Since the second half of 2019, in the cold rolling process for automotive steel plates, varying degrees of slag inclusion defects have been detected, which in severe cases lead to batch blockages and downgrading. Moreover, during the stamping of various parts by automotive sheet users, part cracking due to slag inclusion defects has occurred, sometimes resulting in irreparable scrap. The slag inclusion defects in automotive steel cast slabs severely impact production efficiency and quality costs, and continuously reducing these defects has been a long-term goal in steel production. In this article, I will delve into the causes, analysis, and control measures for slag inclusion defects, leveraging data, tables, and formulas to provide a comprehensive understanding.
The slag inclusion defect typically manifests as elongated streaks parallel to the rolling direction on the surface of thin plates (0.5–1.0 mm after rolling), and in severe cases, it forms large cracks. To analyze this, samples from defect locations were examined using scanning electron microscopy (SEM) and composition analysis. The subsurface inclusions were found to contain elements such as carbon, oxygen, fluorine, sodium, magnesium, aluminum, silicon, sulfur, calcium, titanium, and iron. This confirms that the defect originates from slag inclusions during the steelmaking process. The presence of these elements can be summarized using a composition table, which highlights the complex nature of the slag inclusion defect.

The formation of slag inclusion defects is primarily attributed to the entrapment of mold flux into the slab surface during continuous casting. When the mold flux has a high melting point and poor fluidity, it tends to cause slag inclusion issues. The sources of slag inclusions include ladle slag entering the tundish due to delayed nozzle closure, erosion of tundish refractories and nozzles, and aluminates from the steel melt. These sources contribute to the slag inclusion defect through two main mechanisms: entrainment into the steel interior and adhesion to the slab surface.
During casting, if the submerged entry nozzle (SEN) is centered, the flow field should be symmetric. However, turbulence in the steel flow can create vortices near the nozzle, especially with mold level fluctuations. This turbulence increases the probability of slag entrainment. The relationship between vortex formation and stability can be expressed using fluid dynamics formulas. For instance, the velocity at the meniscus region can be modeled as: $$ v_m = \frac{Q}{A} \cdot f(\text{Re}) $$ where \( v_m \) is the meniscus velocity, \( Q \) is the flow rate, \( A \) is the cross-sectional area, and \( f(\text{Re}) \) is a function of the Reynolds number accounting for turbulence. As casting speed increases, the upward flow intensifies, exacerbating level fluctuations and slag entrainment, thereby worsening the slag inclusion defect.
At higher casting speeds, slag entrainment often occurs near the narrow face of the mold. The steel flow rises along the narrow face, changes direction upon impacting the meniscus, and flows back toward the nozzle. This thins the slag layer, leading to steel exposure and slag accumulation. The interfacial tension between steel and slag plays a role, with some slag particles being dragged into the steel. The force balance can be described as: $$ F_{\text{drag}} = \rho \cdot v^2 \cdot C_d \cdot A_s $$ where \( F_{\text{drag}} \) is the drag force, \( \rho \) is density, \( v \) is velocity, \( C_d \) is the drag coefficient, and \( A_s \) is the surface area. This contributes to the slag inclusion defect through adhesion or entrainment.
Several factors influence the severity of the slag inclusion defect. I will discuss each in detail, using tables and formulas to summarize key points.
1. Mold Flux Properties: The physical properties of mold flux, such as melting point, viscosity, and basicity, critically affect slag inclusion formation. An ideal flux should have a molten layer thickness of 12–15 mm and a consumption of 0.40–0.60 kg/t of steel. For automotive steels, the flux typically has higher viscosity and lower basicity compared to other grades. This requires a balance between inflow and melting rates. The viscosity (\( \eta \)) influences flux consumption, while melting speed determines how quickly it melts. The relationship can be approximated by: $$ \eta = A \cdot e^{\frac{B}{T}} $$ where \( A \) and \( B \) are constants, and \( T \) is temperature. Table 1 compares the physical properties of automotive steel flux with other fluxes, highlighting factors that exacerbate the slag inclusion defect.
| Property | Automotive Steel Flux | Other Fluxes |
|---|---|---|
| Melting Point (°C) | 1040–1120 | 910–1130 |
| Viscosity at 1300°C (Pa·s) | 0.17–0.45 | 0.09–0.13 |
| Basicity (CaO/SiO₂) | 0.87–1.18 | 1.1–1.2 |
Low consumption or excessive consumption can lead to uneven flux inflow, causing poor heat transfer and increased slag inclusion defect risk.
2. Argon Blowing Rate: Argon is injected into the SEN to prevent nozzle clogging and reduce reoxidation. However, excessive argon flow can cause severe mold level fluctuations and slag entrainment. The argon volume flow rate \( Q_{\text{Ar}} \) should be optimized to minimize turbulence. From production data, reducing the argon equivalent from an average of 2.2 to 1.6 (a 27.8% decrease) significantly lowered slag inclusion defects. The effect can be modeled using the Weber number: $$ We = \frac{\rho \cdot v^2 \cdot L}{\sigma} $$ where \( We \) is the Weber number, \( \rho \) is density, \( v \) is velocity, \( L \) is characteristic length, and \( \sigma \) is surface tension. Higher \( We \) indicates greater interface instability, promoting slag inclusion defect formation.
3. Tundish Steel Temperature: Superheat temperature affects steel fluidity and inclusion floatation. For automotive steels, superheat should be maintained at 15–40°C. Lower superheat (<15°C) increases viscosity, hindering inclusion removal and leading to clogging and higher argon use. The viscosity-temperature relationship is: $$ \mu = \mu_0 \cdot e^{\frac{E}{RT}} $$ where \( \mu \) is dynamic viscosity, \( \mu_0 \) is a constant, \( E \) is activation energy, \( R \) is the gas constant, and \( T \) is temperature. This directly impacts the slag inclusion defect by affecting flow dynamics.
4. Mold Level Fluctuation: Level stability is crucial; fluctuations can entrain unmelted flux or inclusions like Al₂O₃ into the slab. The standard deviation of level fluctuation \( \sigma_L \) correlates with defect rate. By optimizing control parameters, the qualification index for level fluctuation improved from 5.05 to 9.33 (a 40% increase), reducing slag inclusion defects. The fluctuation amplitude \( \Delta L \) can be expressed as: $$ \Delta L = k \cdot \frac{v_{\text{cast}}^2}{g} $$ where \( k \) is a constant, \( v_{\text{cast}} \) is casting speed, and \( g \) is gravity. Larger \( \Delta L \) increases slag entrainment risk.
5. Casting Speed: Higher speeds increase flow velocity at the SEN outlet, intensifying meniscus turbulence and slag entrainment. The downward flow velocity also increases, reducing inclusion floatation. The relationship between casting speed \( v_c \) and entrainment probability \( P_e \) can be approximated as: $$ P_e \propto v_c^{2.5} $$ This shows that speed variations exacerbate the slag inclusion defect. Implementing constant speed casting helps mitigate this.
6. Nozzle Alignment and Immersion Depth: SEN misalignment causes biased flow, leading to level翻滚 and slag inclusion. Immersion depth affects slag entrainment and inclusion floatation. Shallow depth increases surface slag entrainment, while deep depth hinders inclusion removal. The optimal depth \( h_{\text{opt}} \) can be derived from fluid flow models: $$ h_{\text{opt}} = 0.5 \cdot W \cdot \left( \frac{\rho_s}{\rho_f} \right)^{0.5} $$ where \( W \) is mold width, \( \rho_s \) is steel density, and \( \rho_f \) is flux density. Proper alignment and depth control are essential to minimize the slag inclusion defect.
To address these factors, I implemented several improvement measures. First, for slag prevention, selecting appropriate flux with higher viscosity and lower wettability can reduce entrainment. The viscosity increase helps dampen meniscus fluctuations, as shown by the formula: $$ \eta_{\text{eff}} = \eta_{\text{flux}} + \Delta \eta_{\text{additive}} $$ where \( \eta_{\text{eff}} \) is effective viscosity, and \( \Delta \eta_{\text{additive}} \) is from additives. Additionally, enhancing ladle slag detection system accuracy prevents slag carryover, and using larger tundishes improves inclusion floatation.
Second, for argon control, I collaborated with refractory suppliers to modify SEN structure and strengthen argon pipeline maintenance. The argon flow rate was precisely regulated based on mold level stability. Table 2 shows the reduction in argon equivalent before and after improvements, highlighting its impact on the slag inclusion defect.
| Period | Average Argon Equivalent | Change |
|---|---|---|
| Before (Jul–Dec 2019) | 2.2 | — |
| After (Jan–Jun 2020) | 1.6 | -27.8% |
Third, for level fluctuation control, I optimized measurement devices and operational parameters, adopting fixed slag line practices. The qualification index improved steadily, as summarized in Table 3. This stabilization directly reduced slag entrainment and the occurrence of slag inclusion defects.
| Month | Qualification Index | Cumulative Improvement |
|---|---|---|
| January | 5.05 | — |
| February | 6.5 | 28.7% |
| March | 5.87 | 16.2% |
| April | 7.8 | 54.5% |
| May | 8.25 | 63.4% |
| June | 9.33 | 84.8% |
Fourth, for casting speed, I promoted constant speed casting by setting strict standards for section, speed, and throughput. The constant speed ratio increased from 82.83% in January 2020 to 93.18% in June 2020, as shown in Table 4. This consistency enhanced inclusion floatation and reduced slag entrainment, thereby controlling the slag inclusion defect.
| Month | Constant Speed Ratio (%) | Increase |
|---|---|---|
| January | 82.83 | — |
| February | 85.30 | 2.98% |
| March | 86.10 | 3.95% |
| April | 89.30 | 7.81% |
| May | 94.51 | 14.10% |
| June | 93.18 | 12.50% |
Other measures included maintaining optimal superheat, ensuring nozzle alignment, and preventing low-tonnage issues during grade changes. These collectively improved steel cleanliness and process stability, reducing the risk of slag inclusion defects.
The implementation of these measures led to significant outcomes. The blockage rate due to slag inclusion defects in rolling decreased from 6.3% in January 2020 to 3.6% in June 2020. The amount of slabs returned for rework dropped substantially, saving mechanical and labor costs for defect cleaning and reducing logistics-related production expenses. Moreover, customer complaints regarding slag inclusion defects fell by 50%, and direct economic losses from customer claims decreased by 70%. The economic benefit can be quantified using a cost-saving formula: $$ S = (R_b – R_a) \cdot C_b + D_c \cdot C_c $$ where \( S \) is total savings, \( R_b \) and \( R_a \) are defect rates before and after, \( C_b \) is cost per defect blockage, \( D_c \) is reduction in complaint cases, and \( C_c \) is cost per complaint. This demonstrates the tangible impact of controlling the slag inclusion defect.
In conclusion, I found that stabilizing the mold level during continuous casting is key to reducing steel slag entrainment and the slag inclusion defect. Reducing argon blowing and increasing constant casting speed contribute to level stability, thereby mitigating slag inclusion defects. Enhancing steel purity minimizes inclusions, alleviates nozzle clogging, and stabilizes the casting process, further lowering slag entrainment risk. The improvement in product quality effectively reduces production costs, yielding significant economic benefits. The slag inclusion defect, once a major challenge, can be systematically controlled through integrated process optimizations.
To generalize, the mechanisms behind slag inclusion defects involve complex interactions between fluid dynamics, thermal properties, and material characteristics. Future work could focus on real-time monitoring using sensors and advanced modeling, such as computational fluid dynamics (CFD), to predict and prevent slag inclusion defects. For instance, a predictive model could use parameters like argon flow \( Q_{\text{Ar}} \), casting speed \( v_c \), and superheat \( \Delta T \) to estimate defect probability: $$ P_{\text{defect}} = \alpha \cdot Q_{\text{Ar}}^{2} + \beta \cdot v_c^{2.5} + \gamma \cdot \frac{1}{\Delta T} $$ where \( \alpha, \beta, \gamma \) are coefficients derived from production data. This proactive approach could further minimize the occurrence of slag inclusion defects in automotive steel production.
In summary, addressing the slag inclusion defect requires a holistic view of the continuous casting process. By understanding the sources, formation processes, and influencing factors, and implementing targeted improvements, I have shown that significant reductions in defect rates and associated costs are achievable. The slag inclusion defect remains a critical quality issue, but with continuous optimization and innovation, it can be effectively managed to enhance product performance and customer satisfaction in the automotive steel industry.
