In recent years, the research focus within our nation’s foundry sector has progressively shifted towards large-section, ultra-large-section, and lightweight nodular cast iron components. As a prominent research topic, lightweight nodular cast iron castings have successfully opened up new application markets. The rationality of casting wall thickness design largely determines the difficulty of the casting process, with performance inhomogeneity being one of its primary defects. Consequently, precise control over the solidification process of castings is paramount, an issue that has garnered widespread attention in academia.
With the rapid advancement of the wind power industry, the structural design of wind turbine components has become increasingly complex. This rise in complexity has led to the widespread occurrence of uneven wall thicknesses, posing significant challenges for both production and application. To address this issue, this investigation focuses on studying the variations in microstructure and mechanical properties of nodular cast iron within stepped test blocks cast under identical process conditions. Through experimentation and analysis, the influence patterns of uneven casting wall thickness on its microstructure and mechanical properties are derived. These findings are intended to contribute to enhancing the performance and quality of wind power castings and to promote the sustainable development of the wind energy sector.
The core objective of this experimental study was to conduct an in-depth analysis of how changes in casting wall thickness affect the microstructure and resultant properties of nodular cast iron. A stepped test block was designed and produced, with its dimensions detailed in the methodology section.
1. Experimental Methodology
1.1 Selection of Chemical Composition
The chemical composition of the base iron was carefully selected to target a predominantly ferritic matrix in the final nodular cast iron, ensuring the study focused on thickness effects rather than composition variations. The target composition is presented in Table 1.
| C | Si | Mn | P | S | Mg | RE | Sb |
|---|---|---|---|---|---|---|---|
| 3.80 | 2.10 | <0.25 | <0.035 | ≤0.020 | 0.035-0.045 | <0.025 | 0.006 |
1.2 Melting and Casting Process
Melting was conducted using a medium-frequency induction furnace. Upon reaching a temperature between 1420°C and 1480°C, the base iron was tapped. Nodularization treatment was performed using the sandwich method. Specifically, 0.9% to 1.05% of a commercial Fe-Si-Mg nodularizing alloy was placed in the bottom of a preheated pouring ladle’s reaction pocket. This was covered with a layer of steel scrap, upon which 0.15% to 0.50% of a primary Fe-Si inoculant was placed. The molten iron was then poured onto this sandwich setup.
Following the nodularization reaction, slag removal was performed. At a temperature ranging from 1350°C to 1370°C, a post-inoculation was carried out by adding 0.20% of a Si-Al-Ba type inoculant during the transfer of the iron to the pouring ladle. The melt was then poured into a resin sand mold designed to produce a stepped test block. The casting system included a sprue, runners, filters, and ingates to ensure smooth and clean filling.
1.3 Specimen Preparation and Sampling Strategy
Prior to sectioning, the entire stepped test block underwent non-destructive testing to ensure internal soundness and validate the sampling plan. Given the block’s size, a comprehensive and representative sampling strategy was essential. The sampling locations were determined based on symmetry and the characteristic solidification modulus of each step, as predicted by numerical simulation software. For each thickness step, specimens were extracted from three distinct layers (designated L1, L2, L3 from the surface to the interior) along the central axis. From each layer, samples for tensile testing, impact testing, metallographic examination, and hardness testing were obtained. The sampling matrix is summarized in Table 2.
| Step Thickness (mm) | Layers Sampled | Samples per Layer |
|---|---|---|
| 100, 135, 170, 205, 240, 275, 310, 345 | 3 (L1: Near-surface, L2: Mid-thickness, L3: Near-center) | 1 Tensile, 2 Impact, 1 Metallographic, 1 Hardness |
2. Results, Analysis, and Discussion
2.1 Solidification Behavior and Macro-Modulus
The solidification sequence and thermal history are critical for understanding the final microstructure of nodular cast iron. The local solidification time, often characterized by the geometric modulus (Volume/Surface Area ratio), directly influences graphite nucleation and growth, as well as the formation of the metallic matrix. For a plate-like geometry, the modulus $M$ can be approximated by:
$$ M \approx \frac{t}{2} $$
where $t$ is the plate thickness, assuming heat extraction occurs predominantly through two large faces. This simplification highlights that thicker sections have a larger modulus and consequently, a longer local solidification time $t_f$. The relationship between secondary dendrite arm spacing $\lambda_2$, a indicator of microstructural scale, and solidification time is often expressed as:
$$ \lambda_2 = k \cdot (t_f)^n $$
where $k$ and $n$ are material constants. A longer $t_f$ promotes coarser microstructure. In our stepped block, the modulus increases with step thickness, leading to progressively slower cooling. This fundamental principle underpins the observed variations in the microstructure and properties of the nodular cast iron across different thicknesses.
2.2 Analysis of Graphite Morphology
Metallographic samples from all steps and layers were prepared, etched with 4% nital, and examined under an optical microscope at 100x magnification. The graphite morphology was assessed according to relevant standards. Representative micrographs confirmed that the graphite was predominantly spheroidal.
Graphite Size and Distribution: A clear trend was observed: as the step thickness increased, the average graphite nodule diameter in the upper layers of the block tended to increase. This is attributed to the decreased cooling rate in thicker sections. The prolonged liquid state allows for both Ostwald ripening (growth of larger nodules at the expense of smaller ones) and potential flotation of graphite nodules. Notably, in regions corresponding to thermal junctions between steps (simulated hotspots), graphite nodules were fewer in number and larger in size compared to areas further from these junctions. Areas distant from hotspots cool more rapidly, promoting a higher nucleation rate for graphite, resulting in a larger population of finer nodules. The relationship between undercooling $\Delta T$ and nodule count $N_v$ (number per unit volume) can be conceptually described as:
$$ N_v \propto \exp\left(-\frac{\Delta G^*}{k_B \cdot \Delta T}\right) $$
where $\Delta G^*$ is the critical nucleation energy barrier and $k_B$ is the Boltzmann constant. Faster cooling (thinner sections) generates higher undercooling, favoring a higher $N_v$.
Nodule Count and Spheroidization: Quantitative image analysis was performed to determine the nodule count per unit area and the nodularity (percentage of graphite in nodular form). The results are summarized in Table 3.
| Avg. Thickness (mm) | Layer | Avg. Nodule Count (per mm²) | Avg. Nodularity (%) | Graphite Size (Scale) |
|---|---|---|---|---|
| ~138 | L1 | 122 | 92 | 6 |
| L2 | 118 | 90 | 6 | |
| L3 | 105 | 88 | 5-6 | |
| ~238 | L1 | 95 | 91 | 5-6 |
| L2 | 87 | 85 | 5 | |
| L3 | 82 | 83 | 5 | |
| ~328 | L1 | 78 | 93 | 5 |
| L2 | 70 | 84 | 5 | |
| L3 | 65 | 80 | 5 |
The data indicates that both nodule count and nodularity generally decrease with increasing section thickness and from the surface (L1) to the center (L3) of a given step. The decrease in nodule count with thickness is directly linked to the reduced cooling rate and undercooling. The lower nodularity in the center of thicker sections can be attributed to longer exposure to surface-active elements fading and a greater propensity for graphite degeneration in the slowly cooling, solute-rich last-to-freeze liquid.
2.3 Matrix Microstructure Characteristics
The matrix of all examined samples consisted of a mixture of ferrite (white etching constituent) and pearlite (dark etching lamellar structure). The relative amounts of ferrite and pearlite determine key mechanical properties like tensile strength and elongation. The pearlite content did not exhibit a strong, systematic variation with increasing step thickness alone. This is likely because the final matrix structure in ferritic-pearlitic nodular cast iron is strongly influenced by both the cooling rate through the eutectoid transformation range and the composition (particularly Si and Cu/Mn/Sn content). While thicker sections cool slower through the eutectoid range, favoring pearlite formation, the high silicon content (2.1%) and specific inoculation practice in this study promoted ferrite formation, leading to a relatively stable, high-ferrite matrix across the thickness range. The fraction of pearlite $f_P$ can be influenced by the transformation kinetics described by time-temperature-transformation diagrams. The tendency can be summarized as a competition between factors promoting ferrite (high Si, slow cooling from eutectic to eutectoid) and those promoting pearlite (slow cooling through eutectoid, presence of pearlite-stabilizers).
2.4 Comprehensive Evaluation of Mechanical Properties
Standard tensile tests, Charpy V-notch impact tests at -20°C, and Brinell hardness tests were conducted on specimens from each layer and step. The results are compiled in Table 4.
| Avg. Thickness (mm) | Layer | Tensile Strength (MPa) | Elongation (%) | Impact Energy (-20°C, J) | Hardness (HBW) |
|---|---|---|---|---|---|
| ~138 | L1 | 405 | 24 | 15.5 | 152 |
| L2 | 398 | 22 | 14.0 | 149 | |
| L3 | 392 | 19 | 12.5 | 147 | |
| ~238 | L1 | 390 | 21 | 13.0 | 145 |
| L2 | 385 | 18 | 11.0 | 142 | |
| L3 | 378 | 16 | 9.5 | 140 | |
| ~328 | L1 | 395 | 23 | 14.5 | 148 |
| L2 | 382 | 17 | 10.5 | 141 | |
| L3 | 375 | 15 | 9.0 | 138 |
Tensile Strength and Hardness: The tensile strength and hardness showed a mild but consistent decreasing trend with increasing section thickness and from surface to center. The tensile strength ranged from approximately 415 MPa in thinner sections to about 365 MPa in the center of the thickest section. Hardness followed a similar pattern, from around 155 HBW down to 135 HBW. This decrease is primarily linked to the coarsening of the microstructure (both graphite and matrix). The yield strength of ferritic-pearlitic nodular cast iron can be related to microstructural parameters via a Hall-Petch type relationship and the rule of mixtures for the two-phase matrix:
$$ \sigma_y = f_F \cdot \sigma_{y_F} + f_P \cdot \sigma_{y_P} + k \cdot (d)^{-1/2} + \Delta\sigma_{graphite} $$
where $f_F$ and $f_P$ are ferrite and pearlite fractions, $\sigma_{y_F}$ and $\sigma_{y_P}$ their respective yield strengths, $d$ is the ferrite grain size, $k$ is a constant, and $\Delta\sigma_{graphite}$ accounts for strengthening from the graphite nodules. Coarser microstructure (larger $d$, larger graphite nodules) generally leads to lower strength and hardness.
Elongation and Impact Toughness: The ductility and toughness parameters, elongation and impact energy, exhibited more pronounced sensitivity to section thickness and location. Elongation varied significantly, from highs of 26% in thin sections to lows near 15% in the center of thick sections. Impact energy showed a similar declining trend with thickness, with values dropping from around 16.5 J to 9 J. This marked deterioration is critically linked to the reduction in graphite nodule count and potential degradation of nodularity in slowly cooled regions. Fewer, larger nodules and the presence of degenerate graphite forms act as more potent stress concentrators and crack initiation sites, drastically reducing ductility and toughness. The relationship between fracture toughness $K_{IC}$ and microstructural features in nodular cast iron is complex, but it is well-established that a high count of small, well-spheroidized graphite nodules is essential for maximizing toughness. The energy absorbed during fracture $U$ can be considered to have components related to plastic deformation around nodules and matrix cracking:
$$ U \propto N_v \cdot \bar{d}^2 \cdot \epsilon_f $$
where $N_v$ is the nodule count per volume, $\bar{d}$ is the mean nodule diameter, and $\epsilon_f$ is the fracture strain of the inter-nodular matrix. A higher $N_v$ and smaller $\bar{d}$ typically lead to greater $U$.
3. Summary and Conclusions
This systematic investigation into the effects of wall thickness on the microstructure and mechanical properties of nodular cast iron, using a stepped test block cast under a single set of process parameters, yields the following key conclusions:
- The matrix microstructure across all section thicknesses consisted of ferrite and pearlite (F+P). The graphite was primarily of types VI and V, with a spheroidization grade between 4 and 5, and a size rating between 5 and 6, according to standard scales.
- Graphite morphology is strongly influenced by cooling rate. Increasing section thickness leads to a decrease in graphite nodule count, an increase in nodule size, and a tendency for reduced nodularity, especially in the thermal centers of thick sections. The design of wall thickness transitions (hotspots) significantly affects the local distribution of graphite morphology.
- Tensile strength and hardness exhibit a relatively modest decrease with increasing section thickness. This is attributed to the general coarsening of the microstructure. The relationship can be partially described by strengthening models incorporating phase fractions and microstructural scale.
- Ductility (elongation) and toughness (impact energy) are far more sensitive to section thickness, showing significant degradation in thicker sections. This is directly correlated with the unfavorable changes in graphite morphology—specifically, lower nodule count and potential nodularity degradation—which promote early crack initiation and propagation.
- A clear gradient in properties exists from the surface to the center within a given thick section, with the center exhibiting the lowest nodule count, nodularity, and mechanical performance.
The findings underscore that achieving consistent and superior properties in heavy-section or varying-thickness nodular cast iron components cannot rely on melting and treatment practices alone. The results emphatically demonstrate that casting geometry and the resultant solidification kinetics are dominant factors controlling the final graphite structure, which in turn dictates mechanical performance. Therefore, optimal casting quality requires an integrated approach where metallurgical control (composition, inoculation, nodularization) is synergistically combined with optimized casting design and molding techniques aimed at controlling and, where possible, increasing the solidification rate to promote a fine, uniform, and well-nodularized graphite structure throughout the casting. This holistic approach is essential for advancing the application of nodular cast iron in demanding, lightweight, and structurally complex components such as those used in the wind power industry.