The pursuit of high-strength, high-hardness cast iron materials with excellent comprehensive properties is a significant focus in foundry engineering. Among these, nodular cast iron, or ductile iron, stands out due to its favorable combination of strength, ductility, and cost-effectiveness compared to cast steel. For demanding applications requiring superior mechanical properties, such as in heavy-section components for industrial machinery, achieving a consistently high level of performance throughout the entire cross-section is critical. Two primary methodologies are employed to enhance the properties of nodular cast iron: alloying and heat treatment. While alloying generally promotes microstructural uniformity, heat treatment, particularly for thick-walled castings, introduces the challenge of potential gradients in structure and properties due to non-uniform heating and cooling. This study focuses on the latter, investigating the specific effects of a normalizing heat treatment cycle on the microstructure and hardness at varying depths within a heavy-section nodular cast iron casting. The objective is to quantify the homogeneity achieved after heat treatment and understand the underlying microstructural evolution from the surface to the core of a thick wall.
The experimental component used in this investigation was a substantial casting with wall thicknesses ranging from approximately 147 mm to a maximum of 253 mm. The geometry featured both internal and external vertical faces, providing an opportunity to sample locations with differing thermal histories during solidification and subsequent heat treatment. The entire casting was subjected to a full normalizing treatment followed by a tempering process to relieve residual stresses. To establish a baseline, standard attached test blocks (Y-blocks) were cast alongside the main component. Analysis of these blocks confirmed a high-quality nodular cast iron structure prior to heat treatment, with a nodularity exceeding 85% (Type VI graphite) and a predominantly pearlitic matrix. After normalizing, the test blocks exhibited a very fine pearlitic structure and achieved tensile properties meeting a high-grade specification (e.g., akin to QT900), with a hardness of approximately 345 HB.
To evaluate the through-thickness properties of the actual casting, cylindrical samples were extracted via core drilling from two key areas: the inner ring (with a nominal wall thickness of 175 mm) and the outer vertical face (with a nominal wall thickness of 253 mm). Multiple sampling points were selected within each area to account for positional variations. Each cylindrical sample was machined to create a flat surface, upon which Brinell hardness measurements were systematically taken at increasing depths from the cast surface: surface, 10 mm, 30 mm, 70 mm, 100 mm, and at the geometric center (core). The results are consolidated in the table below, which highlights the hardness uniformity both at specific depths across different locations and through the thickness at single locations.
| Sampling Area | Sample ID | Surface | 10 mm Depth | 30 mm Depth | 70 mm Depth | 100 mm Depth | Core (~Mid-Thickness) |
|---|---|---|---|---|---|---|---|
| Inner Ring (~175 mm wall) | #1 (Near Boss) | 329 | 326 | 326 | 325 | 323 | 321 |
| #2 (Middle) | 347 | 343 | 340 | 338 | 336 | 334 | |
| #3 (Middle) | 345 | 341 | 339 | 337 | 335 | 332 | |
| Outer Face (~253 mm wall) | #4 (Behind Boss) | 307 | 300 | 292 | 285 | 278 | 271 |
| #5 (Middle) | 311 | 308 | 305 | 303 | 301 | 299 | |
| #6 (Middle) | 337 | 325 | 315 | 305 | 298 | 290 |
The data reveals several key trends regarding the heat treatment of heavy-section nodular cast iron. For the 175 mm thick inner ring section, the hardness deviation through the entire thickness at any given location is remarkably small, less than 10 HB for samples #2 and #3. This indicates excellent control over the cooling rate during the normalizing process for this wall thickness, leading to highly uniform transformation kinetics. Sample #1, located near a thicker boss section, shows slightly lower overall hardness, attributable to a locally slower cooling rate. Nevertheless, the total hardness range from surface to 100 mm depth is only 21 HB, confirming good through-thickness homogeneity.
The situation for the thicker 253 mm outer section is more complex but still demonstrates good control. Examining hardness at a constant depth across different positions (e.g., 30 mm depth: 292-315 HB, a 23 HB spread) shows acceptable uniformity for a casting of this size. The through-thickness hardness drop is more pronounced, particularly for sample #4, which was taken from behind a massive boss, creating an effective section thickness of over 400 mm. Here, the hardness decreases by 56 HB from surface to core. This can be directly related to the cooling rate gradient. The transformation of austenite to pearlite is a diffusion-controlled process. The cooling rate $V_c$ is inversely proportional to the square of the section thickness $D$ for a given quenching intensity, often approximated by relationships like:
$$
V_c \propto \frac{1}{D^n}
$$
where $n$ is an exponent typically between 1 and 2. A slower cooling rate in the core results in the formation of coarser pearlite, which directly lowers hardness. The relationship between pearlite interlamellar spacing $S$ and hardness $HV$ (or HB) is well-established, often following a Hall-Petch type relationship:
$$
HV = k_1 + \frac{k_2}{\sqrt{S}}
$$
where $k_1$ and $k_2$ are material constants. A larger spacing $S$, resulting from slower cooling, leads to a decrease in hardness $HV$. This fundamental metallurgical principle explains the observed hardness gradients in the thick-section nodular cast iron.
To correlate the mechanical property gradients with microstructural changes, detailed metallographic analysis was performed on sample #4 (from the thickest section) at various depths. The findings are summarized in the following table, tracking the evolution of graphite characteristics and matrix structure.
| Depth from Surface (mm) | Graphite Morphology (VI/V Types) | Nodularity (%) | Nodule Count (nodules/mm²) | Matrix Structure | Approx. Hardness (HB) |
|---|---|---|---|---|---|
| 0 (Surface) | 85% VI + 15% V | 100 | 144 | Fine Pearlite | 307 |
| 30 | 84% VI + 12% V | 96 | 64 | Fine Pearlite | 300 |
| 70 | 60% VI + 34% V | 94 | 31 | Fine Pearlite | 292 |
| 100 | 55% VI + 37% V | 92 | 29 | Coarse Pearlite | 285 |
| 130 (Core) | 50% VI + 36% V | 86 | 16 | Coarse Pearlite | 278 |
The microstructural analysis provides a clear explanation for the hardness trends. First, the graphite structure shows a significant degradation from surface to core. The nodule count drastically decreases from 144 nodules/mm² at the surface to only 16 nodules/mm² at the core. Furthermore, the proportion of perfectly spherical Type VI graphite decreases, with an increase in more irregular (exploded/flake-like) forms. This is a classic characteristic of heavy-section nodular cast iron, where the extended solidification time in the core leads to nodule growth, impingement, and degeneration due to fading of nodularizing elements and longer diffusion times for carbon. The graphite degeneration, while present, did not catastrophically deteriorate; even at 100 mm depth, the morphology was still within an acceptable grade.
Second, and more critically for the hardness in this normalized condition, the matrix structure evolves. While the volume fraction of pearlite remained consistently high (>98%) at all depths, its morphology changed. Near the surface, faster cooling promoted the formation of very fine pearlite with a small interlamellar spacing $S_{surface}$. As depth increased and cooling slowed, the transformation occurred at a higher temperature, allowing for greater carbon diffusion and resulting in coarser pearlite with a larger interlamellar spacing $S_{core}$, where $S_{core} > S_{surface}$. Applying the hardness-spacing relationship $HV \propto 1/\sqrt{S}$, the increase in $S$ directly accounts for the systematic decrease in hardness with depth, as observed. The change from “fine” to “coarse” pearlite is a continuum, and the measured hardness values are a direct reflection of this gradient in micro-scale structure within the seemingly uniform pearlitic matrix of the nodular cast iron.
The successful application of normalizing to heavy-section nodular cast iron hinges on managing the cooling rate. The process aims to austenitize the casting fully and then cool it in still air. For thin sections, this yields a uniform, fine pearlitic structure. For thick sections, the surface cools rapidly in air, while the core cools more slowly, behaving like a much larger thermal mass. The effective cooling rate $V_{eff}$ at a depth $x$ can be modeled considering Fourier’s law of heat conduction:
$$
\frac{\partial T}{\partial t} = \alpha \nabla^2 T
$$
where $\alpha$ is the thermal diffusivity of the nodular cast iron. The solution to this equation for a plate of finite thickness shows that the temperature history $T(x,t)$, and hence the cooling rate, is a strong function of position $x$. The goal in heat treating heavy-section nodular cast iron is to select an austenitizing temperature $T_A$ and a cooling medium (air, forced air, fog) that produces a $T(x,t)$ profile resulting in an acceptable gradient of pearlite fineness and, consequently, hardness. The data shows that for a 250 mm wall, air cooling can maintain a useful hardness level even at the core, though a gradient is inevitable.
In conclusion, this investigation into the normalizing of heavy-section nodular cast iron demonstrates that while inherent gradients exist, they can be managed to produce components with acceptable and predictable properties. For a 250 mm thick wall, the normalizing treatment resulted in relatively small hardness deviations at different depths within the same location, particularly when the cooling was uniform. The hardness at a given depth was consistent across different positions on the casting. The primary source of the through-thickness hardness gradient was identified not as a change in phase fraction, but as a systematic coarsening of the pearlitic matrix due to decreasing cooling rates from surface to core, perfectly described by the inverse relationship between hardness and pearlite interlamellar spacing. Concurrently, graphite nodule count decreased and morphology slightly degenerated in the core, but this had a secondary effect on hardness in this high-pearlite material. These findings underscore the importance of considering thermal mass and cooling dynamics when designing heat treatment processes for heavy-section nodular cast iron components to ensure they meet stringent performance requirements throughout their cross-section.