In our foundry’s pursuit of advanced ductile iron production techniques, we implemented two novel processes: pneumatic desulfurization and mold-on-demand nodularization. While these promised efficiency gains, they introduced a significant and recurring casting defect issue in our crankshaft castings. Specifically, after machining, white and black spot-like imperfections were consistently observed on the upper cope surfaces of the main and connecting rod journals. These defects, which we termed “White Spots” and “Black Spots,” severely impacted product quality and yield, necessitating a thorough investigation into their nature and root cause. This article details our first-person analysis of this perplexing casting defect.
The macroscopic appearance of these defects was distinctive. The White Spots, approximately 1-2 mm in size, consisted of a white powdery substance that could be easily dislodged with a needle, leaving behind a pit. The Black Spots were identified as areas of concentrated micro-shrinkage and slag inclusions. To systematically characterize this casting defect, we employed a multi-technique analytical approach: hot acid etching, spectroscopic analysis, and metallographic examination.
Defect Characterization and Initial Analysis
The initial step involved macro and micro examination to understand the physical and chemical nature of the casting defect. The results are summarized below.
| Defect Type | Macroscopic Appearance | Post-Etching Observation | Initial Hypothesis |
|---|---|---|---|
| White Spot | White powder, pit upon removal. | Powder dissolved, revealing underlying shrinkage and pores. | Magnesium-rich compound accumulation. |
| Black Spot | Dark, dense cluster of spots. | Area of shrinkage expanded, slag more evident. | Slag inclusion cluster (likely rare earth based). |
Spectroscopic and Chemical Analysis
To identify the elemental composition of the defect zones, we performed spectroscopic analysis directly on the White and Black Spot regions, comparing them to sound areas of the casting. The goal was to trace the origin of this specific casting defect.
| Analysis Area | Mg Content | Ce Content | Si Content | Mn Content | Inference |
|---|---|---|---|---|---|
| White Spot Zone | Spectrum line very bright (High) | ~0.01-0.03% | ~2.0% | ~0.6% | Severe Mg enrichment. |
| Sound Area (far from defect) | Normal/Uniform | ~0.01-0.02% | ~2.0% | ~0.6% | Base composition. |
| Black Spot Zone | Spectrum line weak (Low) | ~0.08% | ~2.0% | ~0.6% | Ce enrichment, Mg depletion. |
| Sound Area (far from defect) | Normal | ~0.01-0.02% | ~2.0% | ~0.6% | Base composition. |
The spectroscopic data led to a clear preliminary conclusion: The White Spot powder was primarily composed of magnesium oxide (MgO). The Black Spot areas were rich in rare earth (Ce) oxides, indicating they were slag zones. This pointed directly to post-nodularization slag as the source of the casting defect.
Metallographic Investigation and Mechanism Elucidation
Metallography provided the microstructural context for this casting defect. Samples containing the defects were polished and examined. The findings were critical in linking the chemical analysis to the physical manifestation of the flaw.
White Spot Microstructure: After etching, the pits still contained residual white particles with an octahedral, sharp-angled morphology—characteristic of MgO. The pits and surrounding areas contained a significant amount of gray oxide and silver-gray sulfide inclusions. Notably, the matrix adjacent to these defects showed clear evidence of surface oxidation and decarburization. Furthermore, the regions surrounding both White and Black Spots exhibited severe graphite flotation, characterized by exploded, dendritic, and thick flake graphite forms.
Black Spot Microstructure: The etched samples revealed that the shrinkage cavities at the Black Spots were bordered by a network of gray rare earth oxides and titanium nitride inclusions. The surrounding area was, again, a zone of graphite flotation containing various degenerate graphite shapes along with complex magnesium sulfide inclusions.
The combined data allows us to model the thermodynamic conditions promoting this casting defect. The formation of MgO, a primary slag, is highly exothermic and occurs readily upon magnesium treatment:
$$ 2Mg_{(in\ Fe)} + O_{2\ (dissolved)} \rightarrow 2MgO_{(s)} \quad \Delta H \ll 0 $$
Similarly, the oxidation of cerium contributes to slag formation:
$$ 4Ce_{(in\ Fe)} + 3O_{2\ (dissolved)} \rightarrow 2Ce_2O_{3\ (s)} $$
The formation of sulfides, often found adjacent to these slags, follows:
$$ Mg_{(in\ Fe)} + S_{(in\ Fe)} \rightarrow MgS_{(s)} $$
$$ MgS_{(s)} + Ce_2O_{3\ (s)} \rightarrow \text{Complex Slag Compounds} $$
The presence of these hard, non-metallic particles creates stress concentrators and disrupts the matrix continuity. The associated graphite flotation, a separate but related casting defect, results from the local enrichment of carbon in the last-to-freeze regions, which are also the preferred sites for slag accumulation. The carbon equivalent (CE) in these zones can be locally much higher. The standard carbon equivalent formula is:
$$ CE = \%C + \frac{1}{3}(\%Si + \%P) $$
In flotation zones, effective %C near the slag particles is elevated, leading to the precipitation of degenerate graphite forms ahead of the solidification front.
Root Cause Analysis: Process-Related Genesis of the Defect
The core of this casting defect problem lies in the unique sequence and kinetics of the mold-on-demand nodularization process. Our analysis identified a chain of causality.
- High-Purity, Low-Sulfur Iron: The preliminary pneumatic desulfurization was highly effective, reducing sulfur to very low levels (“double-zero” grade). While beneficial for graphite morphology, this eliminated the beneficial slag-coalescing effect of MnS formation during treatment. The conventional slag formed from sulfur reactions acts as a collector for other oxides.
- Short Process Window: In mold-on-demand treatment, the time between magnesium/rare earth addition and the complete filling and solidification of the casting is extremely short.
- Let $t_{float}$ be the time required for a slag particle of radius $r$ and density $\rho_s$ to float a distance $h$ through an iron melt of density $\rho_{Fe}$ and viscosity $\eta$. Stokes’ law gives an approximation:
$$ v = \frac{2gr^2(\rho_{Fe} – \rho_s)}{9\eta}, \quad t_{float} = \frac{h}{v} $$
- For fine, powder-like slag particles (small $r$) generated during treatment, $t_{float}$ is large. The actual available time $t_{process}$ (from treatment to solidification) is very small. Therefore:
$$ t_{float} \gg t_{process} $$
- Slag Morphology and Entrapment: The combination of clean iron (no coalescing agents) and violent reaction kinetics produced a fine, dispersed, powder-like primary slag instead of a coalesced, buoyant, viscous slag mass. This fine slag was easily carried by the turbulent metal flow into the mold cavity. During solidification, these particles were pushed to the last-freezing regions—the upper surfaces of the casting (cope side) or dead zones near cores.
The final casting defect manifestation is then a result of post-casting machining:
- Black Spot: A cluster of brittle slag particles (Ce-oxides, etc.) and associated micro-shrinkage. During machining, some particles are torn out, revealing a porous, dark-appearing area.
- White Spot: A localized concentration of powdery MgO. Machining exposes this powder directly on the surface, creating the white, crumbly appearance.
The detrimental effect of this casting defect on mechanical properties, particularly fatigue strength and impact toughness, is severe. The slag particles act as potent stress raisers. The fatigue limit $\sigma_{w}$ can be empirically related to the defect size $\sqrt{area}$ (projected area of the defect perpendicular to the stress axis):
$$ \sigma_{w} = \frac{C}{( \sqrt{area} )^{1/6}} $$
Where $C$ is a constant. Even small slag clusters significantly reduce $\sigma_{w}$.
Comprehensive Solution Strategy and Process Optimization
To eliminate this casting defect, the solution must address the root causes: slag formation, morphology, and removal. The objective is to modify the slag from a fine powder to a cohesive, viscous agglomerate that can be effectively trapped before entering the casting cavity. Our implemented solution involved a multi-component slag modifier added to the reaction chamber.
| Additive | Chemical Formula | Primary Function | Mechanism |
|---|---|---|---|
| Fluorspar | CaF2 | Fluxing Agent / Slag Thinner (initially) | Lowers slag melting point and viscosity, promoting coalescence of fine particles. |
| Sodium Fluosilicate | Na2SiF6 | Fluxing & Agglomeration Agent | Reacts to form low-melting compounds that bind fine oxides into a single slag mass. |
| Glass Powder / Fibers | SiO2-based | Viscosity Modifier & Collector | Melts to form a viscous, silicate-based network that entraps fine particles, increasing overall slag bulk and buoyancy. |
The combined action of these additives transforms the slag system. The process can be described in stages:
- Coalescence: Fine MgO and Ce2O3 particles are wetted and aggregated by the low-melting flux phase.
$$ \text{MgO}_{(fine)} + \text{Ce}_2\text{O}_{3(fine)} + \text{Flux} \rightarrow \text{Viscous Slag Agglomerate} $$ - Growth and Separation: The agglomerate grows, and its buoyancy velocity $v$ increases significantly (as $r^2$ in Stokes’ law). In a well-designed reaction chamber with centrifugal flow or baffles ($\text{Function: Centrifugal Separation}$), the forces acting on the slag particle of mass $m$ in a rotating flow with angular velocity $\omega$ at radius $R$ can be simplified. The centrifugal acceleration $a_c$ is:
$$ a_c = \omega^2 R $$
This acceleration, much larger than gravity $g$, forces the dense slag agglomerate toward the center of the vortex or against trap walls, while the cleaner metal flows to the outlet. - Entrapment: The designed geometry of the treatment chamber includes dams, baffles, or a quiet “holding” zone where the coalesced slag is permanently retained.
The effectiveness of this solution is measured by the final casting defect count. After implementation, the occurrence rate of White and Black Spots in crankshafts dropped effectively to zero. The key process parameters before and after the fix are contrasted below:
| Parameter | Original Process (Defect-Prone) | Optimized Process (Defect-Free) |
|---|---|---|
| Slag Morphology | Fine, dispersed powder. | Coarse, viscous, cohesive agglomerate. |
| Primary Slag Composition | MgO, Ce2O3 powders. | Complex calcium-alumino-silicate slag containing MgO & RE oxides. |
| Slag Removal Mechanism | Gravity flotation only (ineffective). | Coalescence + Centrifugal/Mechanical separation in chamber. |
| Metal Cleanliness Post-Treatment | High inclusion count (fine). | Low inclusion count. |
| Crankshaft Defect Rate | High (Batch occurrence). | Negligible. |
Generalized Learnings and Preventative Framework
This investigation into a specific casting defect yields broader principles for foundries employing in-mold or late stream treatment processes. The fundamental challenge is managing the by-products of violent metallurgical reactions within a constrained timeframe. A preventative framework can be established:
1. Slag Engineering is Critical: The thermodynamic drive to form stable oxides (MgO, RE-oxides) is unavoidable. The goal is not to prevent their formation entirely but to control their physical state. The slag system must be designed, via additives, to have:
- A low enough initial viscosity to promote particle coalescence.
- A high enough final viscosity at casting temperature to prevent re-entrainment.
- A density sufficiently different from the molten metal to aid separation.
The ideal slag viscosity $\eta_{slag}$ should follow a trajectory over time $t$:
$$ \eta_{slag}(t) : \text{Low} \rightarrow \text{Moderate} \rightarrow \text{High (at entrapment site)} $$
2. Process Kinetics Dictate Design: The available time $t_{process}$ is the master parameter. All separation mechanisms (gravity, centrifugal, filtration) must be designed with this constraint. The required floating/separating velocity $v_{required}$ becomes:
$$ v_{required} = \frac{h_{separation}}{t_{process}} $$
The chamber must be designed to impart this velocity, often requiring passive (baffles, weirs) or active (rotation) means beyond simple gravity.
3. Interaction with Other Defects: A primary slag-related casting defect often nucleates or attracts secondary issues. In our case, slag particles acted as:
- Heterogeneous nucleation sites for graphite flotation (local carbon enrichment).
- Barriers to feeding, promoting localized micro-shrinkage.
- Initiation points for surface oxidation during solidification.
Thus, solving the primary slag defect frequently mitigates a cluster of associated quality problems.
In conclusion, the resolution of the White and Black Spot casting defect underscores a fundamental principle in advanced casting: process innovation must be accompanied by comprehensive by-product management. By shifting the focus from merely achieving a metallurgical reaction (like nodularization) to actively engineering the entire reaction product system—including slag phase morphology and dynamics—it is possible to eliminate even the most persistent and damaging casting defect, thereby achieving high integrity in critical cast components like crankshafts.