Scale Formation on HSLA Steel during Continuous Casting Part I: The Effect of Temperature–Time on Oxidation Kinetics

Round 1

Reviewer 1 Report

The manuscript entitled Scale Formation on HSLA Steel during Continuous Casting Part I of author Rosa M. Pineda Huitron et al. is very ambitiously set. The topic is highly relevant and very important for the scientific field.

Unfortunately - according to the submitted text - the manuscript cannot be recommended for publication as it does not meet the minimum quality requirements for a scientific article. In spite of the fact that a lot of effort has been put into the work, several details are poorly described and in some parts even misinterpreted. Moreover, despite the relevant topic and investigation, the conclusions given do not provide new findings.

Author(s) do(es) not show a full competence of describing, evaluating, and discussing at least the following basic concepts and phenomena. For instance:

• Principles of oxidation sequence (e.g. adsorption, dissociation, diffusion of various species) – see Fig. 2b
• Principles of oxidation of non-alloyed vs. alloyed steels are not adequately discussed (e.g. possible influence of alloying elements C, Si, Mn, etc??).
• Principles of oxidation in wet atmosphere. From Fig. 6 a relatively large distance between furnace and steam generator can be seen. However, the dew point(s) of the gas mixture during the selected experimental condition(s) are not provided.
• The methodology of the experimental work is very poorly described. Please, distinguish between expressions Experimental setup vs. Characterization (see paragraph 2).
• Full characterization? Please, be more specific.
• Microstructure characterization is poor (see Fig. 8.) Many details remained overlooked and undescribed.
  Please, distinguish between expressions Microstructure vs. Phase analysis. How about the (decarburized) steel matrix??
• Decarburization under "dry" vs. "wet" conditions?? Comparison of microstructures of nonetched cross-sections are missing...
• Basics of technical article are not respected – i.e. chemical composition of the steel under investigation is not provided.
• etc.

All these factors undermine the scientific soundness of the manuscript. After a thorough revision of the manuscript a resubmission is recommended.

Author Response

Rosa Maria Pineda Huitron

Ph.D Candidate at Luleå University of Technolgy.

Materials Science and Engineering Department.

Professor Reese Kong,

Assistant Editor in Chief, MDPI, Journal Metals

Monday, 07 September 2020

Subject: Authors Reply

Journal: Metals

ISSN: 2075-4701

Manuscript ID: metals-910587

Manuscript: “Scale formation on HSLA steel during continuous casting

Part I: The Effect of Temperature-Time on Oxidation Kinetics”

Dear Prof. Reese Kong,

Thank you for the reviewers’ comments on the above paper under your consideration. We thought the comments were very pertinent and helpful in improving the manuscript.

We have addressed all of the reviewers’ comments as detailed below. Clarification and expansion of the discussion was also suggested in a few areas and this has also been done in the revised paper.

I hope you find all in order and I look forward to hearing from you.

Yours sincerely,

Rosa M. Pineda Huitron

Response to the Reviewer 1

Comments:

The manuscript entitled Scale Formation on HSLA Steel during Continuous Casting Part I of author Rosa M. Pineda Huitron et al. is very ambitiously set. The topic is highly relevant and very important for the scientific field.

Unfortunately - according to the submitted text - the manuscript cannot be recommended for publication as it does not meet the minimum quality requirements for a scientific article. In spite of the fact that a lot of effort has been put into the work, several details are poorly described and in some parts even misinterpreted. Moreover, despite the relevant topic and investigation, the conclusions given do not provide new findings.

Author(s) do(es) not show a full competence of describing, evaluating, and discussing at least the following basic concepts and phenomena. For instance:

• Principles of oxidation sequence (e.g. adsorption, dissociation, diffusion of various species) – see Fig. 2b

Authors reply: Authors consider this comment is important to emphasize. Furthermore, Figure 2 has been modified according to reviewer’s comments with a proper explanation of the main stages during oxidation in pure iron (lines 65-72 in manuscript).

• Principles of oxidation of non-alloyed vs. alloyed steels are not adequately discussed (e.g. possible influence of alloying elements C, Si, Mn, etc??).

Authors reply: A paragraph regarding the effect of some alloying elements containing the steel was added to the text (lines 83-93 in manuscript).

• Principles of oxidation in wet atmosphere. From Fig. 6 a relatively large distance between furnace and steam generator can be seen. However, the dew point(s) of the gas mixture during the selected experimental condition(s) are not provided.

Authors reply: Authors consider that observations from the reviewer are extremely interesting and definitely have to be consider in future work. Unfortunately, we did not have the required equipment for measuring the relative humidity during oxidation (e.g. dew point meter). Thus, the oxygen concentration (partial pressure) could not be determined for further calculations. The proposed measurements are planned to be performed in future work (industrial project next year).

• The methodology of the experimental work is very poorly described. Please, distinguish between expressions Experimental setup vs. Characterization (see paragraph 2).

Authors reply: Authors consider that the description of experiments is complete since it includes subsections that describes separately the identification of samples, oxidation tests under air and water vapour as well as the description of the characterization of specimens via different techniques used in laboratory. However, small adjustments has been done in the text.

Experiment sep-up is a term that refers directly to the experimental procedure (how experiments were conducted) during oxidation of specimens (under dry and water vapour atmospheres), including the equipment, parameters and samples used during experiments. All described in a macro-scale.  

Characterization, is an important aspect in materials science that we use to learn about materials performance during/after a certain test, which includes a variety of techniques available in laboratory to enable the characterization (e.g. Microscopy, Diffraction techniques, etc.).

In this work, the characterization section aims to describe the methodology for revealing the oxide scale components, properties and morphology by using different laboratory techniques. For instance, in order to obtain data to analyze the phase morphology, phases formed in oxide scales, defects, etc., we applied metallographic methods including grinding, polishing and different etchants. In addition, we used different microscopic techniques for the analysis of the already revealed structure of oxides.

• Full characterization? Please, be more specific.

Authors reply: Authors removes this term for avoiding misunderstandings. However, a full characterization is referred to a number of techniques needed/used for the internal analysis of the oxide scales obtained after experiments (described above).

• Microstructure characterization is poor (see Fig. 8.) Many details remained overlooked and undescribed.
  
  

Authors reply: Authors agreed that the microstructure analysis is not well described in this section. Furthermore, we decided to change the name of this section as “oxide scale growth” since this only described the thickness changes as a function of time and temperature with presence of defects. The microstructure analysis is now included together with the phase analysis, which aims to identify the phases present in oxides (obtained with help of etched samples) including their quantitative analysis by XDR technique.

• Please, distinguish between expressions Microstructure vs. Phase analysis. How about the (decarburized) steel matrix??

Authors reply: Microstructure is the structure obtained after etching the oxide scale (transverse direction) in order to identify microscopically the phases in oxides according to their morphology and colour. Phase analysis, refers to measurements obtained from phases in oxides, which in this case were determined via XRD (quantitative analysis).

• Decarburization under "dry" vs. "wet" conditions?? Comparison of microstructures of nonetched cross-sections are missing...

Authors reply: Although, decarburization/carburization of the steel is a phenomenon that occurs during oxidation, it was not considered to be included within the scope of this work. We considered that decarburization was important to mention in the text to back up the side effect of carbon during oxidation of the steel. Furthermore, this topic is extensive and it requires more specific work and analysis to explain the effect of decarburization on oxide scale formation. Unfortunately, we don’t count with enough data for that purpose, being then not ideal to include in this particular work since it aims to analyze the oxide scale behaviour (kinetics) formed at different temperatures and times not based on decarburization phenomenon.

• Basics of technical article are not respected – i.e. chemical composition of the steel under investigation is not provided.

Authors reply: Since this project is in collaboration with SSAB Europe, they restricted the possibility to mention the alloying elements content in this work. However, some of the main alloying elements were included in the text (line 165 of manuscript).

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have done comprehensive oxidation studies by characterizing the oxide composition using XRD, SEM and mechanical testing. Here are some comments the authors may address.

1. For XRD characterization, what is the detection depth? When the oxide scale is thick, some oxide products may not be measured. For Rietveld refinement, authors may provide the refinement results as supplemental information or list the reliability factor in the composition table.
2. In table 9, similar to those in water vapor, the standard deviation of oxide scale thickness should also be added for those in dry air.
3. In Figure 17, how to distinguish the phases between FeO, Fe3O4 and Fe2O3 in terms of morphology? Some refs are needed here. 
4. The language could be further polished. e.g. in abstract, "phenomena is"; in the 1st sentence of Introduction, "is focuses on"

Author Response

Rosa Maria Pineda Huitron

Ph.D Candidate at Luleå University of Technolgy.

Materials Science and Engineering Department.

Professor Reese Kong,

Assistant Editor in Chief, MDPI, Journal Metals

Monday, 07 September 2020

Subject: Authors Reply

Journal: Metals

ISSN: 2075-4701

Manuscript ID: metals-910587

Manuscript: “Scale formation on HSLA steel during continuous casting

Part I: The Effect of Temperature-Time on Oxidation Kinetics”

Dear Prof. Reese Kong,

Thank you for the reviewers’ comments on the above paper under your consideration. We thought the comments were very pertinent and helpful in improving the manuscript.

We have addressed all of the reviewers’ comments as detailed below. Clarification and expansion of the discussion was also suggested in a few areas and this has also been done in the revised paper.

I hope you find all in order and I look forward to hearing from you.

Yours sincerely,

Rosa M. Pineda Huitron

Response to the Reviewer 2

Comment 1:

The authors have done comprehensive oxidation studies by characterizing the oxide composition using XRD, SEM and mechanical testing. Here are some comments the authors may address.

1. For XRD characterization, what is the detection depth? When the oxide scale is thick, some oxide products may not be measured. For Rietveld refinement, authors may provide the refinement results as supplemental information or list the reliability factor in the composition table.

Authors reply: Authors agree on the reviewer’s observation and this is one of the reasons we decided to made a comment in this regard in the manuscript. Indeed, the penetration depth during experiments will depend on the incident and diffracted angles, the direct beam and diffracted X-ray intensity but also the detection limit of the detector itself. Generally, the maximum penetration depth is around 2-20 µm in reflection mode, which will depend on the samples condition (density, surface condition) that will have an effect on the analysis.  In fact, for thicker oxide scales, the measurements are limited and possibly not accurate data can be obtained during XRD analysis.

Since the specimens under dry oxidation were directly analyzed on top of the surface by XRD (bulk samples), this can overestimate some of the information obtained after the analysis. Additionally, the rough surface, also observed on specimens, was consider to have an effect during measurements leading to a large dispersion of the x-rays diffracted from the oxide scale surface resulting in low intensity and less accuracy of volume fraction values. This last, was emphasized in the present work.

The manuscript focuses more in analyzing the phases obtained under water vapour conditions, since the volume fraction of phases is valuable to obtain because in combination with the micro-mechanics can explain the oxide scale behaviour and the steel performance during continuous casting process. However, the specimens obtained under dry air (10 minutes and 40 minutes holding time) were used as a comparison of possible phases found under similar temperatures and times used for water vapour oxidation. Please find an example of these measurements (bulk samples of oxides formed under dry air conditions) by using Rietveld refinement in report attached to in a separate file.

Finally, we would like to remark that powder samples were used during the XRD measurements for oxide scales formed under water vapour conditions to obtained more accurate values.

1. In table 9, similar to those in water vapor, the standard deviation of oxide scale thickness should also be added for those in dry air.

Authors reply: Changes has been done according to reviewer suggestion. Figure 9, line 257 in manuscript.

1. In Figure 17, how to distinguish the phases between FeO, Fe3O4 and Fe2O3 in terms of morphology? Some refs are needed here. 

Authors reply: Iron oxides such as Wüstite (FeO) magnetite (Fe3O4) and hematite (Fe2O3) were studied through the basics of oxidation in pure iron and steels (see references below). Based on the theory of iron oxides, which in terms of chemical reactions and their formation in equilibrium (Fe-O diagram) leads to a specific order that they appear during oxidation. Wüstite the first at the steel substrate, magnetite the second and hematite the third and thinner in contact with the environment.

In terms of morphology, the methodology used in this work (not used before) by combining different etchants to reveal different phases, was optimal to observe the delimitation between them. In fact, the identification of oxides is quite complex because they can be identified differently by using different characterization techniques such as SEM, LOM (light optical microscopy), where they may look different (see figures below).

Comparison of a) SEM vs. B) LOM techniques used of identification of phase sin oxides. Specimen oxidized at 1000  per 30 minutes holding time under water vapour conditions.

As it was explained in the experimental work, the use of different etchants helped us to identify phases such as Wüstite, magnetite or hematite. Such etchants reacted on the oxide and shading the surface in different colors. This type of results (color etchants) can only be analyzed by using LOM. This method was our main tool to identify more accurately the morphology of different phases in oxides, but we applied our own method for this purpose. Indeed, previous work were used to compare the data, which have been already referred within the manuscript.

1. The language could be further polished. e.g. in abstract, "phenomena is"; in the 1st sentence of Introduction, "is focuses on"

Authors reply: Changes have been done according to the reviewer suggestion.

Author Response File: Author Response.docx

Reviewer 3 Report

General Comments

The manuscript contains interesting test results concerning scale formation on HSLA steel during continuous casting. However, before the publication it requires the following changes:

1. The paper has an unusual structure. I suggest that Section 3 "Results" should be combined with Sections 4 and 5, as is the case with most papers. Section 3 would be then entitled “Results and discussion”.
2. In Section 2.2 it says: “Oxidation under dry air is conducted by heating the specimens at 1000 and 1100 ℃ and subsequently removing a specimen every 5 minutes with the last one removed after 40 minutes”. However, diagram 7a shows something quite different. Please explain.
3. In Figure 8 there are too many pictures in a small space. This makes the pictures unreadable. The figures should be divided into individual temperature variants (separately for 1000, 1100 and 1200°C, or there should be less of them).
4. The results of thickness measurements of the obtained oxide scales are puzzling. The results presented in Fig. 9 and Table 1 seem disputable. Such high thicknesses of oxide scale at such short oxidation times would be difficult to obtain even on materials that are highly susceptible to oxidation, e.g. titanium. Please explain.
5. Part of the information from Section 3.3 (first paragraph) should be moved to Experimental setup.
6. In Fig. 13, the unit of the parabolic rate constant of oxidation, K_p, is of concern. Please explain.
7. Arrhenius graph should be also added to the manuscript.

Author Response

Rosa Maria Pineda Huitron

Ph.D Candidate at Luleå University of Technolgy.

Materials Science and Engineering Department.

Professor Reese Kong,

Assistant Editor in Chief, MDPI, Journal Metals

Monday, 07 September 2020

Subject: Authors Reply

Journal: Metals

ISSN: 2075-4701

Manuscript ID: metals-910614

Manuscript: “Scale formation on HSLA steel during continuous casting

Part I: The Effect of Temperature-Time on Oxidation Kinetics”

Dear Prof. Reese Kong,

Thank you for the reviewers’ comments on the above paper under your consideration. We thought the comments were very pertinent and helpful in improving the manuscript.

We have addressed all of the reviewers’ comments as detailed below. Clarification and expansion of the discussion was also suggested in a few areas and this has also been done in the revised paper.

I hope you find all in order and I look forward to hearing from you.

Yours sincerely,

Rosa M. Pineda Huitron

Response to the Reviewer 3

Comments

The manuscript contains interesting test results concerning scale formation on HSLA steel during continuous casting. However, before the publication it requires the following changes:

1. The paper has an unusual structure. I suggest that Section 3 "Results" should be combined with Sections 4 and 5, as is the case with most papers. Section 3 would be then entitled “Results and discussion”.

Authors reply: Authors consider reviwer´s comments valuable. Section 4 (modelling) is now integrated in Results. However, we prefer to keep in separate sections the results and discussions since both are extensive. By having both separately, the following up of the manuscript becomes more clear with a better structure. Now Discussion is modified as section 4, line 409 in manuscript.

2. In Section 2.2 it says: “Oxidation under dry air is conducted by heating the specimens at 1000 and 1100 ℃ and subsequently removing a specimen every 5 minutes with the last one removed after 40 minutes”. However, diagram 7a shows something quite different. Please explain.

Authors reply: Authors have adjust the thermal cycle according to reviewer’s comments (Figure 7-a). The time described during oxidation of specimens is the one that is taken only during holding time. This means that when the specimen has reached the target temperature (i.e. 1000 and 1100  ), the holding time during oxidation begins. Furthermore, a zoom in of the plots has been done in order to observed the time for each sample. The temperature changes (in Figure 7-a) indicate the moment at which the specimens were taken out from the chamber after reaching the desired time (holding time).

1. In Figure 8 there are too many pictures in a small space. This makes the pictures unreadable. The figures should be divided into individual temperature variants (separately for 1000, 1100 and 1200°C, or there should be less of them).

Authors reply: Changes has been done accordingly in Figure 8.

1. The results of thickness measurements of the obtained oxide scales are puzzling. The results presented in Fig. 9 and Table 1 seem disputable. Such high thicknesses of oxide scale at such short oxidation times would be difficult to obtain even on materials that are highly susceptible to oxidation, e.g. titanium. Please explain.

Authors reply: As it was mentioned in the manuscript (lines 483-485) this type of experiments were performed partially under non controlled atmosphere in pilot scale. This type of experiments are quite complex to perform since they are not 100% atmosphere control during oxidation. Meaning that the interaction of both oxygen from the air and from the water vapour (steam) are mixed during experiments. This type of tests were designed to have a better understanding and situation as close as possible to the real process (continuous casting). During continuous casting process, the strand is subjected to continuous cooling with water coming from the nozzles. This type of process makes the situation more interesting because the steel is also in contact with the oxygen in the air, which may lead to thicker oxide scales. In fact, during our experiments, a number of limitations make the tests only being an approximation. For instance, the contact with the rolls in the caster machine, the pressure of the water and considering that the strand is under cooling and not under heating as it was the case in our experiments.

We must consider as well, that most of the oxidation experiments for other materials that try to mimic other processes such as hot rolling, stamping, etc., they are usually performed under isothermal conditions by using thermogravimetry (TGA) techniques, which keeps the material under protected atmosphere and more control of the oxidation can be done. Such experiments also takes the weight loss for determination of kinetics (oxidation rate, diffusion, etc) while in our case we consider the thickness of the oxide scale for that matter. Moreover, we consider our experiments have a better approach of the real situation when comparing with those only performed under isothermal conditions. We, however, aim to understand what the behaviour of the steel is when “simulating” (including) the parameters encountered during the real process at a smaller scale (pilot scale).

1. Part of the information from Section 3.3 (first paragraph) should be moved to Experimental setup.

Authors reply: Authors agree on the reviewer’s comment. Changes has been done accordingly and those can be seen in section 2.4, line 219-225 in manuscript.

1. In Fig. 13, the unit of the parabolic rate constant of oxidation, Kp, is of concern. Please explain.

Authors reply: The purpose of showing results of the oxidation rate constant (k_p) in [cm²/s] was only to compare the results obtained from this work to previous studies. Most of the references cited in this manuscript use the same units (cm²/s). See references below numbered in the same order as in manuscript:

[1] H. Abuluwefa, R. Guthrie, F. Ajersch, The effect of oxygen concentration on the oxidation of low-carbon steel in the temperature range 1000 to 1250° C, Oxidation Metals 1996, 46, 5-6.

[7] R. Chen, W. Yeun, Review of the high-temperature oxidation of iron and carbon steels in air or oxygen, Oxidation Metals 2003, 59, 5-6.

[10] S. Liu, D. Tang, H. Wu, L. Wang, Oxide scales characterization of micro-alloyed steel at high temperature, J. Mater. Process. Technol. 2013, 213, 7.

[19] J. Slowik, G. Borchardt, C. Köhler, R. Jeschar, R. Scholz, Influence of oxide scales on heat transfer in secondary cooling zones in the continuous casting process, part 2: determination of material properties of oxide scales on steel under spray‐water cooling conditions, steel research international 1990, 61, 7.

[43] H. Yin, W. Yuen, D. Young, Effects of water vapour and oxygen partial pressures on low carbon steel oxidation in N2–H2–H2O mixtures, Materials and Corrosion 2012, 63, 10.

[44] C. Issartel, H. Buscail, Y. Wang, R. Rolland, M. Vilasi, L. Aranda, Water vapour effect on ferritic 4509 steel oxidation between 800 and 1000 C, Oxidation Metals 2011, 76, 3-4.

[45] H. Buscail, R. Rolland, S. Perrier, Influence of water vapour on 316L oxidation at high temperature-in situ X-Ray diffraction 2015.

1. Arrhenius graph should be also added to the manuscript.

Authors reply: Changes has been done according to reviewer’s comments. The determination of the activation energy from Eq. 9 (Arrhenius form) has been moved now to section 3.4.1. (Results, linesn377-) that includes the plots obtained for calculations (figure 14, line 387). Additionally, it was observed an error during calculations of the activation energy in dry air conditions that has been fixed in text, table and reflected in the plot (Figure 14).

Author Response File: Author Response.pdf