Round 1

Reviewer 1 Report

The abstract and the introduction nicely depict the problem highlighting both the theoretical and the applied points of view. The main challenge is well focused and the methodology is detailed. 

In section 2.4 the balance equation for the particles is computed with several forces acting on them. It should be interesting to show which of them can be neglected and which leads the particle motion. The geometry influence shows a relevant effect in the deposition phenomenon. In the conclusion, those aspects should be deeply recalled instead of a list of statements in order to make the evidence of the new insight coming from the paper. The overall work is very nice and original. It shows why the results can improve the knowledge of the scientific community. The paper is easy to read and the extremely detailed.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript presents some CFD calculation results to analyze the clogging tendency in the SEN during continuous casting. A Eulerian-Lagrangian method is used to calculated the flow and the transport of non-metallic inclusions (NMI) in the SEN; and a sub-laminar boundary layer theory is used to consider the NMI-adhesion frequency on the SEN wall. However, this manuscript needs major revision before it can be considered for publication.

 

Major points:

1.    Authors stated at the beginning in the Abstract that clogging can be decreased through 3 measures: flow modification, raw materials, and smoothed internal surfaces. I think this statement is probably correct. However, the main content of this manuscript can only deal with one aspect, i.e. the flow. No further discussions were made about the role of “raw materials” and the “surface roughness” in the clogging. Only suggesting “to use high purity raw materials and smooth internal surfaces for nozzle manufacturing” in the introduction part is not sufficient.  One may not easily address the ‘raw materials’ issue, as it may concerns the chemical reaction (out of the scope of the current study), but the ‘surface roughness’ is a part of CFD issue. As soon as the first layer of deposition of NMIs on the SEN surface, the surface cannot be smooth any more. Obviously, the authors have missed some most important overview of the state-of-the-art works. Example of literatures can be read: Lee J.H., Kang Y.B., et al., ISIJ Int., 2020, pp426; ISIJ Int., 2018, pp.1257; ISIJ Int., 2019, pp. 749; Barati H., Wu M., et al., Powder Techn., 2018, pp. 181; Metall. Mater. Trans. B, 2019, pp. 1428. Some other important aspects during nozzle clogging are also ignored. For example, the transient growth of clog structure and its dynamic interaction with the turbulent flow, and role of (possible) solidification in the porous structure of clog are also ignored. 

I mention above point is not to say that the flow is not important in the clogging. In the opposite, the reviewer agrees that the flow and transport of NMI are dominant factor in the clogging process. However, the authors should put their study in a right context. At least, one should avoid self-conflicting. You raise the importance of “raw materials” and “smoothed surfaces” at the beginning, but you don’t mention how can you consider them into your work. If you did not consider them in the recent version of the model, you should at least discuss them, and guide the readers to understand your modeling results and conclusions.

2.  The paper structure needs revision. Chapter §3 is for ‘Result’. However, §3.4.1 read like ‘Modeling’, §3.4.2 read like ‘Boundary condition’, and §3.4.3. ‘Validation’. Actually, they are mixed with some simulation result. This kind of structure is not accepted for a scientific paper.

3.  3.1 Model Validation. This part is presented very qualitative. Firstly, Figure 2 did not give any quantitative information, no time, no scale, no dimension, etc. Secondly, do you think that this simulation-experiment comparison of the mold flow can help to evaluate the flow inside the SEN? It is the flow inside the SEN that plays role for the clogging. The mold flow can only reflect what has happened inside the SEN, for example, as a consequence of blockage of SEN section by clogging.

 

Other points in the manuscript.

• 1. Introduction. State-of-the-art literature study on modeling SEN clogging is missing. Maybe, a little review of the CFD model history should be made as well. The reason is that the model used by the current authors for clogging is not new. Authors have used a whole paragraph in the introduction part to describe the origin of NMIs (can be shorten to 1~2 sentences, and not-directly-related references can be removed).
• 2. M&M. The assumption list must be extended by mentioning ignorance of other possible mechanisms of clogging, e.g. chemical reactions, transient growth of clog, solidification, etc.
• Please move some part of §3.4 to §2. See my previous comments.
• In Equation (2), does the last term “gravity g” influence your flow result. The flotation of NMIs is considered in equation (7). As assumption, the melt density is constant, and the motion of NMIs has no impact on the flow.
• Equation 11 repeats the equation 8.
• A vague expression “The particle trajectories' predictions are possible using the instantaneous value of the fluctuating flow velocity based on a Gaussian distributed random number determined by the stochastic method Random Walk Model (RWM).” Please make it clearer.
• 2.5, a time step 0.01 s is used for the calculation. What is the mesh size, what is typical flow velocity in SEN, and what is the Courant number here? I think the time step 0.01 s is too large to fulfill the numerical stability condition.
• Figure 4. What is difference between “Vertical view” and “Longitudnal view”. How to distinguish two nozzles? I can guess what the authors mean, but the figure or figure captions were not sufficiently explained.
• Figure 5. How are numbers of touching inclusions calculated? And how is the ‘percentage’ calculated?
• Did you study the non-symmetry of the NMI touching frequency inside the nozzle? Why no result is presented in this paper regarding to the non-symmetry of the NMI touching frequency? We see the flow in Figure 6 non-symmetry, but we cannot see the distribution of NMI touching points in two side of a SEN inner chamber.
• I do not understand Figure 10. It should be clearly described.
• Reference list. Obviously, many no relevant (less relevant) literature were cited, while missing some most important contributions.

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Brief summary:

The authors present an interesting study internal geometry of the SEN nozzle has been modified to prevent clogging. Although the problem of clogging has been widely research for many years some problems still remain, especially in the implementation of solutions. This paper provides a possible implementation to reduce the clogging behavior.

Broad comments:

The scope of the manuscript is interesting and provides a realizable method for implementation.

The exact design of the two nozzles, A and B, are not clearly described by the drawings in figure 1 alone. Some complementary drawings in other perspectives are needed to show how the chambers and deflectors are mechanically connected to the nozzle itself.

The choice of using the standard k-epsilon turbulence model is not motivated. A more obvious choice would be the k-omega SST which blends the near wall treatment of the k-omega model and the freestream of the standard k-epsilon model.

Furthermore, no attempt is made at a model verification, i.e. mesh independence study. The authors are referred to the ASME standard [1]

Moreover, two important references have been neglected in the literature review, specifically the work by Yang et al. [2] and Timmel et al. [3]. These studies highlight the conditions in the SEN that have been lacking in many numerical investigations over the years, namely the gas pocket present in the nozzle.

Specific comments:

L159: The qualitative validation attempt can be significantly improved by adding a tracer also in the numerical model and comparing the corresponding times that were photographed in the physical model.

 

[1] ASME V&V 20-2009, Standard for Verification and Validation in Computational Fluid Dynamics and Heat Transfer, The American Society of Mechanical Engineers (ASME), New York

[2] Yang, Hyunjin, Surya P. Vanka, and Brian G. Thomas. "Modeling Argon Gas Behavior in Continuous Casting of Steel." JOM 70, no. 10 (2018): 2148-2156.

[3] Timmel, Klaus, Natalia Shevchenko, Michael Röder, Marc Anderhuber, Pascal Gardin, Sven Eckert, and Gunter Gerbeth. "Visualization of liquid metal two-phase flows in a physical model of the continuous casting process of steel." Metallurgical and Materials Transactions B 46, no. 2 (2015): 700-710.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

No additional comment.

 

Reviewer 3 Report

The authors have responded adequately to the reviewers comments.