In our foundry, during the trial implementation of two novel processes—pneumatic desulfurization and in-mold spheroidization—we encountered atypical metal casting defects manifesting as white and black spots on the cast components. These metal casting defects were particularly observed on ductile iron crankshafts after machining, appearing on the upper cope sections of the main bearing journals and connecting rod journals. The pervasive occurrence of these defects severely compromised product quality and yield, necessitating a thorough investigation into their nature and root causes. This article details our first-person analytical journey, employing various metallurgical techniques to unravel the origins of these specific metal casting defects and the subsequent process modifications that successfully eliminated them.
The defects, termed “white spots” and “black spots,” presented distinct macro-morphologies. White spots appeared as patches with linear dimensions approximately between 2 to 10 mm. These patches contained a white, powdery substance that could be easily dislodged with a needle, leaving behind pits or cavities. In contrast, black spots corresponded to areas of concentrated micro-shrinkage porosity and slag inclusions. The initial hypothesis pointed towards these being slag-related metal casting defects introduced during the molten metal treatment stage.
To characterize these metal casting defects, a series of experiments were conducted: hot acid etching, spectroscopic analysis, and metallographic examination. The hot acid etching test involved immersing samples in a 50% hydrochloric acid aqueous solution heated to 70±5°C for 30 minutes. After etching, the samples were rinsed, neutralized with a 5% sodium hydroxide solution, dried, and observed. For white spot samples, the etching caused partial dissolution of the white material, revealing underlying substantial porosity and cavities at the defect sites. For black spot samples, the etching intensified the appearance of the defects, enlarging the visible areas of shrinkage porosity.
Spectroscopic analysis (visual emission spectroscopy) provided crucial chemical clues. At the white spot locations, the spectral line for magnesium (Mg) was exceptionally bright, indicating a significant localized enrichment of this element. Quantitative estimates suggested the Mg content in these zones could be several times higher than in the sound matrix, where it was typically around 0.04%. Other elements like cerium (Ce) showed normal levels. At black spot locations, the Mg line was notably weaker and less intense, while lines for elements like silicon (Si) and cerium suggested the presence of complex slag, predominantly silicate-based. The spectroscopic data led to the preliminary inference that the white powder was likely magnesium oxide (MgO), while the black spots consisted of slag rich in rare earth (RE) and magnesium silicates.
| Defect Type | Location Analyzed | Key Spectral Observation | Inferred Primary Compound | Estimated Enrichment/Depletion |
|---|---|---|---|---|
| White Spot | Defect Core | Very intense Mg line | Magnesium Oxide (MgO) | Strong Mg enrichment |
| White Spot | Sound Matrix | Normal Mg, Ce lines | N/A | Mg ~0.04% |
| Black Spot | Defect Core | Weak Mg line, strong Si, Ce lines | RE-Mg Silicate Slag | Mg depletion, Si/RE enrichment |
| Black Spot | Sound Matrix | Normal elemental lines | N/A | Baseline composition |
Metallographic examination after polishing and etching provided microstructural confirmation. In white spot areas, after acid etching, residual white particles with an octahedral shape and sharp edges, characteristic of MgO, were found within the pits. The pits and their surroundings also contained a significant amount of gray oxide slag and silvery-gray sulfide inclusions. Notably, the sulfide inclusions were often observed surrounding graphite nodules. Furthermore, the areas adjacent to these defects showed pronounced decarburization and oxide scale formation. A common feature associated with both white and black spot metal casting defects was the presence of graphite flotation zones nearby, exhibiting exploded, dendritic, thick flake, and large spheroidal graphite forms.
In black spot regions, the etched samples revealed a network-like distribution of grayish rare-earth oxides and titanium nitride inclusions along the edges of the shrinkage porosity zones. The surrounding matrix again displayed severe graphite flotation, along with complex magnesium sulfide-bearing inclusions. The coexistence of slag, porosity, and abnormal graphite structures pointed towards a localized disturbance in solidification dynamics caused by the presence of exogenous particles.
The collective analysis led to the conclusion that both types of metal casting defects originated from primary slag formed during the in-mold spheroidization treatment. The process sequence was critical: the base iron underwent efficient pneumatic desulfurization, achieving ultra-low sulfur levels (“double-zero” grade). While this reduced sulfur-related issues, it also diminished the population of natural impurities that typically act as nuclei for slag agglomeration and flotation. The in-mold process itself was rapid, from treatment to casting solidification. The slag generated during the magnesium-ferrosilicon addition was fine, dispersed, and particulate, lacking the time and mechanism to coalesce into larger, buoyant masses that could separate from the melt.
Consequently, a portion of this finely dispersed slag was entrained into the mold cavity by the flowing metal. During solidification, these particles were trapped in the upper sections of the casting (cope side) or at dead zones near core surfaces. The slag primarily comprised oxides and sulfides of magnesium and rare earth elements. The presence of these solid particles acted as barriers to heat transfer and metal feeding, promoting localized shrinkage porosity (micro-shrinkage) and creating sites for graphite degeneration. The thermodynamic and kinetic conditions favoring slag formation can be partially described by reactions such as:
$$ \text{[Mg]}_{\text{(in iron)}} + \frac{1}{2}\text{O}_2 \rightarrow \text{MgO}_{(s)} $$
$$ 2\text{[Mg]} + \text{SiO}_2_{(s)} \rightarrow 2\text{MgO}_{(s)} + [\text{Si}] $$
$$ \text{[Mg]} + \text{[S]} \rightarrow \text{MgS}_{(s)} $$
$$ \text{[RE]} + \frac{3}{2}\text{O}_2 \rightarrow \text{RE}_2\text{O}_3_{(s)} $$
Where [ ] denotes elements dissolved in the iron melt, and (s) denotes solid phase. The free energy change for these reactions, $\Delta G$, determines their feasibility at casting temperatures. For instance, the stability of MgO is very high, making its formation likely where oxygen potential is sufficient.
Upon machining, the brittle slag particles in the subsurface were exposed. If the dominant phase was powdery MgO, it was removed, leaving a white, chalky appearance—the white spot metal casting defect. If the defect zone contained a mixture of tenacious silicate-based slag and associated shrinkage cavities, machining revealed a dark, porous area—the black spot metal casting defect. The adjacent graphite flotation and decarburization were secondary effects driven by the localized cooling inhibition and possible reaction with atmospheric oxygen through mold gases.
| Defect Name | Macro/Micro Appearance | Key Identified Constituents | Associated Features | Proposed Formation Sequence |
|---|---|---|---|---|
| White Spot | White powder, pits, octahedral MgO particles | MgO (primary), some oxides/sulfides | Local porosity, decarburization, graphite flotation nearby | Fine MgO slag entrainment → trapping in upper mold → solidification disturbance → machining exposes powder. |
| Black Spot | Dark area, micro-shrinkage, network inclusions | RE-Mg Silicates, MgS, TiN, oxides | Severe graphite flotation, complex inclusions | Complex slag particle entrainment → acts as chill & blocks feeding → shrinkage & slag cluster → machining reveals cavity & slag. |
The impact of these metal casting defects on mechanical properties was significant, particularly detrimental to impact toughness and fatigue strength. The brittle inclusions and associated porosity acted as potent stress concentrators and crack initiation sites, drastically reducing the component’s durability under dynamic loading. The fatigue life reduction can be conceptually modeled using a stress-intensity factor approach, where the defect size influences the crack propagation rate. For a surface defect acting as a crack, the stress intensity factor range $\Delta K$ is approximated by:
$$ \Delta K = Y \Delta \sigma \sqrt{\pi a} $$
where $Y$ is a geometric factor, $\Delta \sigma$ is the applied stress range, and $a$ is the defect size (e.g., equivalent diameter of the slag cluster or pore). The presence of such defects would significantly reduce the fatigue threshold $\Delta K_{th}$ and accelerate crack growth, leading to premature failure—a critical concern for high-stress components like crankshafts.
To eliminate these pervasive metal casting defects, the strategy focused on modifying the slag’s physical characteristics and enhancing its removal from the melt before entering the casting cavity. The goal was threefold: (1) purify the iron melt by promoting slag separation, (2) transform the slag from a fine powder into a viscous, cohesive liquid phase to facilitate agglomeration, and (3) provide a mechanism to trap the coalesced slag before the metal enters the mold. This was achieved by modifying the in-mold reaction chamber design and adding specific fluxing agents to the spheroidizing alloy or the reaction chamber.
The key modification involved the addition of small amounts of fluorspar (CaF₂), sodium fluosilicate (Na₂SiF₆), and either glass powder or glass fibers into the spheroidizing well or reaction chamber. The rationale and mechanisms are outlined below:
- Fluorspar (CaF₂): Acts as a powerful flux. It lowers the melting point and viscosity of the existing slag system. The reaction can be simplified as CaF₂ helping to dissolve refractory oxides like MgO and Al₂O₃ into a fluid calcium-aluminosilicate slag.
$$ \text{CaF}_2 + \text{MgO} + \text{SiO}_2 \rightarrow \text{CaO·MgO·SiO}_2 \text{ based slag (low melting point)} $$
This promotes the coalescence of fine particles into larger droplets. - Sodium Fluosilicate (Na₂SiF₆): Decomposes at high temperatures, providing active fluorine and sodium which further enhance fluidity and cleaning action.
$$ \text{Na}_2\text{SiF}_6 \rightarrow 2\text{NaF} + \text{SiF}_4 $$
The generated NaF is an excellent flux for silicate systems. - Glass (SiO₂-rich): Provides a readily fusible silica source to form a low-melting-point, viscous slag matrix that effectively wets and encapsulates the refractory MgO and RE oxide particles, enabling their agglomeration.
The combined effect of these additives transformed the primary slag from a solid, particulate dispersion into a liquid, cohesive globule. Furthermore, the design of the in-mold system, often incorporating a centrifugal or swirling flow pattern in a mixing chamber, helped concentrate the now-coalesced slag droplet in the center due to centrifugal force. A strategically placed baffle or dam then retained this slag mass within the reaction/mixing chamber, allowing only the cleansed iron to flow into the gating system and mold cavity. This principle of centrifugal separation relies on the density difference between slag ($\rho_{slag}$) and iron ($\rho_{iron}$), where the radial acceleration ($a_r = \omega^2 r$) forces the less dense slag towards the center or a specific collection zone. The Stokes’ law modified for centrifugal force gives the settling velocity of a slag droplet:
$$ v_c = \frac{d^2 (\rho_{iron} – \rho_{slag}) \omega^2 r}{18 \eta} $$
where $d$ is droplet diameter, $\omega$ is angular velocity, $r$ is radius, and $\eta$ is melt viscosity. Increasing $d$ via coalescence dramatically increases $v_c$, making separation efficient.
After implementing these process modifications—optimizing flux addition and leveraging centrifugal separation in the reaction chamber—the subsequent production batches of crankshaft castings were thoroughly inspected. The results were definitive: the white spot and black spot metal casting defects were completely eliminated. The machined surfaces of the crankshaft journals were clean and free from the previously observed patches and pits. This led to a substantial improvement in the product’s dimensional integrity, aesthetic quality, and, most importantly, its mechanical performance consistency. The yield rate increased significantly, validating the effectiveness of the targeted approach to slag management in the in-mold spheroidization process.
This case study underscores a critical aspect of advanced metal casting processes: the interplay between metallurgical treatment and fluid dynamics. While achieving high-purity base iron through desulfurization is beneficial, it can inadvertently create conditions that stabilize undesirable fine reaction products. The solution often lies not in reversing the purification but in actively engineering the physical state and behavior of the by-products. The successful remediation of these specific metal casting defects highlights the importance of a holistic view that considers chemistry, thermodynamics, kinetics, and flow dynamics. It serves as a reminder that the elimination of metal casting defects frequently requires a multifaceted strategy targeting the entire process chain, from treatment to solidification.
Further generalizing from this experience, the formation of slag-related metal casting defects follows a common pathway: generation → dispersion → entrapment → manifestation. The strategies to combat them can be mapped to each stage, as summarized in the table below:
| Stage in Defect Genesis | Key Phenomena | Potential Countermeasures | Technical Objective |
|---|---|---|---|
| 1. Slag Generation | Oxidation, desulfurization, spheroidization reactions producing solid products. | Control atmosphere (use of covers, inert gas), use of cleaner alloys, optimized treatment temperature. | Minimize total slag volume; favor liquid over solid slag formation. |
| 2. Slag Dispersion & Growth | Fine particles remain dispersed; lack of agglomeration due to high surface energy, low wetting. | Add fluxing agents (CaF₂, Na₂SiF₆, glass) to reduce melting point/viscosity; promote coalescence. | Transform slag to liquid phase; increase particle/droplet size ($d$). |
| 3. Slag Separation & Removal | Entrainment in turbulent flow; insufficient buoyancy or time for flotation. | Design quiescent filling systems; employ centrifugal separation in chambers; use ceramic filters; adequate holding time. | Maximize separation force/时间; physically trap slag before mold entry. |
| 4. Slag Entrapment in Casting | Trapping in solidifying metal at cope, dross plates, or core junctions. | Optimize gating to minimize turbulence; control solidification direction (use of chills, risers); improve mold coat integrity. | Prevent slag from reaching critical casting regions or facilitate its floatation to dedicated areas (e.g., risers). |
The economic and qualitative implications of controlling such metal casting defects are profound. In high-integrity castings like automotive crankshafts, the cost of a scrap part includes not just the material and energy but also the lost machining value-add. Implementing a robust slag management protocol, as described, enhances overall process capability (Cp, Cpk) and reduces variability. This investigation into the white and black spot metal casting defects reinforces that continuous process monitoring, combined with root-cause metallurgical analysis, is indispensable for advancing casting quality and reliability in demanding applications.