Heuristic Design of Advanced Martensitic Steels That Are Highly Resistant to Hydrogen Embrittlement by ε-Carbide

Round 1

Reviewer 1 Report

Page 4 lines 145-146: “η-carbide and ε-carbide that were observed by electron diffraction; η-carbide and ε-carbide that were observed by electron diffraction patterns were nearly the same”. It seems being repeated and incomplete sentence.

Comments for author File: Comments.pdf

Author Response

Response to Reviewer 3 Comments

Point 1: Page 4 lines 145-146: “η-carbide and ε-carbide that were observed by electron diffraction; η-carbide and ε-carbide that were observed by electron diffraction patterns were nearly the same”. It seems being repeated and incomplete sentence.

Response 1: The duplicate sentence has been deleted as follows: η-carbide and ε-carbide that were observed by electron diffraction patterns were nearly the same. I feel sorry for that mistake.

Reviewer 2 Report

The strength of this manuscript is the detailed description of physical metallurgy of high strength martensitic steel, by tying together the roles of composition, microstructure, phases, and precipitation in the properties of martensitic steels.   Clearly this is the area of expertise of the author. 

Principal concern:

The treatment of the topic of hydrogen embrittlement, on the other hand, is the weaker part of the manuscript, which is unfortunate because the premise is based on the effect of e-carbide on HE.

The experimental work and results, as they are presented, are not sufficient to conclude that presence of e-carbide is the reason for the lower HE susceptibility. More precisely, the comparison of a quenched and tempered steel with an as-quenched steel, without providing any information on mechanical properties of the latter is a serious flaw. Given that strength has a first order effect on HE susceptibility, if the as-quenched steel has higher strength (which it should), it is normal for it to exhibit greater HE susceptibility. In fact, a detailed characterization of the mechanical properties of both materials is necessary: Tensile Strength, Yield Strength, ductility, impact toughness, hardness.

A true analysis of the effect of microstructure on HE susceptibility must be done on materials that are at EQUAL strength.

Secondary concerns:

a. A more detailed description of the mechanical U-bending method is needed, along with figures and schematic.

b. The thermal desorption methodology is not adequately described. Notably the size and location of the sample is important to describe. Also, the time lag between hydrogen charging and TDS measurement is not known. This is an important aspect that can affect the results.

c. A DISCUSSION section seems warranted. The last paragraph of the experimental section (lines 189-201) belongs in the DISCUSSION, along with a more in-depth discussion of the results, notably Figure 1.

Minor concerns:

a. I recommend using the vocabulary of HE that is more universally adopted. Notably "HE susceptibility" rather that "HE resistance."

b. Some minor correction of the English is needed.

SEE ATTACHED DOCUMENTS FOR SPECIFIC COMMENTS

Comments for author File: Comments.pdf

Author Response

Point 1: Principal concern:

The treatment of the topic of hydrogen embrittlement, on the other hand, is the weaker part of the manuscript, which is unfortunate because the premise is based on the effect of e-carbide on HE.

The experimental work and results, as they are presented, are not sufficient to conclude that presence of e-carbide is the reason for the lower HE susceptibility. More precisely, the comparison of a quenched and tempered steel with an as-quenched steel, without providing any information on mechanical properties of the latter is a serious flaw. Given that strength has a first order effect on HE susceptibility, if the as-quenched steel has higher strength (which it should), it is normal for it to exhibit greater HE susceptibility. In fact, a detailed characterization of the mechanical properties of both materials is necessary: Tensile Strength, Yield Strength, ductility, impact toughness, hardness.

A true analysis of the effect of microstructure on HE susceptibility must be done on materials that are at EQUAL strength.

Response: This manuscript has been written to respond to the repeated call for contribution to a special issue of “Heat Treatment of Steels” from the editorial board. It is not intended to be a full paper. The objective of the paper is to communicate rapidly a conceptual design of advanced martensitic steels resistant to HE. The author will be glad if the research arouses a lot of interest in the ε-carbide and stimulates the research. It is added that the experiments were carried out under special friendship of my old colleagues.

 Secondary concerns:

1. A more detailed description of the mechanical U-bending method is needed, along with figures and schematic.

Response: The author supposes that the references in the manuscript are enough to try the experiments.

1. The thermal desorption methodology is not adequately described. Notably the size and location of the sample is important to describe. Also, the time lag between hydrogen charging and TDS measurement is not known. This is an important aspect that can affect the results.

Response: The author is not able to demand much for the disclosures.

1. A DISCUSSION section seems warranted. The last paragraph of the experimental section (lines 189-201) belongs in the DISCUSSION, along with a more in-depth discussion of the results, notably Figure 1.

 Response: The main point of the paper is to communicate rapidly a conceptual design of advanced steels. Each section is planned to be self-sufficient. In future, the author would like to publish a paper written in traditional style including "Materials and Methods" and “Discussion” sections where the statements are complete and satisfying..

Minor concerns:

1. I recommend using the vocabulary of HE that is more universally adopted. Notably "HE susceptibility" rather than "HE resistance.

Response: The “HE resistance” in the manuscript has been replaced by “HE susceptibility”. Thanks for the comment.

1. Some minor correction of the English is needed.

Response: Revised manuscript will be subjected to the English editing service provided by MDPI..  

SEE ATTACHED DOCUMENTS FOR SPECIFIC COMMENTS

Response: No comments, because I was unable to find the attached documents in the letter from the editor.

Reviewer 3 Report

Dear author, thanks for sharing the very promising insights of your manuscript on the use of ε carbides for enhancing HE resistance of martensitic steels.

Short summary: The manuscript performs a literature review of the hydrogen solubility of ε-carbide and its precipitaion hardening effect. This survey serves as an introduction to a hydrogen absorption experiment that was performed comparing back-to-back hydrogenation of ε-carbide containing and non ε-carbide samples of a given heat, by means of TDA-GC. The manuscript ends with a steel design suggestion related to elements influencing Ms, ε-carbide stability and nucleation.

General comments: The manuscript is well written, easy to understand and even didactic. The approach that is shown is very interesting and matchess one of the trends in HE resistant steel design: that of increasing the trapping capacity by promoting the presence of stable traps. It would have been enriching thought to include also the percolation path control design strategy too (e.g. https://doi.org/10.1098/rspa.2014.0108) and carbide based HE resistance enhancement strategies from other authors for discussion (e.g. https://doi.org/10.1098/rspa.2006.1688). In terms of the manuscript organization, the literature survey would fit nicely as an extended introdution and there it would be very welcome to have a traditional "Materials and Methods" section including the info required for any third party to be able to reproduce the results.

Specific comments:

Line 15. Stating that "the carbide has a high resistance to hydrogen embrittlement" may be misleading. Would the author accept the recommendation to rephrase as " the proper use of carbides in steels can promote a high resistance to hydrogen embrittlement".

Line 31. Could you please cite a source to support this statement "...the strength of steel bolts for automobiles has been limited to 1400 MPa because of HE"?

Paragraph starting in Line 49. Would it be possible to further explain why H trapping the mechanism is pointed towards interstitial solution inside the ε-carbides and not towards the trapping in their surface? The importance of having a high surface per unit volume of carbides is frequently underlined in literature. If it is not like that for ε-carbides, the reader would really thank this point to be clarified.

Line 147. Fasteners and automotive applications should also be considered, not oly H₂S handling applications. Furthermore, ISO 15156-1:2015 strongly limits the strength of the materials that can be used in sour service and the appication of high strength LTT martensitic steels would not be feasible if the standard is followed. Would it be possible to clarify what sour service applications are refered to in this paragraph so that the reader is not misled?

Section 3. As for other colleagues in the fields may be interested in repeating and expanding this work, some further inforamtion would really be welcome such as:

• Original product format: rolled bar? rolled sheet? what reduction was used? was it cast? were specimens machined or rolled to the desired thickness? Was it achieved by cold rolling? These factors should be relevant to understanding that the observed effects are actually due to ε-carbides or there may be other factors interacting too.
• Chemical composition: what was the nitrogen content? Could Ti have been fixed into TiN? Were there any other nitride/carbide formers (Al, V, Nb, Cr)? What were the S and P contents? This
• Could you please provide the results of elongation and yield strength for the tested material conditions? Hardness valus would also be very helpful in the interpretation. This should help distinguishing the equilibrium between restauration/stress relief/carbide precipitation and its effect in the compared specimens.
• What was the number of repeats of each test?
• Please complete with micrographs if possible.

Section 4.1. Please explain why the Ms is important for this specific case and how it sohould be tuned. Is the ides to keep the %C the highest possible with the smallest risk of reatiend austenite? IS there any preferred Ms target?

Thanks again for sharing your very interesting work.

Author Response

Point 1: General comments: The manuscript is well written, easy to understand and even didactic. The approach that is shown is very interesting and matches one of the trends in HE resistant steel design: that of increasing the trapping capacity by promoting the presence of stable traps. It would have been enriching thought to include also the percolation path control design strategy too (e.g. https://doi.org/10.1098/rspa.2014.0108) and carbide based HE resistance enhancement strategies from other authors for discussion (e.g. https://doi.org/10.1098/rspa.2006.1688). In terms of the manuscript organization, the literature survey would fit nicely as an extended introduction and there it would be very welcome to have a traditional "Materials and Methods" section including the info required for any third party to be able to reproduce the results.

Response: Thanks for warm and encouraging words.

As you have raised, the problem of hydrogen transport through percolation will be an important subject for future study. In that case, one has to discuss the site percolation problem if the precipitates of ε-carbide are isolated and the bond percolation one if they are connected with dislocations for pipe diffusion.

The article (https://doi.org/10.1098/rspa.2006.1688) deals with the trapping of hydrogen by M4C3 precipitates. Although I am not versed in the hydrogen trap state of M4C3, a paper by Kawakami & Matsumiya (ISIJ Int., 52(2012), 1693-1697) is interesting to me in that ab-initio calculation has suggested a possibility of hydrogen trapping by carbon vacancies in V4C3.

This manuscript has been written to respond to the repeated call for contribution to a special issue of “Heat Treatment of Steels” from the editorial board and is not intended to be a full and perfet paper. The objective of the manuscript is to communicate rapidly a heuristic insight into advanced martensitic steels resistant to HE. Furthermore, the experiment was carried out under special friendship of my old colleagues, so I was unable to demand much. I suppose that the references in my manuscript are enough to reproduce the present results.

Point 2: Specific comments:

Line 15. Stating that "the carbide has a high resistance to hydrogen embrittlement" may be misleading. Would the author accept the recommendation to rephrase as " the proper use of carbides in steels can promote a high resistance to hydrogen embrittlement".

Response: Thanks. I adopt your recommendation.

Line 31. Could you please cite a source to support this statement "...the strength of steel bolts for automobiles has been limited to 1400 MPa because of HE"?

Response: The statement was cited from Technical Journal of Daido Steel, Vol. 90, No 2(2019),87-91.(https://www.daido.co.jp/common/pdf/pages/technology/journal/backno/2019/90_2/04_technicaldata01_ando.pdf) The report is written in Japanese. It writes in the introduction that the maximum strength of bolts for automobiles has been 1400 MPa. The carbon content is 0.31-0.35%. The main objective of the report is that 1600 MPa-class bolts are developed for special use. Its C content is 0.58%.

Paragraph starting in Line 49. Would it be possible to further explain why H trapping the mechanism is pointed towards interstitial solution inside the ε-carbides and not towards the trapping in their surface? The importance of having a high surface per unit volume of carbides is frequently underlined in literature. If it is not like that for ε-carbides, the reader would really thank this point to be clarified.

Response: I do not deny the possibility of hydrogen trapping at the interface. However, the crystal structure of ε-carbide has many C-vacancy site. In hydrogen storage compound such as LaNi5, hydrogen atoms can occupy interstitial sites easily. If the crystal of a carbide contains C-vacancy sites, hydrogen atoms will first occupy that sites and next interstitial sites. Then, the hydrogen atoms will be trapped at the interfaces. Nagakura (J. Phys. Soc. Japan, 1959, 186-95) noticed the carbon content of Fe2.4C (=Fe24C10) is exactly halfway between Fe3C (=Fe24C8) and Fe2C (=Fe24C12). In short, ε-carbide has two C vacancies for the formula of Fe24C10. An elastic energy consideration will predict hydrogen atoms to occupy the vacant sites. As cited above, Kawakami & Matsumiya (ISIJ Int., 52(2012), 1693-1697) considered the carbon vacancies as possible trapping sites for hydrogen atoms. It is added that in the case of TiC, they concluded that Fe/TiC interface is the main trap site. By the way, I am not sure if the iron carbide Fe2C will absorb hydrogen atoms up to Fe2CH, although this composition was erroneously claimed by Berg (Am. Ceramic Soc. Bulletin, 1961, 78) As for the possible capacity of hydrogen absorption, experimental and theoretical studies are necessary.

In this study, the desorption curve indicates a negligibly small desorption from the interface. Most of desorption takes place at temperatures when the ε-carbide precipitates decompose into cementite. This leads to an interpretation that hydrogen atoms are trapped inside the carbide.

 Line 147. Fasteners and automotive applications should also be considered, not oly H₂S handling applications. Furthermore, ISO 15156-1:2015 strongly limits the strength of the materials that can be used in sour service and the application of high strength LTT martensitic steels would not be feasible if the standard is followed. Would it be possible to clarify what sour service applications are referred to in this paragraph so that the reader is not misled?

 Response: The present author is sorry to be ignorant of fasteners and automotive applications of LTT. As long as the citation of Krauss’s report is concerned, the readers would not be misled.

Section 3. As for other colleagues in the fields may be interested in repeating and expanding this work, some further information would really be welcome such as:

• Original product format: rolled bar? rolled sheet? what reduction was used? was it cast? were specimens machined or rolled to the desired thickness? Was it achieved by cold rolling?

These factors should be relevant to understanding that the observed effects are actually due to ε-carbides or there may be other factors interacting too.

Response: Hot rolling of sheet (27mm thick reduced to 3mm). The sheet is finally machined to 1 mm thick.

• Chemical composition: what was the nitrogen content? Could Ti have been fixed into TiN? Were there any other nitride/carbide formers (Al, V, Nb, Cr)? What were the S and P contents?

Response: N: 0.0028%, S: less than 0.007, P:0.005%, S: less than 0.007%, Al: 0.038%, V~0, Nb~0, and Cr~0.

• Could you please provide the results of elongation and yield strength for the tested material conditions? Hardness values would also be very helpful in the interpretation. This should help distinguishing the equilibrium between restauration/stress relief/carbide precipitation and its effect in the compared specimens. What was the number of repeats of each test?

Response: Hv=517 for quenched steel, Hv=491 for tempered one. Elongation and yield strength values are unknown.

Response: N=5 for hardness, N=2 for embrittlement test, N=1 for desorption test.

• Please complete with micrographs if possible.

Response: The micrographs have not been provided to the author, although I viewed them.

Section 4.1. Please explain why the Ms is important for this specific case and how it sohould be tuned. Is the ides to keep the %C the highest possible with the smallest risk of reatiend austenite? IS there any preferred Ms target?

 Response: The Ms temperature should be higher than the tempering temperature. The Ms temperature and the cooling rate determine the precipitation of fine carbides and/or nitrides for the inoculation of ε-carbide. The necessary cooling rates are dependent on the Ms.

Round 2

Reviewer 2 Report

I am a bit puzzled about the reply that the manuscript is not intended to be a full paper. If that is the case, the author should be more explicit and clear in stating the objectives, beginning with the title. Furthermore the proposition that a "hydrogen resistant" steel has been developed has not been proven experimentally. The manuscript should refer to the literature where there is information on the role of e carbides on HE susceptibility,  rather than on insufficient experimental data. This approach would focus the objective on the conceptual design model for advanced steel.

I strongly recommend the manuscript be revised to remove the focus on the  the experimental data for susceptibility. Rather, the experimental approach may be worthy of proposing.

The editorial and terminology changes made by the author are noted and appreciated.

Author Response

Thanks for the 2nd reviewing. The objectives of the paper is revised in line 34-36 lines. The word "Heuristic" is added to the title. I have believed that an open-access on-line journal is open not only to studies based on experimental positivism but also to new heuristic way of scientific thinking and finding on materials science and engineering. I also believe that  one of the functions of this kind of journal is rapid communications of scientific ideas and preliminary results to the public to stimulate and encourage the young research people.

I regret to admit that my experimental work is insufficient. However, the results are substantiated with the previous reports [2 & 9]. This is stated in the line 187-189.

Thanks again for stimulating discussion.

Reviewer 3 Report

Dear author, thanks for your detailed answers, I consider all the points on the review satisfied. 

It would have been very nice if you could  have found a way to address the answers in the text.

Regards. 

Author Response

Dear author, thanks for your detailed answers, I consider all the points on the review satisfied. 

Response:  I am much pleased, if I have cleared the points. 

It would have been very nice if you could  have found a way to address the answers in the text.

Response: If I could, It would be nice.  However, I owe much to young engineers at a steel labo for the hydrogen studies. I would like to reserve  discussion on the questions for  their future study and publication. 

Round 3

Reviewer 2 Report

Thank you for the revision. I have attached a revised set of comments intended as improvements. Please review each of them carefully as they appear in the Comments Pane. Especially the ones related to standard terminology for HE (e.g., diffusible, trapping). Most of these appeared on my first revision, but were not addressed. 

My most significant critique is on Line 168. The specimen without tempering. This is a critical point. If the specimen was not tempered, it follows that it has a higher strength than 1500 MPa. Therefore strength of the untempered specimen MUST be listed. The reason being that strength has a first order effect on HE susceptibility. So naturally, this specimen will be more susceptible. Second, by virtue of the difference in strength the two materials  cannot be compared "side by side." You will have to support your conclusion (lines 173-176) about the role of e carbide on something more than the preliminary observation of cracking. At a minimum you must explain your conclusions in spite of the higher strength (and greater susceptibility) of the untempered sample.  My suggestion is to base your hypothesis on your findings of Figure 1, which seemingly show that the presence of carbides results in more trapping.

Please address this fundamental oversight. I think it can be done some some rewriting of Section 3.

Wishing you success.

Comments for author File: Comments.pdf