My research focuses on the long-term durability of ductile iron castings used in coastal wind power equipment. Wind turbine components such as hubs, wheel shafts, and planetary carriers are often manufactured from QT400-18L, a ferritic ductile iron grade that provides excellent elongation and acceptable strength. Coastal environments, however, expose these components to sea water, tidal moisture, salt spray, and high atmospheric humidity. Since critical components of a wind turbine are expected to remain in service for about twenty years without major replacement, the corrosion resistance of ductile iron castings is as important as their mechanical properties. In order to improve the performance of ductile iron castings under such severe conditions, I studied the influence of multi-scale ceramic particles, abbreviated as MSCP, on the humidity-dependent corrosion behavior of QT400-18L.
The idea of adding ceramic particles to molten metal is not new, but the use of activated multi-scale ceramic particles in ductile iron castings is a promising route for improving both microstructure and corrosion properties. In this study, I added MSCP to the melt of QT400-18L at concentrations between 0.05% and 0.15% by mass. The particles were based on activated multi-scale composite SiC. I then evaluated the corrosion resistance under controlled humidity conditions, because moisture is one of the main degradation factors for ductile iron castings in coastal wind farms. My objective was to determine whether MSCP can reduce the corrosion rate of ductile iron castings and to explain the mechanisms behind the improved behavior.
Materials and Experimental Procedure
All test samples were produced from a commercial QT400-18L ductile iron melt. The chemical composition of the base material is given in Table 1. The carbon and silicon contents are typical for ductile iron castings, while the magnesium and rare earth additions ensure nodular graphite formation. The base composition was kept constant throughout the experiments so that the only intentional variable was the MSCP addition.
| Element | Mass fraction / % |
|---|---|
| C | 3.56–3.68 |
| Si | 2.19–2.40 |
| Mn | 0.148–0.150 |
| P | 0.030–0.032 |
| S | 0.010–0.012 |
| Mg | 0.029–0.062 |
| Rare earth (Re) | 0.032–0.047 |
| Fe | Balance |
MSCP were added to the liquid iron at three different levels: 0.05%, 0.10%, and 0.15% by mass. A control batch without MSCP was also produced. After casting, test coupons were machined from the center of the cast blocks. The final specimens were circular discs with a diameter of 25 mm and a thickness of 3 mm. The exposed surface area was calculated from the specimen geometry. The surface condition of each specimen was carefully controlled by grinding and cleaning before the corrosion tests began.
The humidity corrosion test was carried out in a constant-temperature constant-humidity chamber. According to the standard procedure, the temperature was fixed at 60 °C and the test duration was 168 h. I selected four relative humidity levels: 60%, 80%, 90%, and 98%. For each combination of MSCP addition and humidity, three parallel specimens were tested. The corrosion rate was obtained from the change in specimen mass during exposure. The test matrix is summarized in Table 2.
| MSCP addition / % | Relative humidity / % | Temperature / °C | Duration / h |
|---|---|---|---|
| 0 (control) | 60, 80, 90, 98 | 60 | 168 |
| 0.05 | 60, 80, 90, 98 | 60 | 168 |
| 0.10 | 60, 80, 90, 98 | 60 | 168 |
| 0.15 | 60, 80, 90, 98 | 60 | 168 |
The corrosion rate was calculated using the mass-change method. I used the following equation:
$$ V = \frac{M_2 – M_1}{A t} $$
In this equation, \(V\) is the corrosion rate in \(\mathrm{g/(m^2 \cdot h)}\), \(M_1\) is the initial mass of the specimen in grams, \(M_2\) is the mass after exposure in grams, \(A\) is the exposed surface area in square meters, and \(t\) is the exposure time in hours. Because corrosion products may remain on the surface, the measured mass change represents the net accumulation of reaction products. The final value for each condition was taken as the average of three parallel specimens.
For microstructural analysis, I examined the graphite morphology using optical microscopy combined with image analysis software. The nodularity grade, nodularity rate, and graphite size grade were evaluated according to the relevant ductile iron standard. The matrix constitution, especially the amount of ferrite, was also determined. Scanning electron microscopy was used to observe the corrosion product morphology. Energy-dispersive X-ray spectroscopy, EDS, was used to identify the elements present in the corrosion layers. X-ray diffraction, XRD, was used to identify the crystalline corrosion products.
Results and Discussion
Effect of MSCP on Graphite Morphology in Ductile Iron Castings
The first important observation is that MSCP had a clear effect on the graphite structure of ductile iron castings. In the control material without MSCP, the graphite nodules were not uniform in size and many nodules appeared irregular. Some vermicular graphite particles were also detected. When MSCP were added, the graphite became smaller and more spherical. The nodularity level improved, and the graphite distribution became more homogeneous.
| MSCP addition / % | Nodularity grade | Nodularity rate / % | Graphite size grade |
|---|---|---|---|
| 0 | 4 | 79 | 6 |
| 0.05 | 3 | 82 | 7 |
| 0.10 | 2 | 91 | 7 |
| 0.15 | 3 | 86 | 7 |
As shown in Table 3, the best graphite morphology was obtained at an MSCP addition of 0.10%. At this concentration, the nodularity grade reached the high level of grade 2, and the nodularity rate exceeded 90%. The addition of 0.05% and 0.15% also improved the graphite structure compared to the control, but the effect was slightly weaker. The refinement of graphite is important because graphite particles act as cathodic sites during corrosion. Smaller and more regular graphite nodules reduce the effective cathodic area and make the graphite-matrix interface more regular, which can slow down the development of localized corrosion.
Effect of MSCP on the Metallic Matrix
In addition to modifying the graphite phase, MSCP changed the matrix of the ductile iron castings. The standard QT400-18L has a predominantly ferritic matrix, but small amounts of pearlite are often present. In my experiments, the control material contained about 79.16% ferrite. After the addition of MSCP, the ferrite content increased to more than 85% for all additions. Table 4 summarizes the measured ferrite contents and the corresponding approximate pearlite contents.
| MSCP addition / % | Ferrite content / % | Pearlite content (approx.) / % |
|---|---|---|
| 0 | 79.16 | 20.84 |
| 0.05 | 86.42 | 13.58 |
| 0.10 | 85.84 | 14.16 |
| 0.15 | 86.95 | 13.05 |
The increase in ferrite content is beneficial for corrosion resistance of ductile iron castings. Pearlite consists of ferrite and cementite, and these two phases have different electrochemical potentials. The presence of pearlite increases the number of galvanic couples within the matrix and accelerates local corrosion. By promoting a more ferritic microstructure, MSCP reduce the amount of pearlite and therefore weaken the micro-galvanic effect between ferrite and cementite. This is one of the important reasons why MSCP improve the corrosion behavior of ductile iron castings.
Corrosion Rate under Different Humidity Conditions
The corrosion rate of QT400-18L ductile iron castings increased with increasing relative humidity. This result was expected because atmospheric corrosion requires moisture on the metal surface. At low humidity, the amount of adsorbed water is limited, and the electrochemical reactions cannot proceed rapidly. At high humidity, a continuous or nearly continuous electrolyte film can form on the surface, and the corrosion rate becomes much higher.
An interesting feature in the corrosion rate data is the existence of an inflection point. The inflection point appeared near 80% relative humidity. Below this humidity, the corrosion rate increased slowly with humidity, and the presence of MSCP did not cause a significant difference between the control and the modified samples. Above this humidity, the corrosion rate increased much more sharply. However, the increase was smaller for the samples containing MSCP than for the control. In other words, MSCP became more effective when the environment was more humid.
The relative reduction in corrosion rate at 98% humidity is shown in Table 5. Compared with the control, the addition of 0.05%, 0.10%, and 0.15% MSCP reduced the corrosion rate by 10.10%, 21.21%, and 29.29%, respectively. The highest addition of 0.15% gave the lowest corrosion rate and therefore the best corrosion protection among the tested formulations.
| MSCP addition / % | Reduction in corrosion rate at 98% RH / % |
|---|---|
| 0.05 | 10.10 |
| 0.10 | 21.21 |
| 0.15 | 29.29 |
To express this effect mathematically, I used the following formula for the relative corrosion rate reduction:
$$ R_d = \frac{V_0 – V_M}{V_0} \times 100\% $$
Here, \(V_0\) is the corrosion rate of the control specimen without MSCP, \(V_M\) is the corrosion rate of the specimen with a given MSCP addition, and \(R_d\) is the percentage reduction. This calculation made it possible to compare the effectiveness of different MSCP additions at the same humidity level.
The humidity inflection point near 80% can be attributed to two factors. The first is the ability of moisture to form droplets on the metal surface. When the relative humidity is below 80%, water vapor is not easily condensed into a continuous thin film. Even if a film forms, it is too thin to provide a stable electrolyte path for the anodic and cathodic reactions. As a result, the corrosion rate remains moderate. The second factor is the electrochemical corrosion under thin electrolyte films. At high humidity, water molecules condense more easily on active surface sites, such as graphite-matrix interfaces, and form larger droplets. Larger droplets cover a large anode area, and the anodic metal dissolution process is accelerated. The corrosion mechanism at high humidity can be described by the following anodic and cathodic reaction equations:
$$ \mathrm{Fe} \rightarrow \mathrm{Fe^{2+}} + 2e^- $$
$$ \mathrm{O_2} + 2\mathrm{H_2O} + 4e^- \rightarrow 4\mathrm{OH^-} $$
In the presence of a thin electrolyte layer, the cathodic reaction is usually controlled by oxygen diffusion. The limiting current density for oxygen reduction can be written as:
$$ i_L = \frac{n F D c}{\delta} $$
where \(n\) is the number of electrons involved in oxygen reduction, \(F\) is Faraday’s constant, \(D\) is the diffusion coefficient of dissolved oxygen, \(c\) is the oxygen concentration, and \(\delta\) is the effective diffusion path length through the electrolyte film. The value of \(\delta\) depends strongly on the amount of water on the surface. At very low humidity, the film is not continuous enough for efficient transport of dissolved oxygen; at very high humidity, the film may become thicker and more continuous, which supports more active electrochemical reactions.
Corrosion Product Morphology and Composition
After 168 h of exposure, I observed the surface of the specimens by scanning electron microscopy. The corrosion products were not uniformly distributed. Instead, most corrosion products were concentrated around the graphite nodules. This is because graphite is electrochemically nobler than the ferrite matrix, so graphite particles act as local cathodes while the surrounding iron matrix acts as an anode. The galvanic corrosion process therefore tends to occur at the graphite-matrix interface and spreads outward into the matrix.
A remarkable feature was the appearance of corrosion products in a double-ring configuration around the graphite particles. This double-ring morphology is related to the formation and spreading of microscopic electrolyte droplets. In atmospheric corrosion studies, it has been reported that when a small droplet forms on a metal surface, the main droplet can develop smaller micro-droplets around it. These micro-droplets arise because the concentrated ionic solution inside the main droplet changes the local surface tension and electrochemical activity. As the micro-droplets spread outward, they create concentric anodic and cathodic zones. The inner ring and outer ring correspond to different stages of droplet spreading and electrochemical activity.
The presence of graphite-matrix defects or irregularities provides high-energy sites where moisture is preferentially adsorbed. When relative humidity is high enough, microscopic droplets form at these sites. The main droplet becomes a region of anodic dissolution, while the edge of the droplet becomes a cathodic region. The corrosion products accumulate around the graphite and appear as a ring. As the droplet spreads, another ring of corrosion products can form around the first one, creating the observed double-ring pattern.
Energy-dispersive X-ray spectroscopy was used to analyze the corrosion products at 98% humidity after 168 h. Table 6 presents the representative atomic percentages for the control sample and the sample with 0.15% MSCP.
| Condition | C / at.% | O / at.% | Fe / at.% | Si / at.% |
|---|---|---|---|---|
| 0% MSCP | 46.16 | 33.77 | 18.46 | 1.16 |
| 0.15% MSCP | 21.92 | 31.28 | 43.59 | 3.21 |
The corrosion products contained carbon, oxygen, iron, and silicon. The carbon came from the graphite phase and from reaction products, while the oxygen and iron indicated the formation of iron oxides. The silicon content was likely related to the silicon present in the ductile iron matrix and possibly to the SiC-based MSCP particles. The control sample showed a higher carbon content in the corrosion layer, which is consistent with the more extensive corrosion around the graphite nodules. The sample with 0.15% MSCP showed a higher iron content, indicating that the corrosion layer contained a larger amount of iron oxides and probably a more protective scale.
X-ray diffraction analysis revealed that the main corrosion products were iron oxides, specifically FeO, Fe2O3, and Fe3O4. The diffraction peak intensities were higher for the control sample than for the samples containing MSCP. Since the peak intensity is related to the quantity of the corresponding phase, this result suggests that the control sample formed more corrosion products. The addition of MSCP reduced the overall amount of corrosion products, particularly at additions of 0.10% and 0.15%. This is consistent with the mass-change measurements and confirms that MSCP improve the corrosion resistance of ductile iron castings under high humidity conditions.
Mechanism of Corrosion Resistance Improvement
The improvement in corrosion resistance of ductile iron castings caused by MSCP can be understood through several mechanisms. The first mechanism is the refinement of graphite. Graphite acts as a cathode in the micro-galvanic corrosion system of ductile iron. The potential difference between graphite and the ferrite matrix is a driving force for corrosion. When the graphite particles are smaller, more spherical, and more uniformly distributed, the cathode area is reduced and the graphite-matrix interface becomes more regular. This reduces the number of high-energy sites where moisture can preferentially adsorb. As a result, the initiation and spreading of corrosion around the graphite nodules become more difficult.
The second mechanism is related to the matrix composition. MSCP promote the formation of ferrite and reduce the amount of pearlite in ductile iron castings. Pearlite contains alternating layers of ferrite and cementite. Cementite is nobler than ferrite, so pearlite itself acts as a galvanic system. The presence of pearlite creates additional galvanic couples and accelerates corrosion. By increasing the ferrite content, MSCP reduce the number of pearlite-related galvanic cells and therefore improve the corrosion resistance of the material.
The third mechanism involves electrochemical potential. I measured or inferred that the activated MSCP can introduce metallic elements with a higher electrode potential into the iron matrix. These elements may dissolve into ferrite and raise its electrode potential. The electrochemical potential of a metal electrode can be described by the Nernst equation. For the iron electrode, the potential can be expressed as:
$$ E_{\mathrm{Fe}} = E_{\mathrm{Fe}}^0 + \frac{RT}{2F} \ln [\mathrm{Fe^{2+}}] $$
Here, \(E_{\mathrm{Fe}}^0\) is the standard electrode potential of iron, \(R\) is the universal gas constant, \(T\) is the absolute temperature, and \([\mathrm{Fe^{2+}}]\) is the activity of ferrous ions in the electrolyte. If the ferrite matrix is alloyed with elements that are more noble than iron, its electrode potential becomes more positive. This reduces the potential difference between the graphite cathode and the ferrite anode. The potential difference for the galvanic cell can be written as:
$$ \Delta E = E_c – E_a $$
where \(E_c\) is the cathode potential and \(E_a\) is the anode potential. The corrosion current is then approximately equal to:
$$ I_{\mathrm{corr}} = \frac{\Delta E}{R_p} $$
where \(R_p\) is the polarization resistance of the system. A smaller potential difference leads to a smaller driving force for galvanic corrosion, and therefore the corrosion current is reduced.
The fourth mechanism is the formation of a more compact and protective oxide layer. MSCP can act as nucleating agents during solidification and promote a finer, more homogeneous microstructure. The finer microstructure helps to form a continuous and dense oxide film on the surface. This film provides a physical barrier that restricts the diffusion of corrosive species and slows down the electrochemical reactions. The presence of passivating metallic elements in the activated MSCP can further improve the stability and compactness of the oxide film.
In addition, the uniform distribution of MSCP in the melt reduces the undercooling required for nucleation and refines the solidification structure. The refined grain structure of ductile iron castings tends to have fewer defects and more uniform stress distribution, which also contributes to better corrosion resistance. The interaction between MSCP and the liquid iron may promote the formation of fine, well-distributed graphite particles, which is one of the most important microstructural benefits for the corrosion performance of ductile iron castings.
The relationship between corrosion rate and humidity can be understood using capillary condensation theory. At a given relative humidity, water vapor can condense in small pores and crevices according to the Kelvin equation:
$$ r_K = \frac{2 \gamma V_m}{RT \ln(\mathrm{RH}^{-1})} $$
In this equation, \(r_K\) is the critical radius for capillary condensation, \(\gamma\) is the surface tension of water, \(V_m\) is the molar volume of water, and \(\mathrm{RH}\) is the relative humidity expressed as a fraction. This equation shows that at higher relative humidity, smaller and even larger pores can be filled with liquid water. This is why the corrosion rate of ductile iron castings increases sharply when the humidity exceeds a critical value. At humidity above the critical value, the amount of condensed water is sufficient to establish active electrochemical cells on the surface.
For the observed double-ring corrosion product morphology, the micro-droplet phenomenon plays a key role. When a main droplet forms at an active site, the high concentration of metal ions inside the droplet changes the local chemical environment. The droplet then spreads outward and small micro-droplets appear around its perimeter. These micro-droplets continue to spread outward, causing the formation of concentric corrosion product rings. The corrosion process is therefore not a uniform front but a local and progressive phenomenon. The regular graphite-matrix interface produced by MSCP helps to suppress this micro-droplet spreading because it reduces the number of active sites available for droplet nucleation.
The effect of MSCP can also be discussed in terms of the electrode kinetics. The Butler-Volmer equation describes the current density at the electrode as a function of the overpotential:
$$ i = i_0 \left[ \exp\left( \frac{\alpha_a n F \eta}{RT} \right) – \exp\left( -\frac{\alpha_c n F \eta}{RT} \right) \right] $$
Here, \(i_0\) is the exchange current density, \(\alpha_a\) and \(\alpha_c\) are the anodic and cathodic transfer coefficients, \(n\) is the number of electrons transferred, \(F\) is Faraday’s constant, \(\eta\) is the overpotential, \(R\) is the gas constant, and \(T\) is the absolute temperature. The corrosion rate is determined by the balance between the anodic and cathodic reactions. When the potential difference between graphite and matrix is reduced by MSCP, the anodic overpotential at a given corrosion potential becomes smaller, and the corresponding anodic current is reduced. This electrochemical explanation is consistent with the observed reduction in corrosion rate for ductile iron castings containing MSCP.
It should also be noted that the addition of 0.15% MSCP was the most effective in reducing the corrosion rate, even though the graphite nodularity was slightly lower than that obtained with 0.10% MSCP. This suggests that the matrix composition and the formation of a protective oxide layer may play a more important role than graphite morphology alone when the MSCP addition is increased. The higher MSCP addition may introduce a larger number of fine particles into the metal matrix, which act as barriers to corrosion propagation and improve the stability of the passive-like film.
Conclusions
Based on the experiments conducted in this work, I draw the following conclusions:
First, the addition of multi-scale ceramic particles to QT400-18L ductile iron castings significantly improves the graphite morphology. The graphite nodules become smaller, more spherical, and more uniformly distributed. The optimum nodularity was observed at an MSCP addition of 0.10%, where the nodularity rate reached 91%.
Second, MSCP increase the ferrite content of ductile iron castings and reduce the amount of pearlite. The highest ferrite content was 86.95%, obtained at 0.15% MSCP addition. This change in matrix composition reduces the number of galvanic cells formed between ferrite and cementite.
Third, the corrosion rate of ductile iron castings increases with relative humidity, but the increase is less pronounced when MSCP are present. The critical humidity in this study was near 80% RH. Below this value, the effect of MSCP was small; above this value, MSCP clearly reduced the corrosion rate. At 98% RH, the corrosion rate reduction reached 29.29% for the 0.15% MSCP addition.
Fourth, the corrosion products of ductile iron castings are mainly iron oxides, including FeO, Fe2O3, and Fe3O4. Most corrosion products are concentrated around graphite nodules, and the corrosion morphology often shows a characteristic double-ring pattern. This pattern is related to the formation and spreading of microscopic water droplets on the metal surface.
Fifth, the mechanisms responsible for the improved corrosion resistance of MSCP-modified ductile iron castings include graphite refinement, increased ferrite content, reduced galvanic potential difference, and the formation of a more protective oxide layer. These mechanisms work together to reduce the electrochemical activity of the material and improve the durability of ductile iron castings in coastal atmospheric environments.
The results of this study demonstrate that multi-scale ceramic particles offer a promising method for enhancing the corrosion resistance of ductile iron castings used in coastal wind power applications. Future work should combine high humidity exposure with salt spray and cyclic wet-dry tests to simulate the real coastal service environment more accurately. Nevertheless, the present findings already provide valuable guidance for the design and production of more durable ductile iron castings for offshore and coastal wind power equipment.