Twin Roll Casting and Secondary Cooling of 6.0 wt.% Silicon Steel

Round 1

Reviewer 1 Report

The presented article deals with the production of electric steel with 6.5 wt. % Si by twin-roll casting and the subsequent possibility to suppress the formation of an ordered structure by cooling. The results show that the method of rapid cooling used through the critical temperature range of formation was not sufficient because the subsequent cooling during coiling was very slow from a relatively high temperature. This raises the question of whether it would be possible to design the technology so that the steel strip can be cooled faster or cooled to lower temperatures before coiling. Although the authors have not been successful in suppressing the formation of an ordered structure, the results are very valuable in the field of electric steel research with the possibility of producing steels with a high Si content with respect to their excellent soft magnetic properties and workability.

 

The article is written clearly and appropriate methods of examining the produced material have been used.

Formal errors were found:
- p. 1: space groups (Pm3m; Fm3m) are not written correctly
- p. 6 second line: dot is written in front of the reference [19]

Author Response

Point 1: p. 1: space groups (Pm3m; Fm3m) are not written correctly

Response 1: Thanks for the comment by the reviewer. We corrected the space group notation to "Pm-3m", "Fm-3m".

Point 2:  p. 6 second line: dot is written in front of the reference [19]

Response 2: Thanks for the comment by the reviewer. We corrected the formal error.

Reviewer 2 Report

The article is written well and is appropriate for publication.

I just want an explanation:

- It is stated in section 3.2: "No dendritic structures were distinguishably measuring with a working distance of 22.5 nm".

What does a working distance of 22.5 nm mean?

 

 

Author Response

Point 1: - It is stated in section 3.2: "No dendritic structures were distinguishably measuring with a working distance of 22.5 nm".

What does a working distance of 22.5 nm mean?

Response 1: Thanks for the question by the reviewer. The working distance in the SEM is the distance at which the beam is focussed. However, the distance between the objective lens and the specimen was not meant in this context, but the distance between two measuring points in the map scan, which is supposed to give a conclusion about the resolution. The term "working distance" is incorrect in this context and must be replaced by "step size". Therefore, we have replaced the term "working distance" by "step size".

Reviewer 3 Report

Two different secondary cooling strategies after twin roll casting in an effort to suppress the cause of embrittlement were carefully studied. Detailed microstructure characterization was conducted. While the attempt to suppress the formation of these super-lattice structures was unsuccessful, the manuscript provides valuable insights and results that can be of interest to the readers of this journal. The manuscript is also well-organized and well-written. There are few minor points, which are stated below, that can be addressed to improve the quality of the paper:

• In section 3.3 the authors stated that “water-cooled sample is composed of more equiaxed grains both in edge and middle of the sheet.” This is not as intuitively noticeable compared to the authors’ following statement “…much more fine grains are received in the edge area, where the highest cooling rate…” It is recommended to use OIM or other EBSD processing software to provide a quantitative support to the argument “…water-cooled sample is composed of more equiaxed grains both in edge and middle of the sheet.”
• In section 3.3 the authors stated “Both trials show a nearly random texture with a slightly preferential formation of <001>//ND oriented grains visible in the color red.” The images seem to be more {001} oriented than random. Please provide inverse pole figure, pole figure, or orientation distribution function map to support the statement.
• An OM image of AC1378 can be added in Fig. 3 alongside AC1379 for direct comparison in terms of surface condition. Furthermore, surface distortion can be better represented using EBSD images by displaying grain misorientation or low angle boundaries.
• The authors in section 3.4 stated “The advantage of the faster cooling rate of water-cooling could not be exploited in the given experiment.” However, I suggest that the phrase changed to “The advantage of the faster cooling rate of water-cooling could not be fully exploited…” because water-cooling clearly affected the microstructure of the sample.
• I suggest that the authors can briefly discuss their expectations on how further cooling would affect the formation of super-lattice structures. Given that faster cooling is an important factor, would it reasonable to expect suppression of the super-lattice structures if the authors were to extract TEM specimen near the surface of the sample? A short discussion about this is appreciated.

Author Response

Point 1: In section 3.3 the authors stated that “water-cooled sample is composed of more equiaxed grains both in edge and middle of the sheet.” This is not as intuitively noticeable compared to the authors’ following statement “…much more fine grains are received in the edge area, where the highest cooling rate…” It is recommended to use OIM or other EBSD processing software to provide a quantitative support to the argument “…water-cooled sample is composed of more equiaxed grains both in edge and middle of the sheet.”

Response 1: We thank the reviewer for the comment. We changed the statement to "In contrast the water-cooled sample AC1379 is composed of more equiaxed grains. ", which is a more softened statement. Sample AC1379 has more equiaxed grains, which is also a consequence of the significantly smaller grain size. However, it is true that the grains in the edge region of AC1379 are not all equiaxed. We think it is more understandable now because of the weakened statement.

Point 2: In section 3.3 the authors stated “Both trials show a nearly random texture with a slightly preferential formation of <001>//ND oriented grains visible in the color red.” The images seem to be more {001} oriented than random. Please provide inverse pole figure, pole figure, or orientation distribution function map to support the statement.

Response 2: Thanks for the comment of the reviewer. We calculated ODFs from EBSD data and added inverse pole figure in the manuscript. It is now clear that the texture is predominantly 001//ND in both AC1378 and AC1379.

Point 3: An OM image of AC1378 can be added in Fig. 3 alongside AC1379 for direct comparison in terms of surface condition. Furthermore, surface distortion can be better represented using EBSD images by displaying grain misorientation or low angle boundaries.

Response 3: Thanks for the suggestion by the reviewer. We added KAM-maps both from AC1378 and AC1379. The difference in surface distortion is now clearly visible.

Point 4: The authors in section 3.4 stated “The advantage of the faster cooling rate of water-cooling could not be exploited in the given experiment.” However, I suggest that the phrase changed to “The advantage of the faster cooling rate of water-cooling could not be fully exploited…” because water-cooling clearly affected the microstructure of the sample.

Response 4: Thanks for the comment by the reviewer.  We have implemented the proposed modification.

Point 5: I suggest that the authors can briefly discuss their expectations on how further cooling would affect the formation of super-lattice structures. Given that faster cooling is an important factor, would it reasonable to expect suppression of the super-lattice structures if the authors were to extract TEM specimen near the surface of the sample? A short discussion about this is appreciated.

Response 5: We thank the reviewer for the comment. It can be assumed that the superlattice structure is most likely to be suppressed in the near-edge area. However, the focus of the present work was on ensuring sufficient plasticity for cold rolling. Assuming that the superlattice structure is responsible for the brittleness of the sheet, it is necessary to ensure suppression of this embrittling phase over the entire volume. Thus, if the B2/DO3 phase could not be detected in the edge region, it could still be present in the center region and lead to cracking during roll forming. Therefore, we took the FIB samples as far as possible from the interior of the sheet. However, we also could not measure too far in the center area, since this is where the segregation is present. The 500 µm distance from the specimen surface is therefore a good compromise between the segregation and the interior of the specimen.