Answers are grounded in this paper and cite their sources.
Continuous casting slag plays dominant role in the protecting fresh melt from the re-oxidation and heat losses at the meniscus region, as well as predefines heat transfer and lubrication phenomena at the gap between the solidifying shell and water-cooled copper mold. We developed a multiphase model for the coupled liquid steel–mold slag system that resolves transient flow in both the melt and the liquid slag and couples it with conjugate heat transfer across the interface. Steel melt solidification is included via latent heat release, and the solid slag skin formation and remelting are represented by evolving liquid, mushy, and solid sublayers. A key novelty is the consistent treatment of the inlet-throughput balance, which allows back-pressure pumping of the liquid core, for example due to bulging. Slag is fed into or rejected from the shell–mold gap based on the local pressure–shear balance, so mold powder consumption is not prescribed but emerges from mass conservation and the governing balances without ad hoc forcing. The 2D configuration was applied to isolate the core mechanisms linking mold flow, initial solidification, and slag skin infiltration, while the underlying geometric VOF formulation with sharp interfacial interpolation is written in vector–tensor form. The underlying formulation is not intrinsically restricted to 2D and can in principle be extended to 3D industrial cases. The model was applied to a representative case at a casting speed of 1.2 m/min and verified against state-of-the-art literature and plant observations, showing agreement with (i) slab surface temperatures, (ii) mold heat flux and hot-face temperature distributions, (iii) the analytical Bikerman meniscus profile, (iv) casting speed-scaled mold flux consumption trends, and (v) surface defects observed on final products. Simulations in the stable operating regime reveal surface ripples governed by a small set of dominant coherent oscillation modes. The predicted ripple depth is typically about 20 μ m, with occasional extremes of 40 to 50 μ m, well below typical mold oscillation marks. Even in this quasi-steady regime, the meniscus exhibits a weak time-dependent “breathing,” where small level and gap fluctuations promote intermittent Bikerman-type overflow of the shell tip and leave wavy surface imperfections. Bulging was prescribed using a roller pitch of 200 mm, giving a forcing frequency of 0.1 Hz (mode M1) at the casting speed of 1.2 m/min and higher harmonics at 0.2, 0.3 and 0.4 Hz (modes M2 to M4), generated by periodic shell deformation between support rolls and transmitted upward to the meniscus. Second-harmonic generation was observed even without explicit superposition of the meniscus oscillation modes, associated with a resonance-like response in defect-prone scenarios. Deep surface defects formation and underlying mechanisms, including cold-slag pumping and subsequent hooks freezing, were studied for bulging-induced mold-level oscillations. These effects were found to become particularly critical for the higher-frequency modes despite their net amplitude. Superposition of the main bulging modes produced sharper meniscus velocity and acceleration transients, resulting in increased surface defect depth. The relative phase shift between the superimposed modes was further identified as an additional governing parameter that modifies the instantaneous meniscus dynamics and, consequently, the resulting defect formation.
No takes yet. Share an insight, caveat, or question.
Vakhrushev et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: