequently specified for thick-section ductile iron castings due to their renowned resistance to nodularizer fading (recession) during extended holding times. Yttrium exhibits superior desulfurization capability, reducing sulfur content to below 0.008% without the risk of sulfur reversion. Furthermore, it maintains stable nodulizing action at temperatures exceeding 1450 °C, which can be advantageous for achieving higher pouring temperatures. The fading rate of yttrium in molten iron is slower than that of magnesium, primarily due to oxidation losses, thereby offering a longer processing window and greater consistency in heavy castings.
The present work focuses on a practical, foundry-floor investigation. The objective was to evaluate how these two distinct classes of nodulizers influence the final quality of ductile iron castings across different cooling rates, simulated by varying casting wall thickness. This is crucial because the solidification kinetics of ductile iron castings significantly affect graphite nucleation and growth, potentially interacting differently with the residual nodulizing elements.
1. Materials and Experimental Methodology
The study was conducted using step-block castings, designed to provide two distinct section thicknesses within a single casting. The geometry provided sections of 150 mm and 250 mm in thickness. These dimensions were selected to represent medium and heavy-section ductile iron castings, relevant to many industrial components like pump housings, valve bodies, and heavy machinery parts. A 2-ton medium-frequency induction furnace was used for melting, and the molds were prepared using resin-bonded sand. To study the interaction between cooling rate and nodulizer type, samples were extracted from specific locations on each thickness section of the blocks cast with different nodulizers.
The base iron chemistry was carefully controlled. The target was a standard ferritic grade of ductile iron castings. The chemical composition of the base iron and the treated iron for both trials is presented in Table 1. Spectrometry was used for verification.
| Material State | C | Si | CE | Mn | P | S | Mg | Ce | La |
|---|---|---|---|---|---|---|---|---|---|
| Base Iron | 3.46 | 2.59 | – | 0.38 | 0.027 | 0.012 | 0.001 | 0.011 | 0.003 |
| Heavy RE Treated | 3.34 | 3.24 | 4.42 | 0.37 | 0.033 | 0.013 | 0.052 | 0.018 | 0.005 |
| Medium RE Treated | 3.42 | 3.40 | 4.55 | 0.37 | 0.032 | 0.011 | 0.051 | 0.019 | 0.005 |
The key variable was the nodulizer. Two commercially available nodulizers were used:
1. A Yttrium-based Heavy Rare Earth nodulizer.
2. A Cerium-based Medium Rare Earth
nodulizer.
Their nominal compositions are detailed in Table 2.
| Nodulizer Type | Ce | La | Y | Mg | Si | Ca | Al | Ba | Sb | Fe |
|---|---|---|---|---|---|---|---|---|---|---|
| Medium Rare Earth | 0.55-0.75 | 0.25-0.45 | 0.30-0.50 | 6.8 | 46 | 2.2 | 0.8 | 1.0 | – | 41-43 |
| Heavy Rare Earth | 0.30-0.50 | 0.65-0.85 | 0.85-1.05 | 6.8 | 45 | 1.5 | 0.7 | 1.0 | 0.5 | 41.5-43.5 |
The treatment process was standardized for both heats to ensure a valid comparison. The sandwich method (pour-over technique) was employed for nodulizing treatment. The treatment temperature was controlled at 1450 ±10 °C. For each mold pour, approximately 250 kg of iron was treated. The nodulizer addition was fixed at 1.1% of the iron weight, covered with 0.7-0.8% silicon steel chips to moderate the reaction and improve magnesium recovery. Inoculation was performed in two stages: a primary addition of 0.4% during tapping and a late stream inoculation of 0.1% during pouring. The only difference between the two trial castings was the type of nodulizer used in the treatment ladle. The pouring temperature ranged from 1320 to 1340 °C.
After casting and cooling, test specimens were machined from designated locations on both the 150 mm and 250 mm thick sections of each step-block. The sampling ensured representation from areas experiencing different solidification rates within the same thickness. For the purpose of this analysis, samples from the heavy rare earth treated block are labeled ‘A’, and those from the medium rare earth treated block are labeled ‘B’.
The evaluation consisted of two main parts:
1. Mechanical Testing: Tensile test bars (ϕ20 mm) were machined and tested on a universal testing machine according to relevant standards. Hardness measurements (Brinell, 3000 kgf load) were also taken.
2. Metallographic Analysis: Samples were prepared by standard grinding, polishing, and etching (4% nital). Microstructures were examined using optical microscopy. Graphite nodule characteristics such as nodularity, size (according to standard charts), and count were evaluated. The matrix structure was also assessed to confirm a predominantly ferritic matrix, as intended for these ductile iron castings.
2. Experimental Results and Data Analysis
2.1. Mechanical Properties: The Effect of Wall Thickness and Nodulizer
The tensile properties and hardness for the 250 mm thick section are summarized in Table 3. The data reveals distinct trends. For the heavy rare earth (Heavy RE) treated ductile iron castings (Samples A1-1 to A1-9), the tensile strength (Rm) ranged from 479 to 493 MPa, with an average of 485 MPa. The yield strength (Rp0.2) ranged from 369 to 383 MPa, averaging 375 MPa. Elongation (A%) showed a wider spread, from 5.5% to 21.5%, averaging 16.1%. Hardness values were between 163 and 172 HBW.
For the medium rare earth (Medium RE) treated ductile iron castings from the same thick section (Samples B1-1 to B1-9), the average tensile strength was slightly lower at 478 MPa (range: 464-490 MPa). The average yield strength was 368 MPa (range: 354-380 MPa). Interestingly, the average elongation was higher at 19.3% (range: 9.0-26.5%). Hardness was comparable, averaging 167 HBW.
A key observation is the consistency of the strength properties. The range (max-min) for tensile strength in the Heavy RE group was 14 MPa, compared to 26 MPa for the Medium RE group. This suggests that the heavy rare earth nodulizer provided more uniform tensile properties in these thick-section ductile iron castings.
| Nodulizer Type | Sample ID | Tensile Strength, Rm (MPa) | Yield Strength, Rp0.2 (MPa) | Elongation, A (%) | Hardness (HBW) |
|---|---|---|---|---|---|
| Heavy RE | A1-9 | 479 | 369 | 5.5 | 163 |
| A1-8 | 487 | 377 | 10.0 | 164 | |
| A1-7 | 490 | 380 | 17.5 | 171 | |
| A1-6 | 488 | 378 | 17.5 | 168 | |
| A1-5 | 489 | 379 | 17.0 | 172 | |
| A1-4 | 493 | 383 | 19.0 | 170 | |
| A1-3 | 486 | 376 | 16.0 | 170 | |
| A1-2 | 484 | 374 | 20.5 | 168 | |
| A1-1 | 470 | 360 | 21.5 | 167 | |
| Average / Range | – | 485 (14) | 375 (14) | 16.1 (16.0) | 168 (9) |
| Medium RE | B1-9 | 480 | 370 | 10.5 | 169 |
| B1-8 | 468 | 358 | 9.0 | 167 | |
| B1-7 | 488 | 378 | 22.5 | 165 | |
| B1-6 | 490 | 380 | 21.0 | 172 | |
| B1-5 | 486 | 376 | 19.0 | 173 | |
| B1-4 | 485 | 375 | 21.5 | 167 | |
| B1-3 | 475 | 365 | 22.0 | 164 | |
| B1-2 | 466 | 356 | 21.5 | 163 | |
| B1-1 | 464 | 354 | 26.5 | 166 | |
| Average / Range | – | 478 (26) | 368 (26) | 19.3 (17.5) | 167 (10) |
The results for the 150 mm thick section, presented in Table 4, show a different pattern regarding consistency. Here, the Heavy RE treated ductile iron castings exhibited higher average tensile strength (488 MPa vs. 475 MPa) and yield strength (378 MPa vs. 365 MPa) than the Medium RE ones. Elongation was again lower for the Heavy RE group (19.3% vs. 22.6%). However, the range of tensile strength values was larger for the Heavy RE group (36 MPa) than for the Medium RE group (19 MPa). This indicates that for this medium wall thickness, the performance of the heavy rare earth nodulizer, while delivering higher strength on average, showed greater variability from one sample location to another.
| Nodulizer Type | Sample ID | Tensile Strength, Rm (MPa) | Yield Strength, Rp0.2 (MPa) | Elongation, A (%) | Hardness (HBW) |
|---|---|---|---|---|---|
| Heavy RE | A2-5 | 491 | 381 | 15.0 | 169 |
| A2-4 | 496 | 386 | 19.0 | 174 | |
| A2-3 | 499 | 389 | 19.5 | 174 | |
| A2-2 | 493 | 383 | 21.5 | 173 | |
| A2-1 | 463 | 353 | 21.5 | 167 | |
| Average / Range | – | 488 (36) | 378 (36) | 19.3 (6.5) | 171 (7) |
| Medium RE | B2-5 | 481 | 371 | 24.0 | 171 |
| B2-4 | 484 | 374 | 19.5 | 171 | |
| B2-3 | 479 | 369 | 21.0 | 168 | |
| B2-2 | 468 | 358 | 23.5 | 166 | |
| B2-1 | 465 | 355 | 25.0 | 165 | |
| Average / Range | – | 475 (19) | 365 (19) | 22.6 (5.5) | 168 (6) |
The relationship between tensile strength and elongation often follows a trade-off, which can be conceptually modeled. A simplified inverse relationship for these ductile iron castings can be represented as:
$$ R_m + k \cdot A \approx C $$
where $R_m$ is tensile strength, $A$ is elongation, $k$ is a material- and process-dependent constant, and $C$ is another constant. The data shows that the Heavy RE treatment generally shifts the balance towards higher $R_m$ and slightly lower $A$ compared to the Medium RE treatment for the same section size.
2.2. Metallographic Analysis: Graphite Morphology and Matrix
The microstructure holds the key to understanding the mechanical property data. For both section thicknesses, a clear and consistent difference in graphite morphology was observed between the two types of ductile iron castings.
In the 250 mm thick sections: The Heavy RE treated samples (A1 series) predominantly exhibited graphite nodules that were more spherical, smaller in size, and greater in number per unit area compared to the Medium RE treated samples (B1 series). One sample from the Heavy RE group (A1-9) showed larger graphite, which correlated with its lower elongation (5.5%) and slightly lower strength, likely indicating a localized area of slower cooling or minor fading. The Medium RE samples showed acceptable nodularity but with a tendency towards slightly larger and occasionally less perfectly round graphite particles.
In the 150 mm thick sections: The same trend was observed even more distinctly. The Heavy RE treated ductile iron castings (A2 series) consistently displayed superior graphite characteristics: higher nodularity grade, smaller nodule size (finer graphite), and a higher nodule count. The Medium RE samples (B2 series) showed good nodularity but with noticeably larger graphite nodules on average.
The matrix in all analyzed samples was predominantly ferritic, as intended, confirming that the heat treatment or in-mold cooling conditions successfully promoted ferrite formation around the graphite nodules. The differences in mechanical properties are therefore primarily attributable to the differences in graphite morphology rather than the matrix structure.
The beneficial effect of finer, more numerous graphite nodules on strength can be rationalized. The graphite-matrix interface acts as a barrier to dislocation movement. Finer and more numerous nodules create a greater total interfacial area per unit volume, strengthening the material. This relationship can be approximated by considering the inter-nodule spacing, $\lambda$. A smaller $\lambda$, resulting from a higher nodule count $N_v$, increases strength. We can express nodule count density as:
$$ N_v \propto \frac{1}{\lambda^3} $$
And strength often follows a Hall-Petch type relationship:
$$ \sigma_y \approx \sigma_0 + \frac{k_y}{\sqrt{\lambda}} $$
where $\sigma_y$ is the yield strength, $\sigma_0$ is the friction stress, and $k_y$ is a strengthening constant. The Heavy RE treatment, by promoting a higher $N_v$ and smaller $\lambda$, contributes to the observed higher strength in the resulting ductile iron castings.
3. Discussion: Interpreting the Interaction Between Nodulizer and Cooling Rate
The results demonstrate a complex interaction between the nodulizer chemistry and the solidification conditions inherent to producing ductile iron castings of different thicknesses.
For Heavy-Section Castings (250 mm): The superior and more consistent mechanical properties, coupled with better graphite morphology, achieved with the yttrium-based heavy rare earth nodulizer align with its documented advantages. The slow cooling rate in thick sections extends the solidification time, increasing the risk of nodulizer fading (e.g., magnesium loss through oxidation and slag reaction) and graphite degeneration (e.g., vermicular or flake formation). Yttrium’s higher boiling point and slower fading kinetics provide a more stable reservoir of nodulizing elements throughout the extended solidification period. This ensures that the late-solidifying regions, such as the thermal center of the 250 mm section, receive sufficient protection to form spherical graphite. This leads to a more uniform microstructure throughout the casting cross-section, explaining the smaller range (higher consistency) in tensile properties. The slightly lower average elongation in the Heavy RE group, despite better graphite shape, may be influenced by other factors such as subtle differences in inclusion populations or pearlite content, but the primary strength advantage is clear.
For Medium-Section Castings (150 mm): The faster cooling rate changes the dynamics. While the Heavy RE nodulizer still produced a finer and more numerous graphite structure—leading to higher average strength—the property variability was greater than with the Medium RE nodulizer. This suggests that in faster solidifying ductile iron castings, the potent nucleation effect of the heavy rare earths might be more sensitive to local process variations (e.g., slight temperature gradients, inoculation efficiency, or local silicon distribution). The cerium-based medium rare earth nodulizer, while producing slightly larger graphite, may provide a more forgiving and consistent result under these conditions, as evidenced by the tighter property ranges. The faster solidification might also reduce the time available for fading, somewhat negating the primary advantage of yttrium, while its powerful but potentially less uniform nucleation effect becomes the dominant factor.
The graphite size can be related to the number of active nuclei and the growth conditions. The cooling rate, $v_c$, influences the undercooling and thus the nucleation rate, $I$. A simple model links nodule count $N_v$ to these parameters:
$$ N_v \propto \int I(v_c(t)) \, dt $$
The Heavy RE nodulizer appears to increase the integral $I(t)$, generating more nuclei, especially in slower cooling conditions. However, in faster cooling, if nucleation is extremely rapid and heterogeneous, local variations can lead to greater scatter in $N_v$ and, hence, in properties.
4. Conclusions and Practical Implications
This detailed investigation into the performance of two distinct nodulizers for producing ferritic ductile iron castings leads to the following conclusions:
- Graphite Morphology: The yttrium-based heavy rare earth nodulizer consistently promotes a superior graphite morphology in both medium and heavy-section ductile iron castings. It results in higher nodularity, smaller graphite nodule size, and a greater number of nodules per unit area compared to the cerium-based medium rare earth nodulizer under identical melting and treatment conditions.
- Mechanical Properties: The refined microstructure obtained with the heavy rare earth nodulizer translates into higher tensile and yield strength for both section thicknesses investigated. The elongation tends to be slightly lower, reflecting the typical strength-ductility balance.
- Property Consistency (Range): The consistency of mechanical properties, as indicated by the range of measured values, is highly dependent on casting section size.
- For heavy-section (250 mm) ductile iron castings, the heavy rare earth nodulizer provides not only higher strength but also significantly more uniform properties (smaller range). This is critically important for ensuring reliability in large, critical castings.
- For medium-section (150 mm) ductile iron castings, while the heavy rare earth nodulizer delivers higher average strength, it can introduce greater property variability compared to the medium rare earth alternative, which showed more consistent results in this thickness.
Practical Recommendation: The selection of a nodulizer for ductile iron castings should be a strategic decision based on the dominant casting section size and the specific performance priorities.
• For heavy-section and thick-walled ductile iron castings where resistance to fading and maximum property uniformity are paramount, a yttrium-based heavy rare earth nodulizer is strongly recommended.
• For medium-section ductile iron castings where a balance of good properties, high consistency, and potentially lower cost is desired, a cerium-based medium rare earth nodulizer remains a robust and reliable choice.
This study underscores that optimizing the production of high-integrity ductile iron castings requires a holistic approach, carefully matching the nodulizer technology to the specific thermal and geometric challenges of the casting itself.