In the field of engineering materials, grey iron castings are widely used due to their excellent castability, machinability, and damping capacity. However, their mechanical properties, particularly tensile strength and ductility, are often limited by the presence of flake graphite, which acts as stress concentrators. To address this, surface modification techniques have been explored to tailor the microstructure and performance of grey iron castings, enabling distinct properties between the surface and core regions. This study focuses on achieving surface layer graphite nodularization in grey iron castings through the application of rare earth (RE)-based coatings on mold surfaces. By pouring molten grey iron into coated molds, a composite structure is formed: a surface layer with spheroidal graphite, a transition layer with vermicular graphite, and a core with flake graphite. This approach enhances surface properties like wear resistance and crack resistance while maintaining the desirable thermal conductivity of the core, making it suitable for applications such as steel ingot molds and crankshafts.
The motivation for this research stems from observations in industrial practices. For instance, in steel ingot molds made of compacted graphite iron, a high nodularity surface layer was found to significantly improve service life by enhancing mechanical properties and reducing thermal stress. Conversely, in ductile iron crankshafts, undesired surface flake graphite layers can lead to premature failure. Therefore, developing a controlled surface nodularization process for grey iron castings is crucial. Previous studies have employed coatings containing rare earth, rare earth-magnesium, or magnesium salts to induce surface graphite modification. This work builds on that by systematically investigating the effects of RE-based coatings, specifically using 1# ferro-silico-rare earth and 75 ferrosilicon, on the graphite morphology and microstructure of grey iron castings. The goal is to optimize the coating composition and process parameters to achieve a consistent and thick nodular graphite layer, exceeding 2.5 mm, which can benefit various engineering components where surface and core requirements differ.

To understand the underlying mechanisms, it is essential to review the fundamentals of graphite formation in iron castings. In grey iron castings, graphite typically precipitates as flakes during eutectic solidification, influenced by cooling rate and composition. The nodularization of graphite is achieved by adding elements like magnesium or rare earths, which alter the surface energy of graphite crystals, promoting spheroidal growth. The effectiveness of these nodularizing agents depends on factors such as carbon equivalent (CE), sulfur content, and cooling conditions. For RE-based treatments, the reaction can be described by the following equilibrium, where rare earth elements (e.g., Ce) react with sulfur to form stable sulfides, reducing the sulfur content in the melt and facilitating graphite spheroidization:
$$ \text{RE} + \text{S} \rightarrow \text{RE}_x\text{S}_y $$
This reaction highlights the importance of low sulfur levels for successful nodularization. Additionally, the carbon equivalent, defined as $$ \text{CE} = \%\text{C} + \frac{1}{3}\%\text{Si} $$, plays a critical role; hypereutectic compositions (CE > 4.3%) tend to favor graphite nodularization with sufficient RE addition, while hypoeutectic ones may only form compacted graphite. In this study, we explore these relationships through experimental designs, aiming to provide a comprehensive analysis for optimizing surface modification in grey iron castings.
The experimental methodology involved the preparation of RE-based coatings and their application to mold surfaces. The coatings were formulated with 1# ferro-silico-rare earth alloy (containing rare earth elements like cerium) and 75 ferrosilicon as inoculant, along with fluxes such as borax and sodium fluoride, and binders like sodium silicate and polyvinyl butyral (PVB). The alloy particle size was controlled between 0.2 and 0.6 mm to ensure effective interaction with the molten iron. The coatings were applied to a thickness of approximately 5 mm on the mold wall opposite the gating system, as illustrated in the浇注工艺简图. The molds were made of clay sand or cold-set resin sand, and after coating, they were dried naturally for 24 hours before use.
Molten iron was prepared in medium-frequency induction furnaces with varying compositions to assess the influence of carbon equivalent and sulfur content. The base iron for grey iron castings had typical compositions, adjusted with ferrosulfide to control sulfur levels. The pouring temperature was maintained around 1550°C to ensure fluidity and proper reaction with the coating. Specimens with variable cross-sections, as shown in the试样简图, were cast to study the effect of wall thickness on graphite morphology. The thickest section (50 mm) was primarily analyzed for graphite nodularization depth and quality.
An exploratory orthogonal experiment (L9 array) was conducted to identify key factors affecting surface nodularization. The factors and levels are summarized in Table 1, focusing on the content of 1# ferro-silico-rare earth, borax, sodium silicate, and sodium fluoride. Graphite powder was used as a filler in the coatings. The results, evaluated based on graphite nodularization grade and depth, indicated that the amount of 1# ferro-silico-rare earth had the most significant impact. The optimal combination from this phase was used as a basis for subsequent comparative tests.
| Level | 1# Ferro-silico-RE (%) | Borax (%) | Sodium Silicate (%) | Sodium Fluoride (%) |
|---|---|---|---|---|
| I | 40 | 1 | 4 | 2 |
| II | 60 | 0 | 10 | 1 |
| III | 80 | 2 | 7 | 0 |
In the comparative experiments, the binder and flux compositions were stabilized, while the contents of 1# ferro-silico-rare earth and 75 ferrosilicon were varied, along with the carbon equivalent and sulfur content of the base iron. The factors and levels are detailed in Table 2. Each coating included fixed proportions of borax and sodium fluoride (2.5% relative to 1# ferro-silico-rare earth weight), sodium silicate (10% of powder weight), and PVB (0.4% of powder weight). The iron melt compositions were carefully controlled, and the results were analyzed for graphite morphology transitions and nodular layer depth.
| Level | Carbon Equivalent (%) | Sulfur Content (%) | 1# Ferro-silico-RE (parts) | 75 Ferrosilicon (parts) |
|---|---|---|---|---|
| I | 4.1 | 0.04 | 100 | 10 |
| II | 4.26 | 0.02 | 120 | 8 |
| III | 4.5 | 0.06 | 80 | 4 |
The outcomes from the comparative experiments are presented in Table 3, showing the graphite morphology and nodular layer depth for each trial. The nodular layer depth was measured in grid units (1 grid ≈ 0.05 mm), with higher values indicating thicker nodular zones. The best result was achieved in Trial 5, where a well-defined nodular graphite layer of about 338 grids (approximately 16.9 mm) was obtained, along with a smooth transition to flake graphite in the core. This highlights the importance of optimized coating composition and base iron parameters for grey iron castings.
| Trial | Carbon Equivalent (%) | Sulfur Content (%) | 1# Ferro-silico-RE (parts) | 75 Ferrosilicon (parts) | Graphite Morphology | Nodular Depth (grids) |
|---|---|---|---|---|---|---|
| 1 | 4.1 | 0.04 | 120 | 8 | Surface flake, rest vermicular | 0 |
| 2 | 4.26 | 0.04 | 100 | 10 | Surface flake, rest vermicular | 0 |
| 3 | 4.5 | 0.04 | 80 | 4 | Surface nodular (grade 2-3), transition vermicular, core flake | 270 |
| 4 | 4.1 | 0.02 | 80 | 10 | Surface flake, rest vermicular + few nodules | 0 |
| 5 | 4.26 | 0.02 | 120 | 4 | Surface nodular (grade 2), core flake | 338 |
| 6 | 4.5 | 0.02 | 100 | 8 | Vermicular + few nodules | 0 |
| 7 | 4.1 | 0.06 | 100 | 4 | Surface flake, vermicular zone, core flake | 0 |
| 8 | 4.26 | 0.06 | 80 | 8 | Surface flake, non-coated layer flake, rest vermicular | 0 |
| 9 | 4.5 | 0.06 | 120 | 10 | Surface flake, rest vermicular + few nodules | 0 |
Analysis of the results reveals several key insights. First, the amount of 1# ferro-silico-rare earth in the coating is critical for initiating graphite nodularization in grey iron castings. A minimum threshold of approximately 80% in the coating composition is necessary to achieve surface spheroidal graphite, as seen in the exploratory tests. However, within the effective range, variations in RE content do not drastically alter the graphite morphology, suggesting that other factors like inoculation and base iron composition play complementary roles. The role of rare earth elements can be modeled using a kinetic equation for nodularization, where the rate of graphite spheroid formation depends on the concentration of RE agents diffusing into the melt:
$$ \frac{dN}{dt} = k \cdot [\text{RE}] \cdot e^{-E_a/RT} $$
Here, \( N \) represents the number of graphite nodules, \( k \) is a rate constant, \( [\text{RE}] \) is the concentration of rare earth elements, \( E_a \) is the activation energy, \( R \) is the gas constant, and \( T \) is the temperature. This equation underscores the diffusion-controlled nature of the process in grey iron castings.
Second, the addition of 75 ferrosilicon as an inoculant significantly enhances the depth of the nodular layer. Inoculation promotes the formation of numerous graphite nucleation sites, which, combined with RE elements, leads to a finer and more uniform nodular structure. The effect can be quantified by the relationship between inoculant content and nodular layer thickness \( \delta \):
$$ \delta = \alpha \cdot [\text{Si}]_{\text{inoc}} + \beta \cdot [\text{RE}] $$
where \( \alpha \) and \( \beta \) are proportionality constants, and \( [\text{Si}]_{\text{inoc}} \) is the silicon content from the inoculant. From the comparative experiments, the optimal coating formulation for grey iron castings was determined as: 91% 1# ferro-silico-rare earth, 9% 75 ferrosilicon, with additions of 2.5% borax and 2.5% sodium fluoride (relative to RE alloy weight), 10% sodium silicate, and 0.4% PVB (relative to powder weight). This composition reliably produces a nodular layer exceeding 2.5 mm in thickness for suitable grey iron castings.
Third, the carbon equivalent and sulfur content of the base iron are pivotal. For grey iron castings, hypereutectic compositions (CE > 4.3%) are conducive to surface nodularization when treated with RE-based coatings. In contrast, hypoeutectic irons tend to form vermicular graphite or even surface flake layers, as observed in Trials 1, 2, and 4. This aligns with Evans et al.’s findings that cerium-induced nodularization is effective only in hypereutectic irons. The sulfur content must be kept below 0.04% to avoid excessive consumption of RE elements through sulfide formation, which inhibits nodularization. The relationship between sulfur content and nodularization efficiency can be expressed as:
$$ \eta = \frac{[\text{RE}]_{\text{eff}}}{[\text{RE}]_{\text{added}}} = 1 – \frac{[\text{S}]}{K_{\text{sp}}} $$
where \( \eta \) is the efficiency, \( [\text{RE}]_{\text{eff}} \) is the effective RE concentration available for graphite modification, \( [\text{RE}]_{\text{added}} \) is the total RE added, \( [\text{S}] \) is the sulfur content, and \( K_{\text{sp}} \) is the solubility product of RE sulfides. For grey iron castings with sulfur above 0.04%, \( \eta \) drops significantly, leading to incomplete nodularization.
Fourth, the microstructure of the surface-modified grey iron castings exhibits a distinct gradient. From the surface inward, the graphite morphology transitions from spheroidal to vermicular and then to flake, with clear boundaries between the nodular and vermicular zones. Correspondingly, the matrix structure changes from a mixture of pearlite and ferrite in the nodular layer, to ferrite in the transition zone, and to pearlite in the core. This gradient is attributed to the diffusion of RE and silicon from the coating into the molten iron. As these elements diffuse inward, they create a carbon-depleted region ahead of the nodular zone, promoting ferrite formation and reducing graphite precipitation. The diffusion process can be described by Fick’s second law:
$$ \frac{\partial C}{\partial t} = D \frac{\partial^2 C}{\partial x^2} $$
where \( C \) is the concentration of RE or silicon, \( D \) is the diffusion coefficient, \( t \) is time, and \( x \) is the distance from the surface. Solving this for boundary conditions relevant to grey iron castings yields a concentration profile that correlates with the observed microstructural layers.
Fifth, the wall thickness of grey iron castings influences the nodularization outcome. Thicker sections have higher modulus, leading to slower cooling rates, which allow more time for RE and inoculant diffusion, thereby increasing the nodular layer depth. However, slower cooling also promotes graphite growth, resulting in larger nodule diameters and fewer nodules per unit area. This trade-off can be modeled using the following empirical equation for nodular layer depth \( d \) as a function of wall thickness \( t \):
$$ d = d_0 + \gamma \cdot t – \lambda \cdot t^2 $$
Here, \( d_0 \) is a base depth, \( \gamma \) is a positive coefficient for diffusion enhancement, and \( \lambda \) is a negative coefficient for graphite coarsening. In thin sections, rapid cooling favors nodularization but may lead to surface flake layers due to insufficient RE penetration, whereas in thick sections, the nodular layer is deeper but with coarser graphite. For instance, in sharp corners of grey iron castings, accelerated cooling can induce local nodularization even when the main surface shows vermicular graphite.
Sixth, thermal analysis during solidification provides insights into the crystallization behavior. Cooling curves recorded from coated and uncoated mold surfaces show that the RE-based coating reduces the cooling rate by approximately 8°C/min, raises the eutectic temperature by about 20°C, and shortens the eutectic solidification time by 1.7 minutes. This indicates that the coating alters the nucleation and growth kinetics of graphite in grey iron castings. The modified cooling curve can be approximated by:
$$ T(t) = T_0 – \dot{T}_c \cdot t + \Delta T_{\text{eut}} \cdot \delta(t – t_{\text{eut}}) $$
where \( T(t) \) is the temperature over time, \( T_0 \) is the initial temperature, \( \dot{T}_c \) is the cooling rate (lower for coated surfaces), \( \Delta T_{\text{eut}} \) is the increase in eutectic temperature, \( t_{\text{eut}} \) is the eutectic start time, and \( \delta \) is a Dirac delta function representing the thermal effect. Additionally, the coating promotes prior austenite dendrite formation before eutectic reaction, as rare earth elements create conditions for austenite growth, which subsequently influences graphite morphology.
Seventh, microanalysis of the nodular graphite reveals that rare earth elements and sulfur are concentrated around graphite particles and within inclusions. Energy-dispersive spectroscopy (EDS) and wavelength-dispersive spectroscopy (WDS) line scans show peaks of cerium and sulfur at inclusion sites, confirming the formation of RE sulfides that deplete sulfur from the matrix. This localized distribution supports the mechanism where RE elements first react with sulfur, and the remaining RE promotes graphite spheroidization. For grey iron castings, this means that controlling sulfur content is essential to maximize the nodularizing effect of RE-based coatings.
In summary, this research demonstrates a viable method for surface graphite nodularization in grey iron castings using RE-based mold coatings. The process involves applying a coating containing 1# ferro-silico-rare earth and 75 ferrosilicon to the mold surface, then pouring hypereutectic iron with low sulfur content (<0.04%). The resulting grey iron castings exhibit a composite microstructure with a surface nodular graphite layer over 2.5 mm thick, a transition vermicular layer, and a flake graphite core. The optimal coating composition has been identified, and the effects of carbon equivalent, sulfur, wall thickness, and cooling conditions have been elucidated through experimental data and theoretical models.
The implications of this work are significant for industries relying on grey iron castings, such as automotive, machinery, and metallurgy. By enabling tailored surface properties, this technique can enhance the performance and longevity of components like ingot molds, engine blocks, and gears. Future research could explore the scalability of the process for large-scale production, the use of alternative RE sources, and the integration with other surface treatments. Additionally, computational modeling of diffusion and solidification could further optimize coating designs for specific grey iron castings applications. Overall, the RE-based coating approach offers a promising route to achieve high-performance grey iron castings with graded microstructures, meeting diverse engineering demands.
