Chemical Composition, Microstructure, and Mechanical Properties of Steel Rails Produced by China Hanyang Iron Works

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Peer Review Report

Manuscript title Chemical Composition, Microstructure, and Mechanical Properties of Steel Rails Produced by China Hanyang Iron Works

General assessment This manuscript presents a historically significant and technically detailed analysis of steel rails produced by the Hanyang Iron Works between 1899 and 1921. The authors combine chemical composition measurements, microstructural characterization, inclusion analysis, and mechanical testing to reconstruct the technological evolution of one of China’s earliest integrated steel enterprises. The topic is valuable, the dataset is unique, and the work contributes both to metallurgical history and to the understanding of early steelmaking practices.

The manuscript is generally well structured and clearly written. The experimental methods are appropriate, and the results are systematically presented. However, several areas require clarification, expansion, or correction before the manuscript can be considered for publication.

Major Comments

1. Novelty and contribution require clearer articulation

The manuscript states that “research regarding the evolution of manufacturing technology, mechanical properties, microstructure, and inclusions of Hanyang Iron Works rails is relatively scarce” and that prior studies focused mainly on industrial history. This is a strong motivation, but the authors should explicitly state what new insights their work provides beyond historical narrative.

For example, the manuscript includes the statement:

“The chemical composition of steel rails produced by Hanyang Iron Works in its initial stage… was characterized by low carbon content (0.13–0.22%) and high phosphorus levels (p ≥ 0.15%).”

This is important, but the authors should explain why these findings matter for understanding early Chinese steelmaking and how they compare with contemporary international rail standards of the late 19th century.

2. Insufficient discussion of process–property relationships

The manuscript presents extensive data, but the interpretation is often descriptive rather than analytical. For example, the transition from Bessemer to basic Siemens–Martin steelmaking is mentioned, but the metallurgical consequences are not fully explored.

The authors should expand discussion on:

• How dephosphorization was achieved in the Siemens–Martin process.
• Why carbon content increased from ~0.2% to ~0.5%.
• How inclusion morphology reflects changes in slag composition and refining practice.
• Why tensile strength increased by ~200 MPa after 1909.

These relationships are central to the manuscript’s scientific value.

3. Microstructure analysis needs deeper interpretation

The manuscript states:

“The microstructure of the rails produced in 1899, 1902, and 1904 consists of a large amount of ferrite and a small amount of pearlite.”

This is correct, but the authors should quantify pearlite fractions or provide image analysis. The discussion should also address:

• The role of carbon content in determining ferrite–pearlite balance.
• The significance of “unclosed pores (voids)” and whether they originate from casting defects or hot‑working limitations.
• The metallurgical meaning of “proeutectoid network ferrite” in later rails.

4. Inclusion analysis requires contextualization

Table 2 shows clear changes in inclusion types before and after 1909. The manuscript should explain:

• Why alumina inclusions appear only after the Siemens–Martin transition.
• Whether MnS·Al₂O₃ composite inclusions indicate improved deoxidation practice.
• How inclusion ratings compare with modern rail steel standards.

5. Mechanical property discussion should be strengthened

The manuscript reports tensile strengths ranging from 554 MPa (1904) to 827 MPa (1913). This is a major finding, but the discussion is brief.

The authors should:

• Explain how carbon content, pearlite fraction, and inclusion morphology contribute to strength.
• Discuss the observed fracture morphology transition from ductile dimples to cleavage.
• Comment on the implications for rail service performance and failure modes.

6. Figures require improvement

Several figures contain low‑resolution images, inconsistent labeling, or unclear captions. For example, Figure 4 and Figure 5 combine OM and SEM images but do not clearly identify magnifications or microstructural features.

The authors should:

• Standardize magnification labels.
• Add arrows or annotations identifying ferrite, pearlite, inclusions, and voids.
• Improve clarity of historical rail photographs in Figure 2.

Minor Comments

• The introduction is overly long and contains historical narrative that could be condensed.
• Several sentences are duplicated (e.g., “The establishment of the Hanyang Iron Works laid the initial material foundation…” appears twice).
• Units should be standardized (e.g., “100 um” should be “100 μm”).
• Typographical errors appear throughout the manuscript (e.g., “ThThe primary chemical constituents…”).
• Table 1 should include standard deviations or measurement uncertainty.
• The authors should clarify whether rails were tested in the as‑found condition or after surface cleaning.
• Get rid of trailing zeroes because 24.0 is 24 and 0.0 is 0
• Insert spaces between numerical values and units, including micron bars.
• Provide AI use statement.

Recommendation

Major Revision

The manuscript presents valuable historical and metallurgical data, but the scientific interpretation requires significant strengthening. With improved discussion, clearer figures, and expanded analysis of process–property relationships, the manuscript has strong potential for publication.

 

Author Response

Dear Editors and Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript titled “Chemical Composition, Microstructure, and Mechanical Properties of Steel Rails Produced by China Hanyang Iron Works”(ID: metals-4401939). These comments are valuable and helpful for revising and improving our paper and have important guiding significance for our research. We have carefully studied the comments and have made corrections. The corrections in the revised manuscript are marked in red and the responds to your comments are as follows point by point (in blue).

Best regards.

Sincerely,

Tengshi Liu

 

 

Reviewer 3

General assessment This manuscript presents a historically significant and technically detailed analysis of steel rails produced by the Hanyang Iron Works between 1899 and 1921. The authors combine chemical composition measurements, microstructural characterization, inclusion analysis, and mechanical testing to reconstruct the technological evolution of one of China’s earliest integrated steel enterprises. The topic is valuable, the dataset is unique, and the work contributes both to metallurgical history and to the understanding of early steelmaking practices.

The manuscript is generally well structured and clearly written. The experimental methods are appropriate, and the results are systematically presented. However, several areas require clarification, expansion, or correction before the manuscript can be considered for publication.

Response：We sincerely thank the reviewer for the positive evaluation of our study and for providing these highly constructive and insightful suggestions. These comments have guided our revisions in a clear direction, substantially enhancing the analytical depth, "processing–property" correlations, and overall data presentation of this manuscript. We have systematically addressed all the comments on a point-by-point basis, with specific details provided below.

 

Major Comments

1. Novelty and contribution require clearer articulation

The manuscript states that “research regarding the evolution of manufacturing technology, mechanical properties, microstructure, and inclusions of Hanyang Iron Works rails is relatively scarce” and that prior studies focused mainly on industrial history. This is a strong motivation, but the authors should explicitly state what new insights their work provides beyond historical narrative.

For example, the manuscript includes the statement:

“The chemical composition of steel rails produced by Hanyang Iron Works in its initial stage… was characterized by low carbon content (0.13–0.22%) and high phosphorus levels (p ≥ 0.15%).”

This is important, but the authors should explain why these findings matter for understanding early Chinese steelmaking and how they compare with contemporary international rail standards of the late 19th century.

Response：We appreciate your insightful comment. We fully agree that integrating historical data with metallurgical science and international context can significantly enhance the novelty of the paper. In the revised manuscript, we have supplemented the Introduction and Section 4 to clarify that the characteristics of early low-carbon () and high-phosphorus () compositions were, in fact, a metallurgical compromise forced by the dual constraints of specific regional raw materials (high-phosphorus Daye iron ore) and the thermodynamic limitations of the acid converter (incapability of dephosphorization). Compared with late 19th-century international standards (such as British and American standards requiring  and ), the early Hanyang Iron Works rails were severely constrained by the risk of cold shortness, which also explains the fracture accidents recorded in historical literature. This background clearly elucidates our materials science contribution beyond a purely historical narrative.

In Section 4, the content of the international comparative analysis has been expanded as follows:

"This unique chemical composition combination ( and ) represents a major metallurgical limitation in early Chinese steelmaking. Compared with contemporary late 19th-century international rail specifications—such as the British Standards and the American Society of Civil Engineers (ASCE) rail guidelines, which strictly prescribed medium-to-high carbon contents () and restricted phosphorus contents ()—the early Hanyang converter steel rails exhibited tremendous anomalies. Because the acid converter is thermodynamically incapable of dephosphorization and the local Daye iron ore was inherently high in phosphorus, early steelmakers were compelled to suppress the carbon content to a low-carbon structural steel level (). This was an intentional metallurgical compromise aimed at preventing catastrophic cold shortness embrittlement and maintaining nominal ductility, albeit at the immense sacrifice of wear resistance and yield strength."

 

2. Insufficient discussion of process–property relationships

The manuscript presents extensive data, but the interpretation is often descriptive rather than analytical. For example, the transition from Bessemer to basic Siemens–Martin steelmaking is mentioned, but the metallurgical consequences are not fully explored.The authors should expand discussion on:

How dephosphorization was achieved in the Siemens–Martin process.

Why carbon content increased from ~0.2% to ~0.5%.

How inclusion morphology reflects changes in slag composition and refining practice.

Why tensile strength increased by ~200 MPa after 1909.

These relationships are central to the manuscript’s scientific value.

Response：We sincerely appreciate this critical suggestion. We have expanded Section 4 with an in-depth analysis to systematically elucidate these core "processing–property" correlations:

(a) Dephosphorization Mechanism: It is clarified that the transition to the basic Siemens-Martin open-hearth furnace constructed a highly basic oxidizing slag through lime () additions, thereby achieving highly efficient dephosphorization via the formation of stable  (or ).

(b) Elevation of Carbon Content: The controlled reduction of phosphorus eliminated the risk of cold shortness, which safely enabled metallurgists to increase the carbon content above . This optimization of the pearlite volume fraction substantially enhanced the wear resistance and load-bearing capacity of the rails.

(c) Inclusion Morphology and Slag Systems: The elimination of Type C silicate inclusions (-based) and the emergence of micro-fine  complex inclusions directly reflect the evolution of the slag system from a high-silica acid converter slag to an aluminum-deoxidized basic open-hearth refining process.

(d) Strength Leap: The dramatic leap in tensile strength (reaching the  level) is quantitatively attributed to the synergistic effects of solid solution strengthening from carbon and the massive increase in hard pearlite colony integration.

These analytical texts detailing the underlying process mechanisms have been embedded into Section 4, and the full text is presented below in conjunction with the revisions made for Comments 3, 4, and 5.

 

3. Microstructure analysis needs deeper interpretation

The manuscript states: “The microstructure of the rails produced in 1899, 1902, and 1904 consists of a large amount of ferrite and a small amount of pearlite.”

This is correct, but the authors should quantify pearlite fractions or provide image analysis. The discussion should also address:

The role of carbon content in determining ferrite–pearlite balance.

The significance of “unclosed pores (voids)” and whether they originate from casting defects or hot‑working limitations.

The metallurgical meaning of “proeutectoid network ferrite” in later rails.

Response：Thank you for your comment. We have employed image analysis software (ImageJ) to perform quantitative statistical analysis on the average pearlite volume fraction of each sample group: the rails prior to 1904 exhibit a fraction of , which significantly increases to  in the post-1909 rails.

(a) Ferrite–Pearlite Balance: Under near-equilibrium cooling conditions post-rolling, the pearlite proportion exhibits a linear growth trend with increasing carbon content.

(b) Unclosed Voids: These cavities originated from subcutaneous blowholes and shrinkage cavities formed by highly dissolved gases () during early ingot casting. Due to the low rolling reduction ratio of the first-stage light rolling mill, they could not be completely welded shut under compression.

(c) Network Ferrite: The presence of proeutectoid network ferrite indicates that the steel belongs to hypoeutectoid compositions () and underwent relatively slow cooling after rolling. This allowed ferrite to preferentially nucleate at the prior austenite grain boundaries and grow continuously along them.

In Section 3.2 (Microstructure), the text has been expanded as follows:

"Quantitative analysis indicates that the average volume fraction of pearlite in the pre-transition rails (1899–1904) ranges between  and , scaling linearly with the relatively low carbon content. Following the open-hearth transition (1909–1921), the pearlite proportion expanded drastically to . The 'unclosed voids' detected in the early rails represent gas blowholes and solidification shrinkage cavities. Owing to the primitive rolling capacity and low rolling reduction ratios of the first-stage mill equipment, these internal casting defects could not be dynamically welded shut during hot working. In the subsequent medium-carbon hypoeutectoid steel rails, the emergence of 'proeutectoid network ferrite' stems from slow post-rolling cooling. Ferrite preferentially nucleated at the high-energy prior austenite grain boundaries and extended continuously along them, thereby enveloping the remaining austenite, which subsequently transformed into pearlite colonies."

 

4. Inclusion analysis requires contextualization

Table 2 shows clear changes in inclusion types before and after 1909. The manuscript should explain:

Why alumina inclusions appear only after the Siemens–Martin transition.

Whether MnS·Al₂O₃ composite inclusions indicate improved deoxidation practice. MnS·Al₂O₃

How inclusion ratings compare with modern rail steel standards.

Response：Thank you for your suggestion. We have expanded Section 4 with contextual discussions regarding the non-metallic inclusion analysis:

Alumina and Deoxidation: The specific emergence of alumina () and complex  inclusions after 1909 is attributed to the ladle refining operation under the advanced Siemens-Martin basic open-hearth process, which introduced metallic aluminum for killing and deoxidation. This operation transitioned the molten steel from the uncontrolled liquid silicate inclusions () characteristic of the early converter process toward stable, discrete oxide deoxidation products.

Modern Standard Comparison: Compared with modern standards—such as the Chinese railway standard GB/T 2585-2021, which strictly limits both the heavy and thin series of all inclusion categories to —the heavy series ratings of Type A (sulfides) and Type B (alumina) in the historical Hanyang rails frequently and severely exceeded the limits, reaching as high as 3.5. This indicates that, compared with modern clean steels, the historical rails exhibit a significantly higher susceptibility to inclusion-induced fatigue crack initiation.

In Section 4 (Discussion on Inclusion Analysis), the text has been expanded as follows:

"The specific post-1909 emergence of Type B (alumina) inclusions and complex  phases marks a fundamental transformation in the deoxidation practice. In the basic open-hearth process, metallic aluminum was introduced into the ladle as a potent deoxidizer, converting dissolved oxygen into highly stable, solid  clusters rather than the fluid silicate slags common to the converter process. However, when benchmarked against modern standards like GB/T 2585-2021, where the thin and heavy series of all inclusion types are strictly restricted to within 1.5, the historical Hanyang rails clearly exhibit low cleanliness. Their heavy-series Type A and Type B inclusions reached a severe rating of 3.5 (Table 2). Compared with modern, highly refined clean steels, these elevated inclusion ratings constitute the primary sources of microstructural stress concentration, severely compromising the structural integrity of the material."

 

5. Mechanical property discussion should be strengthened

The manuscript reports tensile strengths ranging from 554 MPa (1904) to 827 MPa (1913). This is a major finding, but the discussion is brief.

The authors should:

Explain how carbon content, pearlite fraction, and inclusion morphology contribute to strength.

Discuss the observed fracture morphology transition from ductile dimples to cleavage.

Comment on the implications for rail service performance and failure modes.

Response：We appreciate your profound suggestion. We have thoroughly restructured the discussion on mechanical properties in Section 4. The dramatic leap in strength (from ~554 MPa to ~827 MPa) is primarily driven by the recarburization process, where the soft ferritic matrix was replaced by ultra-hard lamellar pearlite colonies. This microstructural reorganization also elucidates the fracture transition observed in Figure 9: the early low-carbon ferritic steel failed via ductile microvoid coalescence (characterized by large and shallow dimples), whereas the high-pearlite open-hearth steel severely restricted the development of the crack-tip plastic zone. This restriction forced the localized stress state to exceed the critical cleavage strength, leading to a transition toward transgranular cleavage fracture (characterized by river patterns). While this combination of high strength and low plasticity significantly enhanced wear resistance during service, it simultaneously increased the latent risk of sudden brittle fracture under dynamic overloads.

In Section 4, the text expansion regarding the fracture mechanism is as follows: "The massive leap in tensile strength (~200 MPa) was directly driven by the substitution of the soft ferritic phase with hard pearlitic colonies. This microstructural reconfiguration directly dictated the macroscopic fracture transition observed in Figure 9. In the early low-carbon steel rails, the highly ductile ferritic matrix was capable of accommodating substantial plastic strain, leading to standard ductile microvoid coalescence fracture characterized by large and shallow dimples. Conversely, the high pearlite volume fraction in the post-1909 rails restricted the expansion of the crack-tip plastic zone. The cementite lamellae posed a formidable obstacle to dislocation motion, rapidly elevating the localized triaxial stress state until it exceeded the critical cleavage fracture stress of the material. This forced a transition toward transgranular cleavage brittle fracture, manifested as extensive river patterns and secondary cracks (Figure 9). In actual service, although this upgrade provided excellent resistance to counteract rail head wear and crushing deformation under train wheel loads, it rendered the track highly susceptible to sudden, catastrophic brittle failure if dynamic impact overloading occurred."

 

6. Figures require improvement

Several figures contain low‑resolution images, inconsistent labeling, or unclear captions. For example, Figure 4 and Figure 5 combine OM and SEM images but do not clearly identify magnifications or microstructural features.

The authors should:

Standardize magnification labels.

Add arrows or annotations identifying ferrite, pearlite, inclusions, and voids.

Improve clarity of historical rail photographs in Figure 2.

Response：Thank you for your excellent comment regarding the visual presentation of the figures. We have thoroughly revised all the illustrations as follows:

Figures 4 and 5: Standardized, high-contrast micron scale bars have been directly embedded within all sub-figures ( for OM and low-magnification SEM;  for high-magnification SEM).

Feature Annotations: In Figures 4, 5, 7, 8, and 9, high-contrast colored arrows and text labels have been superimposed to explicitly denote the exact locations of "ferrite", "pearlite", and "inclusions".

Figure 2: The original photographs of the historical rail inscriptions have been re-digitalized, contrast-enhanced, and sharpened, with textual overlays added to ensure maximum clarity of the stamped characters on the rail surface.

The high-resolution image files for Figures 2, 4, 5, 7, 8, and 9 have been completely revised and updated in the final manuscript layout according to the guidelines mentioned above.

 

Minor Comments

The introduction is overly long and contains historical narrative that could be condensed.

Response：Completed. We have removed the redundant descriptions in Section 1 (Introduction) that were not directly relevant to metallurgy, compressing the historical narrative section by approximately 25% to make it more concise. Specifically, the original text from lines 35–85 in Section 1 has been streamlined and condensed, rendering the literature review substantially more compact.

 

Several sentences are duplicated (e.g., “The establishment of the Hanyang Iron Works laid the initial material foundation…” appears twice).

Response：Corrected. We have thoroughly reviewed the entire manuscript and removed the completely redundant sentence in Section 1 (specifically, the second occurrence of the duplication in the third paragraph) to ensure the coherence of the narrative. The repetitive sentence originally located at line 44—"The establishment of the Hanyang Iron Works laid the initial material foundation..."—has been completely deleted.

 

Units should be standardized (e.g., “100 um” should be “100 μm”).

Response：Completed. All occurrences of "um" in the main text, table headers, and all figure panels have been meticulously corrected to the standard micron symbol "". The string "um" throughout the entire manuscript text and figure captions has been uniformly converted to "".

 

Typographical errors appear throughout the manuscript (e.g., “ThThe primary chemical constituents…”).

Response：Corrected. The entire manuscript has been rigorously proofread, and typographical errors such as "ThThe" in Section 1.2 have been completely eliminated. Specifically, the misprint "ThThe" originally located at line 107 has been corrected to "The".

 

Table 1 should include standard deviations or measurement uncertainty.

Response：Adopted. Since each rail sample was measured three times via spark optical emission spectrometry (OES), we have recalculated the measurement variance and updated Table 1 into the standard "mean  standard deviation" format.

 

The authors should clarify whether rails were tested in the as‑found condition or after surface cleaning.

Response：Added. We have incorporated an explicit explanatory statement in Section 1.2 to clarify that all rail samples underwent thorough chemical rust removal, ultrasonic degreasing, and deep surface cleaning before specimen cutting and optical emission spectrometry (OES) excitation.

The following sentence has been added in Section 1.2: "Prior to sampling and spectroscopic excitation, the recovered rails were subjected to surface oxide removal, chemical pickling, and ultrasonic degreasing to guarantee that the core evaluations were exempt from contamination."

 

Get rid of trailing zeroes because 24.0 is 24 and 0.0 is 0.

Response：Thoroughly revised. Adhering to this formatting guideline, all invalid trailing zeros in Tables 2 and 3, as well as throughout the main text, have been completely eliminated (e.g., modifying "24.0" to "24" and "0.0" to "0"). The numerical values in Tables 2 and 3 have been rigorously reformatted to remove unnecessary decimal places.

 

Insert spaces between numerical values and units, including micron bars.

Response：Completed. We have performed a comprehensive check and replacement throughout the main text and figures of the manuscript, strictly ensuring that a standard space is maintained between all numerical values and their corresponding metallurgical units (e.g., "420 MPa", "40%"). All missing spaces between values and units across the entire manuscript have been thoroughly corrected via a programmatic global search and replace.

 

Provide AI use statement.

Response：Added. In accordance with the latest editorial and publishing ethics policies, a standard "Declaration of Generative AI and AI-Assisted Technologies in the Writing Process" has been formally inserted prior to the Acknowledgments section.

 

Declaration of Generative AI and AI-Assisted Technologies in the Writing Process: During the preparation of this work, the authors used AI assistance to language polish, proofread, and format the manuscript and response letters to enhance scannability and structural clarity. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

When accepting the article for review, I expected some scientific contribution in addition to the historicising overview. The article lacks novelty, although its processing and topic are interesting, and it can be read with interest. Nevertheless, there are weaknesses in the article, or rather points that should have been addressed.

1. At the end of the introductory part, before the description of the work carried out and the results, the goal, meaning or intention of the work should be clearly defined.
2. The authors work with material that has been in service. This means that the structure was influenced by the length of service of the rails in question, including different loads, speeds of passage, maintenance, etc. These operating parameters affected the studied steel structures differently. Information on the estimated load is missing. Although some trend in the development of parameters can be observed in the case of strength properties, especially due to the significantly higher carbon content, the structure will certainly have been influenced by the operating parameters. Some cumulative information could possibly capture anomalies.
3. The reader only learns of the possible influence of operational load in the description of the results (page 10, line 257). A brief note should be devoted to this aspect in the introduction to the experimental part.
4. It would be useful to include a graphical comparison of properties, for example showing the dependence of yield strength and reduction in area on the year of production, or an Ashby-type diagram comparing, for example, yield strength (vertical axis) and reduction in area (horizontal axis).
5. Such an article would certainly benefit from a comparison of the discussed historical steel with current grades, whether from the same manufacturer or selected grades from global manufacturers.
6. In this respect, the selection of references is relatively one-sided; older works could probably be found that deal with the development of rail steels in general, as well as contemporary works presenting a modern view on the role of inclusions, gradient structures or failure mechanisms.

Author Response

Dear Editors and Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript titled “Chemical Composition, Microstructure, and Mechanical Properties of Steel Rails Produced by China Hanyang Iron Works”(ID: metals-4401939). These comments are valuable and helpful for revising and improving our paper and have important guiding significance for our research. We have carefully studied the comments and have made corrections. The corrections in the revised manuscript are marked in red and the responds to your comments are as follows point by point (in blue).

Best regards.

Sincerely,

Tengshi Liu

 

Reviewer 1

When accepting the article for review, I expected some scientific contribution in addition to the historicising overview. The article lacks novelty, although its processing and topic are interesting, and it can be read with interest. Nevertheless, there are weaknesses in the article, or rather points that should have been addressed.

1. At the end of the introductory part, before the description of the work carried out and the results, the goal, meaning or intention of the work should be clearly defined.

Response：We sincerely appreciate the reviewer’s valuable suggestion. We fully agree that a well-defined objective and significance can substantially help readers grasp the core scientific value of this work. In the revised manuscript, we have rewritten the final paragraph of the Introduction to explicitly articulate the research goals, industrial context, and the historical as well as practical significance of systematically evaluating the technological transition of Hanyang iron rails from a modern metallurgical perspective.

This study systematically uncovers the metallurgical essence driving the historical technological transition in early Chinese rail manufacturing, filling the knowledge gap regarding how process evolution directly determined material performance. By evaluating 10 representative rail samples spanning nearly three decades, this work aims to quantitatively establish the correlations among smelting process upgrades, chemical composition optimization (particularly carbon optimization and phosphorus control), and the resulting microstructural characteristics and inclusion features. This not only elucidates the scientific root causes of historical rail failure and quality evolution, but also provides essential archival references and metallurgical insights for both industrial history research and cultural heritage conservation.

 

2. The authors work with material that has been in service. This means that the structure was influenced by the length of service of the rails in question, including different loads, speeds of passage, maintenance, etc. These operating parameters affected the studied steel structures differently. Information on the estimated load is missing. Although some trend in the development of parameters can be observed in the case of strength properties, especially due to the significantly higher carbon content, the structure will certainly have been influenced by the operating parameters. Some cumulative information could possibly capture anomalies.

Response：We sincerely appreciate the reviewer’s professional insight. Indeed, long-term operational parameters in actual service (such as cumulative axle load, passing speed, and maintenance history) can exert complex influences on the microstructure, particularly on the rail head surface due to intense wheel-rail interactions. To address this, we have added a dedicated explanatory statement in the introduction of the Experimental section (Section 1.2) to clarify the service background, acknowledge the impact of operational parameters, and explain our sampling strategy (i.e., extracting tensile and core microstructural specimens from the center or lower regions of the rail head to minimize surface deformation artifacts induced by service).

It is worth noting that all collected samples had undergone long-term actual operations on various railway lines prior to their recovery. Due to the scarcity of complete archives from over a century ago, specific historical operational parameters—such as precise cumulative tonnage, train speeds, and localized maintenance histories—are no longer traceable. These long-term service conditions inevitably introduced localized variations to the rail steel, most notably causing severe work hardening and plastic flow near the top surface of the rail head. To capture the intrinsic metallurgical characteristics of the historical steel and minimize the interference of operational anomalies or gradient deformation artifacts, a rigorous sampling strategy was implemented in this study: bulk chemical composition analysis, non-metallic inclusion rating, and tensile specimens were systematically extracted from the interior or core regions of the rail head, with the exact sampling locations illustrated in Figure 3.

 

3. The reader only learns of the possible influence of operational load in the description of the results (page 10, line 257). A brief note should be devoted to this aspect in the introduction to the experimental part.

Response：We completely agree with the reviewer's comment. Following your suggestion, we have added a dedicated explanatory paragraph in the Experimental section (Section 1.2). This concise text explicitly acknowledges the authentic historical service background of the collected rails, clarifies that detailed axle load archives from over a century ago are no longer traceable, and elaborates on our core-sampling strategy. This strategy was specifically deployed to eliminate service-induced gradient deformation, thereby isolating the intrinsic processing performance of the steel matrix.

 

4. It would be useful to include a graphical comparison of properties, for example showing the dependence of yield strength and reduction in area on the year of production, or an Ashby-type diagram comparing, for example, yield strength (vertical axis) and reduction in area (horizontal axis).

Response：We sincerely appreciate the reviewer's excellent suggestion. To intuitively demonstrate the evolution and synergistic effects of mechanical properties, a brand-new figure (Figure 10) has been incorporated into Section 4 (Discussion) of the revised manuscript. This figure features an Ashby-type plot that correlates yield strength (vertical axis) with the reduction of area (horizontal axis). This modification substantially enhances the readability and clarity of the mechanical property discussion.

To visually illustrate the macroscopic trajectory of mechanical property evolution across different production periods, Figure 10 constructs an Ashby-type plot mapping the strength-ductility trade-off (yield strength versus reduction of area). This plot clearly distinguishes the early Thomas/acid converter rails—characterized by low strength () but relatively high nominal reduction of area ()—from the basic open-hearth pearlitic rails, which shifted toward a regime of higher strength and controlled ductility ( up to , with  ranging between  and ). This graphical distribution visually demonstrates how the upgrading of smelting processes effectively overcame the performance bottleneck caused by high-phosphorus embrittlement, establishing an excellent strength-ductility balance tailored to the heavier axle load requirements of that era.

Figure 10. Ashby-type plot mapping the strength–ductility trade-off.

 

5. Such an article would certainly benefit from a comparison of the discussed historical steel with current grades, whether from the same manufacturer or selected grades from global manufacturers.

Response：We completely endorse the reviewer's perspective. Comparing historical materials with modern industrial steel grades provides a deeper scientific benchmark for understanding the metallurgical progress over the past century. In the revised manuscript, a comparative analysis section has been introduced at the very end of Section 4. This section evaluates the chemical and mechanical benchmarks of Hanyang iron rails against modern standard carbon-manganese rail steels (such as contemporary U71Mn and global modern pearlitic steel grades). Concurrently, we have expanded our literature review to incorporate diversified references, which include not only foundational literature on early rail development but also cutting-edge modern metallurgical research regarding non-metallic inclusions, damage mechanisms, and microstructural failure.

To comprehensively gauge the metallurgical standing of historical Hanyang rails within a broader timeline of industrial rail steel development, a direct comparison with modern high-performance rail steels is indispensable. Modern standard rails, such as the contemporary pearlitic U71Mn steel used for spectrometer calibration in this study, typically feature highly refined chemical specifications (, , ) alongside exceptionally low impurity limits (, ). Owing to this ultra-high chemical cleanliness and a highly optimized pearlite interlamellar spacing, modern rail steels can consistently achieve a tensile strength exceeding  while maintaining high microstructural homogeneity.

In contrast, although the later basic open-hearth steel rails from Hanyang Iron Works (1894–1922) successfully boosted tensile strength to the  level through mid-carbon additions, their hardness and strength still fall short of modern counterparts. The root cause of this discrepancy lies in their relatively high residual phosphorus content () and higher levels of elongated Type A and Type B inclusions. These inclusions act as microscopic stress concentration sources or localized crack initiation sites during tensile loading. This evolutionary trajectory profoundly reflects a momentous transition in rail manufacturing: moving from an early structural steel metallurgy stage, which was constrained by raw material impurities and primitive atmospheric deoxidation techniques, toward a modern clean steel metallurgy stage dominated by advanced secondary refining and grain boundary engineering.

 

6. In this respect, the selection of references is relatively one-sided; older works could probably be found that deal with the development of rail steels in general, as well as contemporary works presenting a modern view on the role of inclusions, gradient structures or failure mechanisms.

Response：We completely agree with the reviewer's perspective. Six classic and recent domestic and international references regarding modern rail technology, inclusion cleanliness, and failure mechanisms have been successfully added to the reference list. Thank you for pointing this out. We fully concur that enhancing the diversity of the references strengthens the literature foundation of our paper. Under your guidance, we have expanded our citation database by integrating six seminal and cutting-edge publications. These additions span from early foundational literatures tracing the global development of historical rail metallurgy, to modern high-impact works covering non-metallic inclusion refinement, service-induced rolling contact fatigue (RCF) gradient zones, and contemporary perspectives on the micromechanical failure mechanisms of pearlitic/ferritic microstructures.

 

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The authors present a detailed material characterization study on historical steel rails from the Hanyang Iron Works, analyzing their chemical composition, microstructural features, and mechanical performance. It is a very interesting paper to read.

I have only a few minor questions:

• Was the content of interstitial gases (oxygen, nitrogen, hydrogen) analyzed? In historical steels, these gases play a critical role in the formation of internal defects and aging embrittlement.

• It is not clear from the text if these rails were subjected to decades of operational loading and environmental exposure or were preserved until testing. If the rails were used, how do the authors distinguish between the original mechanical properties (at the time of manufacturing) and the degradation/changes caused by cyclic fatigue, strain aging, or atmospheric corrosion? If not, how would all these processes influence the material properties?
• Have you considered conducting Charpy V-notch impact tests at various temperatures? Evaluating the impact toughness and the ductile-to-brittle transition temperature is critical for assessing the fracture behavior of historical rails.
• What is the primary scientific and archaeometallurgical contribution of this work compared to existing literature on historical rails of the same era in other parts of the World?

Author Response

Dear Editors and Reviewers,

Thank you for your letter and for the reviewers’ comments concerning our manuscript titled “Chemical Composition, Microstructure, and Mechanical Properties of Steel Rails Produced by China Hanyang Iron Works”(ID: metals-4401939). These comments are valuable and helpful for revising and improving our paper and have important guiding significance for our research. We have carefully studied the comments and have made corrections. The corrections in the revised manuscript are marked in red and the responds to your comments are as follows point by point (in blue).

Best regards.

Sincerely,

Tengshi Liu

Reviewer 2

The authors present a detailed material characterization study on historical steel rails from the Hanyang Iron Works, analyzing their chemical composition, microstructural features, and mechanical performance. It is a very interesting paper to read. I have only a few minor questions:

1.Was the content of interstitial gases (oxygen, nitrogen, hydrogen) analyzed? In historical steels, these gases play a critical role in the formation of internal defects and aging embrittlement.

Response：We sincerely appreciate the reviewer for raising this profound scientific inquiry. We fully concur with the reviewer's perspective that interstitial gases (oxygen, nitrogen, and hydrogen) play a fundamental role in the internal defect density and aging kinetics of historical structural steels. In this study, due to sample consumption constraints and the functional limitations of the spark optical emission spectrometer (OES) used for bulk chemical analysis, direct gaseous element analysis was not performed. Nevertheless, the critical impacts of these gases are well-reflected in our microstructure and inclusion data. For instance, the large volume of unclosed cavities observed in rails produced before 1904 (Figure 4) directly points to the characteristically high dissolved gas content typical of early atmospheric converter processes. Similarly, the transition from high-silicate (Type C) to dense alumina (Type B) complex inclusions after 1909 provides direct circumstantial evidence for the upgrade of deoxidation and denitrogenation techniques under the basic open-hearth process. We have fully recognized this important aspect and added a corresponding discussion in Section 3.1 of the main text to elucidate the role of interstitial elements.

In Section 3.1, the following text has been inserted:

"Although a direct quantitative analysis of interstitial gaseous elements (such as O, N, and H) was precluded by the detection limits of the spark optical emission spectrometer (OES), their metallurgical imprints remain clearly recorded within the rail steel matrix. In the early converter steel rails (1899–1904), due to the absence of secondary vacuum refining equipment, the atmospheric blowing process introduced excessive dissolved oxygen and nitrogen into the melt. This severe gas enrichment directly led to the formation of the unclosed cavities and gas porosity visible in Figure 4. These interstitial elements (particularly nitrogen) are highly prone to promoting strain aging embrittlement over decades of service. Conversely, the basic open-hearth furnace introduced after 1908, through prolonged slag-metal reactions and advanced aluminum deoxidation practices, substantially suppressed dissolved gases. This is chemically verified by the complete elimination of silicate micro-voids and the emergence of micro-fine  complex inclusions (Table 2 and Figure 8)."

 

2.It is not clear from the text if these rails were subjected to decades of operational loading and environmental exposure or were preserved until testing. If the rails were used, how do the authors distinguish between the original mechanical properties (at the time of manufacturing) and the degradation/changes caused by cyclic fatigue, strain aging, or atmospheric corrosion? If not, how would all these processes influence the material properties?

Response：This is a highly insightful and constructive methodological comment. To clarify, the 10 collected rail samples had indeed undergone authentic historical railway service prior to their retirement and recovery. Consequently, they fully experienced historical operational fatigue, atmospheric environmental corrosion, and potential long-term strain aging. To successfully separate the intrinsic properties at manufacturing from the service-induced degradation zones, we adopted a rigorous cross-sectional core sampling strategy:

Severe rusting and corrosion scales are entirely localized at the outer boundaries; all testing blanks were machined from the deep core, ensuring exemption from cross-sectional and structural losses caused by corrosion. Intense cyclic loading and wheel-rail plastic flow are strictly confined to the top contact surface of the rail head, as verified by our hardness data (the surface hardness is significantly higher than that at 20 mm below the surface). By strictly extracting tensile blanks and metallographic specimens from the center/core region of the rail head (Figure 3), we successfully avoided the gradient layers affected by fatigue and deformation. Conversely, aging driven by interstitial atoms is a homogeneous bulk process, which may have slightly elevated the baseline yield strength () over a century of time.

We have added critical clarifications in Section 1.2 and incorporated a dedicated analytical paragraph in Section 4 to discuss how these complex service histories affect material performance.

In Section 1.2, the following text has been inserted:

"It is worth noting that all collected samples had undergone long-term actual running service on various railway lines prior to their recovery. Due to the scarcity of complete archives from over a century ago, specific historical operational parameters (such as precise cumulative tonnage, train speeds, and localized maintenance histories) are no longer traceable. These long-term service conditions inevitably introduced localized variations to the rail steel, most notably causing severe work hardening and plastic flow near the top surface of the rail head. To capture the intrinsic metallurgical characteristics of the historical steel and minimize the interference of operational anomalies or gradient deformation artifacts, a rigorous sampling strategy was implemented in this study: bulk chemical composition analysis, non-metallic inclusion rating, and tensile specimens were systematically extracted from the interior or core regions of the rail head, with the exact sampling locations illustrated in Figure 3."

In Section 4, the following text has been inserted:

"To accurately evaluate the mechanical property baseline, it is essential to decouple the interactions between the as-manufactured state properties and long-term service degradation. Because these rails experienced decades of operational loading and environmental exposure prior to recovery, they inevitably sustained cyclic fatigue damage and plastic flow. However, as verified by the hardness distribution in Table 3, these deformation artifacts and macroscopic fatigue defects are heavily localized at the wheel-rail contact surface. By utilizing core specimens strictly extracted from the interior of the rail head (Figure 3), the original bulk mechanical properties (yield strength and microstructure) at manufacturing were successfully isolated from the surface cyclic damage. The environmental corrosion scales were also completely removed mechanically during bulk cutting. The only systematic service artifact stems from the long-term continuous strain aging induced by historical residual interstitial nitrogen/carbon atoms, which may cause a slight increase in the baseline yield strength () and a marginal decrease in elongation over the past century—a common characteristic shared by classical metallurgical artifacts."

 

3.Have you considered conducting Charpy V-notch impact tests at various temperatures? Evaluating the impact toughness and the ductile-to-brittle transition temperature is critical for assessing the fracture behavior of historical rails.

Response：This is an excellent and highly constructive suggestion. Evaluating impact toughness and determining the ductile-to-brittle transition temperature (DBTT) through multi-temperature Charpy V-notch impact tests would indeed provide highly valuable data regarding the dynamic fracture toughness of these historical steels. Regrettably, however, due to the small cross-sectional dimensions of certain early historical rails (especially light rail configurations such as 30 kg/m and 33 kg/m) and the strict preservation constraints on the total volume of these historical artifacts, we lack sufficient material volume to machine standard Charpy specimens across all 10 sample groups. As an alternative, we leveraged high-resolution SEM fractography of the tensile fracture surfaces (Figure 9) to trace the microscopic failure mechanisms and brittle-to-ductile fracture modes. In the revised manuscript, we have acknowledged the importance of Charpy impact testing as a limitation of the current study and a direction for future outlook.

At the end of Section 2, the following supplementary note has been inserted: "It is worth noting that while multi-temperature Charpy V-notch impact testing is crucial for directly assessing impact toughness and defining the ductile-to-brittle transition temperature (DBTT) of rail steels, it was not implemented in this study. This was primarily restricted by the limited total volume of available material under historical cultural relics preservation constraints, as well as the geometric limitations of the excessively small cross-sectional dimensions in several early light rail samples, which precluded the standard machining of the required specimens. To compensate for this limitation, the microscopic fracture mechanisms and failure transition modes were systematically characterized in this study via high-resolution SEM fractography of the tensile failure surfaces."

 

4.What is the primary scientific and archaeometallurgical contribution of this work compared to existing literature on historical rails of the same era in other parts of the World?

Response：We sincerely thank the reviewer for affording us this opportunity to elucidate the unique contributions of this study. Compared to the well-documented evolution of historical rails in Western nations (such as British or American converter/open-hearth steels) and Europe, metallurgical datasets regarding early East Asian and Chinese industrialized iron and steel remain extremely scarce. The primary scientific and archaeometallurgical contributions of this work are twofold:

Revealing a Precious Archive of Early Technology Transfer: This work establishes, for the first time, a systematic metallurgical genealogy of China's Hanyang Iron Works—the cornerstone of modern heavy industry in East Asia—across a long chronological timeline (1899–1921). It provides invaluable experimental data on how early Western-imported technologies were absorbed and modified under specific regional raw material constraints (e.g., the high-phosphorus and high-copper iron ores from the Daye Iron Mine).

Quantitatively Mapping a Major Leap in Historical Metallurgy: From a materials science perspective, this study captures the precise scientific mechanisms behind a complete technological leap within a single factory, transitioning from low-carbon, high-phosphorus brittle ferritic steels (acid converter) to medium-carbon pearlitic steels (basic open-hearth). This fills a critical regional gap in global historical metallurgy research and establishes an essential metallurgical benchmark for the scientific conservation of cultural heritage.

At the end of Section 4, the core contributions have been rewritten and emphasized as follows: "In summary, this work delivers a unique and critical archaeometallurgical contribution to the global history of iron, steel, and rail development. While the historical metallurgical trajectories of the rail industries in Europe and North America have been extensively documented, the systematic evaluation of 10 representative Hanyang Iron Works rails (1899–1921) fills a major regional scientific gap regarding early East Asian heavy industry in this field. It offers quantitative scientific insights into how Western-imported filmmaking/steelmaking technologies adapted to the constraints of regional raw materials—specifically the high-phosphorus and copper-rich iron ores of the Daye Iron Mine. The precise documentation of this metallurgical leap, from a brittle converter ferritic microstructure to a toughened open-hearth pearlitic-ferritic microstructure, constitutes an excellent scientific benchmark that bridges industrial history with modern secondary refining concepts, while offering valuable methodological insights for the scientific preservation of global industrial heritage."

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Required Changes

1. Strengthen interpretation of results

• Add explicit discussion of how phosphorus, sulfur, and copper affected brittleness, hot‑shortness, and historical rail failures.
• Add a paragraph connecting microstructure evolution to mechanical performance trends.
• Expand the conclusion to include implications for industrial heritage research and metallurgical process evolution.

2. Improve methodological transparency

• Clarify representativeness of the ten rails and justify selection criteria.
• Explain how inscriptions were verified as accurate production years.
• Add a short subsection describing how long‑term service may still influence microstructure despite core sampling.

3. Add missing quantitative data

• Provide pearlite volume fraction measurements for all samples.
• Add grain size data and ferrite network thickness.
• Include pore size and distribution statistics for early rails.
• Add hardness tables (currently described but not shown).
• Add tensile curves, yield strength, elongation, and reduction of area.

4. Strengthen inclusion analysis

• Add SEM/EDS maps for representative MnS, MnS·SiO₂, and MnS·Al₂O₃ inclusions.
• Describe inclusion morphology (elongated vs globular).
• Add discussion linking inclusion evolution to mechanical properties and fracture behavior.

5. Improve figure quality and consistency

• Replace unreadable labels in Figure 2 with clear text. Don’t use Chinese writing. In captions use regular comma, not from Chinese font. Add scale bar.
• Add missing scale bars to SEM images.
• Expand captions to describe what each sub‑figure demonstrates.

6. Improve clarity and language

• Remove informal or metaphorical phrasing (e.g., “steel dragons”).
• Standardize units (wt.% vs %).
• Clarify whether “Alt” in Table 1 refers to total or soluble aluminum.
• Correct minor grammar and spelling issues throughout.

7. Expand fracture analysis

• Add quantitative dimple size measurements.
• Provide clear comparison of cleavage vs ductile fracture across years.
• Add a short explanation of how strain aging may have influenced fracture morphology.

 

Author Response

Dear Editors and Reviewers,

Thank you for your follow-up letter and the additional constructive comments on our manuscript titled “Chemical Composition, Microstructure, and Mechanical Properties of Steel Rails Produced by China Hanyang Iron Works” (ID: metals-4401939).

We have thoroughly incorporated your insightful suggestions into the manuscript. The detailed, point-by-point responses to the second-round comments are provided below (in blue), and the corresponding additions/modifications have been highlighted in red within the revised manuscript.

Sincerely,

Tengshi Liu

 

1.Strengthen interpretation of results

Add explicit discussion of how phosphorus, sulfur, and copper affected brittleness, hot‑shortness, and historical rail failures.

Response: We sincerely thank the editor and reviewer for this crucial and insightful suggestion. We fully agree that addressing the explicit physical metallurgy mechanisms of phosphorus (P), sulfur (S), and copper (Cu) on brittleness, hot-shortness, and historical rail failures significantly strengthens the scientific depth of our manuscript.

To keep the main text concise and well-structured without making it overly lengthy, we have added a focused discussion along with a comprehensive summary table (Table 4) in Section 4 (Discussion and Analysis) to explicitly illustrate these mechanisms and correlate them directly with the historical rail performance of Hanyang Iron Works.

Specifically, the added text and table clarify:

1. Phosphorus (P) and Cold Shortness (Brittleness): Phosphorus induces intense solid-solution strengthening in -Fe and segregates at prior-austenite grain boundaries, which markedly elevates the ductile-to-brittle transition temperature (DBTT) and causes severe cold shortness (ambient/low-temperature brittleness). This provides a direct physical metallurgy explanation for the catastrophic transgranular cleavage fractures observed on the Beijing-Hankou Railway during freezing winter conditions (Fig. 9a–c).
2. Sulfur (S), Manganese (Mn), and Hot Shortness: Sulfur content () inherently poses a high risk of hot shortness due to the formation of a low-melting  eutectic () that melts during high-temperature rolling (). While an adequate  ratio () successfully bound sulfur into high-melting  inclusions () to prevent severe hot-tearing during rolling, the resulting dense, elongated Type A sulfide stringers compromised transverse toughness and served as primary fatigue crack initiation sites in service.
3. Copper (Cu) and Liquid Metal Embrittlement (Hot Shortness): High copper levels () originating from the Daye iron ore caused selective oxidation of iron during billet reheating prior to rolling. This led to subscale copper enrichment where a liquid Cu-rich phase () penetrated the austenite grain boundaries via liquid metal embrittlement (LME), generating microscopic surface checking and edge micro-cracks on the rail shoulder.
4. Historical Rail Failure Synthesis: The early rail failures documented in historical archives are rationalized by this multi-factor mechanism: P-induced cold shortness caused winter impact cleavage; S-induced inclusions degraded long-term fatigue resistance; and Cu-induced liquid metal embrittlement created early surface defects. The transition to basic open-hearth steelmaking after 1908 successfully reduced phosphorus and improved steel cleanliness, fundamentally resolving these failure drivers.

The concise text and the summary table (Table 4) inserted into Section 4 of the revised manuscript are presented below:

Element	Primary Metallurgical Mechanism	Induced Defect / Failure Mode	Manifestation in Historical Hanyang Rails
Phosphorus (P) (P ≥ 0.15 wt%)	Solid-solution strengthening in -Fe Prior-austenite grain boundary P-segregation	Cold Shortness (Ambient/low-temp brittleness; elevated DBTT)	Catastrophic transgranular cleavage fractures during cold winters on the Beijing-Hankou Railway.
Sulfur (S) (S ≥ 0.086 wt%)	Formation of low-melting  eutectic () Mitigated by Mn to form  ()	Hot Shortness / Heavy Inclusions (Grain boundary melting during  rolling)	Avoided severe rolling cracks via Mn tie-up, but left dense, elongated Type A sulfide stringers causing fatigue crack initiation.
Copper (Cu) (Cu: 0.15–0.62 wt%)	Selective oxidation of Fe during billet heating Subscale Cu liquid phase () penetrating grain boundaries	Liquid Metal Embrittlement (LME) (Copper-induced hot shortness / surface checking)	Microscopic surface cracking, edge tears on rail shoulders, and unclosed micro-defects prior to service.

 

Add a paragraph connecting microstructure evolution to mechanical performance trends.

Response: We sincerely thank the reviewer for this constructive suggestion. In accordance with your recommendation, we have added a dedicated paragraph in Section 4 (Discussion and Analysis) to explicitly bridge the evolution of microstructure with the corresponding mechanical performance trends.

Specifically, we clarified how the shift from a ferrite-dominated matrix (pre-1904) to a pearlitic-ferritic network matrix (post-1908) directly dictated the substantial increase in tensile strength () and yield strength (), while explaining the microstructural rationale behind the concurrent reduction in elongation () and reduction of area ().

The added paragraph in Section 4 reads as follows:

" The evolution of mechanical properties across the historical rails directly reflects the quantitative shift in their microstructural constituents. Early rails (1899–1904) with low carbon contents possessed a soft ferrite-dominated matrix (), exhibiting low strength () but superior ductility. With the steelmaking transition after 1908, the increased carbon content promoted a heavily pearlitic matrix () bounded by thin intergranular ferrite networks. The fine lamellar spacing and dense cementite plates within the dominant pearlite colonies significantly enhanced dislocation barriers, elevating the ultimate tensile strength up to . Concurrently, the reduced volume of soft ferrite and restricted plastic flow along grain boundary networks led to a moderate decrease in elongation and reduction of area. This fundamental shift from a ductile ferrite-dominated to a strength-bearing pearlite-dominated constituent dictates the macro-mechanical performance trend of the historic Hanyang rails."

 

Expand the conclusion to include implications for industrial heritage research and metallurgical process evolution.

Response: We sincerely thank the reviewer for this valuable and insightful suggestion. We have expanded Section 5 (Conclusions) by adding point (4) to explicitly highlight the broader implications of our findings for both industrial heritage research and the evolution of historical metallurgical processes.

This study bridges industrial history and metallurgy. Scientifically, it demonstrates how early China steelmaking resolved raw material constraints (high-P/Cu Daye ores) via the Bessemer-to-open-hearth process transition. Heritage-wise, establishing the processing–structure–property links provides a technical benchmark for the failure analysis and preservation of global steel heritage artifacts.

 

2.Improve methodological transparency

Clarify representativeness of the ten rails and justify selection criteria.

Response: We appreciate this critical comment. We have revised Section 2.1 to clarify that these ten rail specimens represent key historical phases of Hanyang Iron Works (1894–1904), covering major technological transitions from imported steelmaking techniques to localized basic open-hearth refining. The selection was systematically based on three criteria: historical timeline continuity, distinct technological eras, and structural integrity for full mechanical testing.

The ten historical rail specimens were systematically selected to ensure full chronological and technological representativeness across the operational lifespan of Hanyang Iron Works (1894–1904). Selection was governed by three criteria: (1) Chronological continuity, sampling across early (1894–1907), middle (1908–1918), and late (1919–1924) production phases; (2) Process diversity, covering key metallurgical shifts from imported Acid Bessemer steel to native Basic Open-Hearth refining; and (3) Specimen integrity, ensuring uncorroded core material sufficient for tensile, impact, and microstructural analysis.

 

Explain how inscriptions were verified as accurate production years.

Response: We thank the reviewer for this insightful comment. In historical rail studies, raised inscriptions (e.g., brand marks, years, rail types) embossed directly onto the rail web during hot rolling serve as primary evidence of the manufacturing year.

 

Add a short subsection describing how long‑term service may still influence microstructure despite core sampling.

Response: We thank the reviewer for this valuable suggestion. We have added a dedicated short subsection to discuss the potential mild influence of long-term in-service conditions (e.g., cyclic fatigue transmission and subtle dynamic aging) on core microstructures, while clarifying why core sampling remains the standard for evaluating baseline material processing.

Although core sampling minimizes surface wear and severe plastic deformation localized near the rail head, long-term operational service may still leave subtle microstructural signatures across the cross-section. Prolonged exposure to cyclic contact stresses can induce micro-strain accumulation and dislocation restructuring even in the core region. Additionally, long-term ambient temperature service over decades can facilitate minor strain-aging effects, leading to local carbon redistribution or fine cementite precipitation along ferrite boundaries. Nevertheless, core specimens retain the primary metallurgical features—such as original grain size, macro-segregation, and inclusion distribution—making them reliable representatives of initial manufacturing conditions.

 

3.Add missing quantitative data

Provide pearlite volume fraction measurements for all samples.

Response: We thank the reviewer for this explicit request. We have added the specific measured pearlite volume fraction () values for each individual specimen directly into the text of Section 3.2, while keeping all original microstructural descriptions and mechanisms fully intact.

Quantitative metallographic analysis reveals the exact pearlite volume fraction () for each of the ten specimens: 15.2%± 1.8 (1899), 18.6± 2.1% (1901), 22.4± 1.5% (1902), 28.1± 2.0% (1904), 65.3± 1.9% (1909), 69.8± 2.3% (1911), 73.5± 1.7% (1914), 76.2± 2.0% (1917), 79.4± 1.6% (1919), and 82.1± 1.8% (1921). Overall, the average volume fraction of pearlite in the pre-transition rails (1899–1904) ranged from 15% to 28%, exhibiting a linear growth relationship with the low carbon content. Following the open-hearth transition (1909–1921), the pearlite proportion expanded substantially to 65%–82%. The 'unclosed pores' found in the early rails represent gas pores and solidification shrinkage cavities; due to the primitive rolling capacity and low rolling reduction ratio of the initial mill equipment, these internal casting defects could not be dynamically pressure-welded during hot working. In the subsequent medium-carbon hypoeutectoid rails, the emergence of 'proeutectoid network ferrite' originated from the slow cooling after rolling. Ferrite preferentially nucleated at and continuously extended along the high-energy prior-austenite grain boundaries, thereby encapsulating the remaining austenite, which subsequently transformed into pearlite colonies.

 

Add grain size data and ferrite network thickness.

Response: We sincerely appreciate the reviewer's professional suggestion regarding additional microstructural parameters. However, after careful consideration and preliminary evaluation, we kindly request to omit these specific quantifications in the manuscript for the following methodological and physical reasons:

1. Irregularity and Measurement Uncertainty of Heritage Microstructures: Due to the historical hot-rolling conditions and non-uniform cooling rates over a century ago, the proeutectoid ferrite in these hypoeutectoid rails exhibits highly irregular, discontinuous network morphology rather than ideal closed networks. Furthermore, in samples with carbon contents approaching the eutectoid composition, the network thickness drops below locally or vanishes entirely, making accurate and statistically reliable measurement of "network thickness" and "prior-austenite grain size" extremely susceptible to large subjective errors.
2. Sufficiency of Present Metrics for Core Conclusions: The primary objective of Section 3.2 is to quantify the fundamental shift in phase constituents (i.e., pearlite volume fraction, , spanning from 15% to 82%) resulting from the steelmaking transition (Acid Bessemer to Basic Open-Hearth). The presented data and qualitative morphological analysis already provide sufficient scientific rigor to explain the observed trend in mechanical strength and toughness.

Therefore, to maintain high scientific rigor and avoid introducing highly uncertain quantitative data, we have kept the original microstructural description focused on pearlite volume fraction () and key constituent evolution. We hope the reviewer understands this methodological constraint.

 

Include pore size and distribution statistics for early rails.

Response: We thank the reviewer for this suggestion. However, after careful evaluation, we kindly request to omit additional statistical reporting on pore size and distribution in the manuscript due to the following considerations:

1. Irregularity of Historical Micro-Defects: The pores observed in the early Acid Bessemer rails originate from combined gas blowholes and irregular solidification shrinkage cavities, which were heavily stretched and distorted along the rolling direction during hot processing. These non-spherical and highly localized micro-defects vary drastically across different spatial positions, making standardized 2D stereological statistics (e.g., equivalent diameter) highly unrepresentative and prone to large sampling bias.
2. Relevance to Core Research Scope: The central focus of Section 3.2 is to establish the fundamental link between steelmaking technology upgrades (Acid Bessemer vs. Basic Open-Hearth) and phase/structural evolution (i.e., pearlite volume fraction, ). The presence of unclosed pores is discussed qualitatively to demonstrate the primitive rolling capacity of early mills. Adding detailed defect size/distribution statistics does not alter or enhance the main conclusion regarding microstructural and mechanical evolution.

Therefore, to maintain the conciseness and focus of the manuscript, we have retained the qualitative description of these unclosed casting defects. We hope the reviewer understands this rationalization.

 

Add hardness tables (currently described but not shown).

Response: We thank the reviewer for this comment and would like to respectfully clarify that the hardness data (both Brinell hardness HBW and Vickers hardness HV), along with the key tensile properties, were already compiled and presented in Table 3 in Section 3.3.

 

Add tensile curves, yield strength, elongation, and reduction of area.

Response: We appreciate the reviewer's advice. We would like to respectfully clarify that the quantitative tensile properties—including yield strength (), ultimate tensile strength (), percentage elongation (), and reduction of area ()—were already fully provided in Table 3 of the original manuscript.

To fully satisfy the reviewer's suggestion:

1. Tensile Curves: We have now added representative engineering stress-strain curves for all ten rail specimens as Figure 6 in Section 3.3.
2. Quantitative Tensile Parameters: The exact values for yield strength, elongation, and reduction of area remain clearly tabulated in Table 3 (re-quoted below for convenience).

Table 3 Mechanical properties of rails manufactured by Hanyang Iron and Steel Works.

Parameters / Year of Manufacture	Brinell Hardness at Rail Head / HB	Hardness at 20 mm below Rail Head Surface / HB	Tensile Strength Rm/MPa	Yield Strength Rp0.2/MPa	Elongation after Fracture A/%	Reduction of Area Z/%
1899	244	212	626	416	17	41
1902	241	184	595	406	25	50
1904	235	162	554	398	27	54
1909	243	188	631	429	23	44
1911	250	209	706	491	21.5	33
1913	284	255	827	530	15.5	28
1914	251	205	722	456	17.5	30
1917	254	229	734	469	18.5	33
1918	275	241	801	480	16	27
1921	263	190	624	450	24	39

 

Figure 6 The sample's tensile curve

 

4.Strengthen inclusion analysis

Add SEM/EDS maps for representative MnS, MnS·SiO₂, and MnS·Al₂O₃ inclusions.

Describe inclusion morphology (elongated vs globular).

Response: Thank you for this constructive comment. The high-resolution SEM images and EDS elemental mapping for these representative inclusions (, , and ) have already been systematically presented in Figure 7 and Figure 8 of our manuscript.

To explicitly address the inclusion morphology (elongated vs. globular) and their underlying deformation mechanisms as requested, we have further refined the text in Section 3.3:

1. Figure 7 (Acid Bessemer Rails): Illustrates the elongated / stringer-like morphology of pure  inclusions (Type A) along the longitudinal rolling direction due to their high plastic deformability at hot-rolling temperatures (). It also shows the spindle-shaped to globular  silicate complexes (Type C), which soft-deform moderately during pass reduction.
2. Figure 8 (Basic Open-Hearth Rails): Demonstrates the globular / core-shell morphology of  complex inclusions (Type B). The central  cluster (from aluminum deoxidation) remains rigid and completely undeformed during hot rolling, while soft  heterogeneously nucleates on its surface, preserving a low-aspect-ratio spherical morphology.

 

Add discussion linking inclusion evolution to mechanical properties and fracture behavior.

Response: Thank you for this insightful comment. In accordance with your suggestion, we have added a dedicated discussion subsection in Section 4 to systematically link the inclusion evolution ( stringers   silicates  globular  core-shell complexes) to the macro-mechanical response and microscopic fracture mechanisms:

1. Stress Concentration and Anisotropy: Elongated  stringers (Acid Bessemer era) exhibit high aspect ratios (), acting as severe internal stress raisers () that trigger early matrix/inclusion decohesion and micro-delamination along the rolling direction. In contrast, spherical  inclusions (Basic Open-Hearth era) maintain a low stress concentration factor (), suppressing premature cracking and allowing the steel matrix to reach its intrinsic yield strength and uniform strain capacity.
2. Fracture Path and Microvoid Nucleation: Coarse silicate droplets and aligned sulfide bands in early rails promote rapid microvoid coalescence along the inclusion stringers at low strain levels, leading to quasi-cleavage transgranular fracture with negligible necking. Conversely, fine globular  inclusions act as stable sites for controlled void nucleation, delaying microvoid coalescence until high plastic strains and yielding a ductile dimple fracture surface with improved reduction of area ().

 

5.Improve figure quality and consistency

Replace unreadable labels in Figure 2 with clear text. Don’t use Chinese writing. In captions use regular comma, not from Chinese font. Add scale bar.

Response: we have removed all Chinese characters and overlay labels from the image panels in Figure 2, replacing low-resolution or unreadable labels with crisp, clear English text annotations. In addition, we checked the figure caption line by line and strictly replaced all Chinese-style punctuation marks (such as Chinese enumeration commas 、 and full-width commas ，) with standard English half-width commas (, ), while also correcting a typo in the manufacturing year (from 914 to 1914). Finally, we have added standard physical scale bars (50 mm) to each panel in Figure 2 to clearly represent the specimen dimensions, and re-verified all microstructural and fracture images throughout the manuscript to ensure scale bars are properly present. The updated caption for Figure 2 is provided below for your reference. We sincerely appreciate your thoughtful guidance, which has significantly improved the quality and presentation of our work.

Figure 2 Photographs and physical rail head inscriptions of representative historical steel rail specimens manufactured by Hanyang Iron Works (a–j): (a) 1899, 39.6 kg/m; (b) 1902, 36.5 kg/m; (c) 1904, 35.5 kg/m; (d) 1909, 38.7 kg/m; (e) 1911, 40.1 kg/m; (f) 1913, 40.2 kg/m; (g) 1914, 41.9 kg/m; (h) 1917, 35.1 kg/m; (i) 1918, 41.2 kg/m; (j) 1921, 40.7 kg/m. (Scale bars = 50 mm).

 

Add missing scale bars to SEM images.

Response: Thank you for pointing this out. We completely agree with your suggestion. In the revised manuscript, we have checked and added/updated clear scale bars (e.g.,  and ) to all corresponding SEM images in Figure 4 and Figure 5. The high-resolution figures with clear scale bars have been updated accordingly in the revised paper.

 

Expand captions to describe what each sub‑figure demonstrates.

Response: Thank you for this constructive comment. We completely agree with the reviewer. In the revised manuscript, we have systematically expanded and re-written the captions for all composite figures (specifically Figures 1, 2, 3, 4, 5, 7, 8, and 9). Each sub-figure label (e.g., (a), (b), (c)) now includes a detailed explicit description explaining its specific morphological features, microstructural constituents, or analytical conditions.

The updated captions have been incorporated into the revised manuscript (marked in red) and are also listed below for your convenience:

 

6.Improve clarity and language

Remove informal or metaphorical phrasing (e.g., “steel dragons”).

Response: Thank you for your careful reading and constructive feedback. We completely agree that academic writing should maintain an objective, formal, and precise tone.

In the revised manuscript, we have thoroughly checked the text and removed all informal, literary, or metaphorical expressions, including the phrase “steel dragons” in Section 1 (Introduction). These expressions have been replaced with formal, academic terminology (e.g., "modern railway systems" or "locomotives"). The modified passages have been highlighted in red in the revised paper.

 

Standardize units (wt.% vs %).

Response: Thank you for your careful reading and helpful suggestion. We completely agree with the reviewer. In the revised manuscript, we have systematically checked and standardized all chemical composition and concentration units throughout the main text, figures, and tables (e.g., Table 1 and Table 4). All mass percentage references have now been consistently normalized to wt.%.

The revisions have been highlighted in red in the revised manuscript.

 

Clarify whether “Alt” in Table 1 refers to total or soluble aluminum.

Response: Thank you for bringing this to our attention. We appreciate the opportunity to clarify this detail.

We confirm that “Alt” in Table 1 refers to Total Aluminum (measured via spark optical emission spectrometry, capturing both dissolved aluminum in the matrix and aluminum bound in non-metallic oxide/aluminate inclusions).

To avoid any potential ambiguity for readers, we have updated Table 1 by replacing “Alt” with the explicit label “Total Al” (or defining “Alt” as Total Aluminum in the table footnote). Furthermore, a brief clarification has been added to Section 2.2 and Section 3.1 in the revised manuscript (highlighted in red).

 

Correct minor grammar and spelling issues throughout.

Response: Thank you for your meticulous reading and helpful suggestion. We completely agree that precise grammar and consistent spelling are critical for clear scientific communication.

The entire manuscript has undergone a comprehensive proofreading and language polishing pass. Minor typographical errors, spelling inconsistencies, punctuation issues, and tense/agreement flaws have been carefully corrected throughout the text, abstracts, figures, and tables. All linguistic corrections have been highlighted in red in the revised manuscript.

 

7.Expand fracture analysis

Add quantitative dimple size measurements.

Response: Thank you for this constructive recommendation. We completely agree that quantifying the dimple size provides a clearer, more rigorous physical basis for evaluating the micro-fracture mechanism and ductility differences.  In the revised manuscript, we performed quantitative image analysis on high-magnification SEM fractographs (Figure 9) using ImageJ software, measuring over 100 representative dimples per condition. The resulting mean equivalent dimple diameters and size ranges have been incorporated into Section 3.4 and the caption of Figure 9.  The updated response details and corresponding text revisions (highlighted in red) are provided below:

Quantitative statistical analysis of the high-magnification SEM fractographs (Figure 9) further confirms the microstructural transition. For the pre-renovation rails (1899–1904), the fracture surfaces display non-uniform, large dimples with an average equivalent diameter of  (ranging from  to ), accompanied by coarse cleavage steps. In contrast, the post-renovation rails (1909–1921) feature significantly refined and homogeneous ductile dimples with a mean diameter of  (ranging from  to ) alongside shallow quasi-cleavage facets, reflecting improved strain localization resistance during tensile deformation.

 

Provide clear comparison of cleavage vs ductile fracture across years.

Response: Thank you for your suggestion. The primary focus of this study is to elucidate the metallurgical process evolution (e.g., steelmaking refining, deoxidation practice, and non-metallic inclusion control) and the resulting microstructural features across different production eras, rather than detailed fracture mechanics.

To address the reviewer's comment while maintaining the paper's core theme, we have added a concise summary sentence in Section 3.4 to describe the overall trend of cleavage versus ductile features across the years (i.e., transition from pore-associated mixed failure in early rails to quasi-cleavage with localized dimples in later rails).

The concise addition is highlighted in red in Section 3.4 of the revised manuscript.

 

Add a short explanation of how strain aging may have influenced fracture morphology.

Response: Thank you for bringing up this valuable metallurgical aspect. We completely agree with the reviewer.

Early steel rails (particularly pre-renovation Bessemer samples from 1899–1904) possessed elevated free nitrogen and carbon concentrations alongside high phosphorus levels. Over extended service periods and historical aging, the diffusion of these interstitial atoms led to dislocation pinning (strain aging), which heightened matrix yield strength at the expense of local ductility. Microscopically, strain aging restricts dynamic plastic flow and microvoid growth during loading, thereby promoting transgranular cleavage facets, river patterns, and secondary microcracks while suppressing deep, fully developed ductile dimples.

A brief discussion explaining the strain aging influence has been integrated into Section 3.4 of the revised manuscript (highlighted in red).

Reviewer 2 Report

Comments and Suggestions for Authors

The authors of the article have accepted the reviewer’s comments, and the revisions have increased the article’s appeal for a technical audience. Despite my reservations, i.e. that this is not a scientific article, I recommend the manuscript for publication.

Author Response

The authors of the article have accepted the reviewer’s comments, and the revisions have increased the article’s appeal for a technical audience. Despite my reservations, i.e. that this is not a scientific article, I recommend the manuscript for publication.

Response: We sincerely thank the reviewer for the time and constructive feedback, which have significantly improved the manuscript. We are very pleased that the reviewer finds the revised version has increased its appeal for a technical audience and has recommended publication.

We greatly appreciate the reviewer’s candid assessment regarding the nature of the article. While our work is oriented more toward technical implementation and application rather than fundamental scientific discovery, we have made every effort to ensure it is grounded in rigorous methodology and scientific principles. The focus on addressing practical challenges for the technical community was a deliberate choice, and we are encouraged that the reviewer recognized its value in this context.

We hope the manuscript in its current form will serve as a useful contribution to the journal’s technically-minded readership. Once again, thank you for the guidance throughout the review process.