Effect of Hydrogen and Hydrogen-Blended Natural Gas on Additive-Manufactured 316L Stainless Steel in Ambient Oil and Gas Environments

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript investigates the effects of hydrogen and hydrogen-blended natural gas on 316L stainless steel produced via Additive Manufacturing (AM), compared to Conventional Manufacturing (CM). The study is motivated by the growing interest in utilizing existing natural gas pipeline infrastructure for hydrogen transport to reduce greenhouse gas emissions. However, concerns over hydrogen embrittlement (HE) remain a major challenge, especially for new materials and manufacturing techniques such as metal 3D printing.

This research focuses on evaluating the mechanical behavior and hydrogen susceptibility of AM and CM 316L exposed to 10 MPa pressure environments with 0%, 50%, and 100% hydrogen concentrations for 5 and 24 weeks. Post-exposure analyses included tensile testing, SEM fractography, hydrogen quantification via hot chemical extraction, and microstructural characterization.

The results show that CM 316L displays consistent resistance to hydrogen exposure across all conditions, with no significant changes in mechanical properties or evidence of embrittlement. In contrast, AM 316L exhibited a ~20% reduction in ductility and a ~15% increase in yield strength when exposed to 100% hydrogen, indicative of HE. This degradation is attributed to microvoids inherent to the AM process, which act as hydrogen traps and promote failure through mechanisms such as Hydrogen-Enhanced Localized Plasticity (HELP).

Interestingly, even though CM samples consistently showed higher hydrogen content than AM samples—due to the presence of MnS inclusions—only the AM samples exhibited mechanical degradation. The authors highlight that these microstructural differences and the nature of hydrogen trapping sites significantly influence embrittlement behavior.

While the study presents meaningful insights, several areas could be improved:

1. Expand the range of testing conditions: All tests were performed at room temperature and 10 MPa. Real-world applications, particularly in oil and gas, often involve higher temperatures and fluctuating conditions. Future studies should broaden the scope to simulate these more complex environments. Refer to Section 1, lines 62–66.
2. Better phase characterization is needed: Although the paper mentions austenite and martensite, no quantitative phase analysis (e.g., XRD, EBSD) is presented. Since transformation-induced plasticity and phase evolution are central to HE, their detailed characterization would strengthen the conclusions. Refer to Section 3.3, lines 209–211.
3. Include post-processing treatments: AM samples often contain porosity and residual stresses that can be mitigated through Hot Isostatic Pressing (HIP) or similar methods. The authors should include or propose such treatments, especially since porosity is identified as a key weakness. Refer to the Conclusions, lines 331–334.
4. Quantify microvoids via XRM: Although X-ray microscopy (XRM) revealed the presence of microvoids, no statistical analysis (e.g., average size, distribution, volume fraction) was provided. Quantitative defect metrics would allow better correlation with mechanical degradation.Refer to Figure 12 and lines 302–307.
5. Explore different AM build orientations:The study only tested vertically printed AM samples, which are known to have the best mechanical properties. Evaluating other build orientations (horizontal, angled) is essential to reflect the behavior of complex, real-world parts.Refer to Section 2, line 96.

 

Suggested Improvement to the Literature Review

To enhance the scientific background, the authors are encouraged to broaden their literature review by incorporating insights on environmental gas distribution and sampling strategies, which are relevant for simulating hydrogen exposure in real-world settings. A particularly useful reference is:Optimization of number and location of sampling points of an air quality monitoring network in an urban context. DOI: 10.3303/CET1974047

Author Response

1. Expand the range of testing conditions: All tests were performed at room temperature and 10 MPa. Real-world applications, particularly in oil and gas, often involve higher temperatures and fluctuating conditions. Future studies should broaden the scope to simulate these more complex environments. Refer to Section 1, lines 62–66. 1. Thank you for pointing this out. We agree with this comment. The samples were exposed to fluctuating temperatures in the soaking process, but the testing (tensile and chemical) was done at room temperature. We do plan on looking at more in-situ testing and accelerated testing in extreme temperatures and pressures and have made a comment in the conclusion in line 341 to indicate future research 2. Better phase characterization is needed: Although the paper mentions austenite and martensite, no quantitative phase analysis (e.g., XRD, EBSD) is presented. Since transformation-induced plasticity and phase evolution are central to HE, their detailed characterization would strengthen the conclusions. Refer to Section 3.3, lines 209–211. 1. Thank you for pointing this out. We agree with this comment. Therefore we have added EBSD to the paper in Figure 13 and lines 318, 319, and 323 and 324   3. Include post-processing treatments: AM samples often contain porosity and residual stresses that can be mitigated through Hot Isostatic Pressing (HIP) or similar methods. The authors should include or propose such treatments, especially since porosity is identified as a key weakness. Refer to the Conclusions, lines 331–334. 1. Thank you for pointing this out. We agree with this comment and have made this more clear in our conclusion as a way to prevent HE in 339 and 340. Post-processing stress relief was carried out as per standard operating procedures at Emerson in line 103 and made more clear.  4. Quantify microvoids via XRM: Although X-ray microscopy (XRM) revealed the presence of microvoids, no statistical analysis (e.g., average size, distribution, volume fraction) was provided. Quantitative defect metrics would allow better correlation with mechanical degradation.Refer to Figure 12 and lines 302–307. 1. Thank you for pointing this out. We agree with this comment and have added some statistical analysis from the XRM of the volume fraction in lines 308 -3010 to more clearly highlight this in volume fraction.  5. Explore different AM build orientations: The study only tested vertically printed AM samples, which are known to have the best mechanical properties. Evaluating other build orientations (horizontal, angled) is essential to reflect the behavior of complex, real-world parts. Refer to Section 2, line 96. 1. Thank you for pointing this out. We agree with this comment as this orientation was chosen for the reason pointed out and we believe would show the effects of HE most clear. We have made some comments in the conclusion in line 343 and 345 address this point.    Suggested Improvement to the Literature Review To enhance the scientific background, the authors are encouraged to broaden their literature review by incorporating insights on environmental gas distribution and sampling strategies, which are relevant for simulating hydrogen exposure in real-world settings. A particularly useful reference is: Optimization of number and location of sampling points of an air quality monitoring network in an urban context. DOI: 10.3303/CET1974047
2. Thank you for this suggestion. We have added this reference to into the intro and some more background from lines 71-75. 

 

Reviewer 2 Report

Comments and Suggestions for Authors

The author performed a sound and viable experiment to define the influence of hydrogen on the mechanical properties of the 316 L stainless steel.
However, some issues should be considered as follows:
1. Line 72: "MPa with ambient temperatures MPa [27]". Please reconsider the phrase (temperatures in MPa?)
2. Figure 4: Please change Mpa /MPa/
3. Please change H2 H₂ in text and tables!
4. Please change CH4 CH₄ in text!
5. Please specify the temperature of resting for 72 hours, this is important for the repeatability of the experiment!
Results and discussion are appropriately arranged together with defined conclusions.

Author Response

1. Line 72: "MPa with ambient temperatures MPa [27]". Please reconsider the phrase (temperatures in MPa?)
Thank you for pointing out this error. We have corrected the error and rephrased lines 71 and 72 to be more clear and give the ambient temperature range
2. Figure 4: Please change Mpa /MPa/
Thank you for pointing out this error. Figure 4 has been corrected to MPa in the paper
3. Please change H2 H2 in text and tables!
Thank you for pointing out this error, this has been corrected throughout.
4. Please change CH4 CH4 in text!
Thank you for pointing out this error. This has been fixed on line 277
5. Please specify the temperature of resting for 72 hours, this is important for the repeatability of the experiment!
Thank you for pointing this out. This has been addressed saying in line 134 “…allow all reversable hydrogen to degas in the lab at room temperature”

Reviewer 3 Report

Comments and Suggestions for Authors

Recommendations and comments:

Line 72: The manuscript incorrectly states, "with ambient temperatures MPa [27]. "

Line 94: Is the 928 mm/s correct parameter?

Chapter 3.1: It is clear from the tensile diagram that not only the yield strength and ductility change, but also the ultimate tensile strength of the analysed and hydrogenated states changed.

Figure 4: The description of the tension diagrams is not correct. (What is the state in Fig. 4a and what is in Fig. 4b?

Figure 5: The microstructures of the analysed states are not properly prepared or do not have adequate quality.

The fact that it is not clear what mechanical properties the analysed states have, it is not possible to take a position on the other results found and discussed.

Author Response

1. Line 72: The manuscript incorrectly states, "with ambient temperatures MPa [27]. "
    Thank you for pointing out this error. We have corrected the error and rephrased lines 71 and  72 to be clearer and give the ambient temperature range
2. Line 94: Is the 928 mm/s correct parameter?
    Thank you for pointing out this error. It has been corrected to mm/s2 for the velocity of the beam in line 97.
3. Chapter 3.1: It is clear from the tensile diagram that not only the yield strength and ductility change, but also the ultimate tensile strength of the analysed and hydrogenated states changed.
    Thank you for pointing this out. This result has also been pointed out in section 3.1 but as it is not traditionally discussed when talking about hydrogen embrittlement, was left out for this discussion. 
4. Figure 4: The description of the tension diagrams is not correct. (What is the state in Fig. 4a and what is in Fig. 4b?
    Thank you for pointing out this oversight. The a and b figures have been swapped to match the description in line 161
5. Figure 5: The microstructures of the analysed states are not properly prepared or do not have adequate quality.
    Thank you for pointing this out. Figure 5 has been updated with a clearer quality picture of the CM sample more comparable to the AM sample. 
6. The fact that it is not clear what mechanical properties the analysed states have, it is not possible to take a position on the other results found and discussed.
    We appreciate this comment. We have added EBSD to help quantify the phases analyzed in the results section. 

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

I thank the authors for the corrections they made.

Unfortunately, the microstructures of the analyzed steel states in Fig. 5, Fig. 10 and Fig. 11 are still unclear, of insufficient quality and contain artifacts. In metallographic preparation of steel samples, it is necessary to use the correct grinding, polishing and etching procedures and suitable metallographic preparations and etchant.

I draw the authors' attention to the microstructures of steels of this type presented in the articles:

1. Microstructure and corrosion behavior of 316L stainless steel prepared using different additive manufacturing methods: A comparative study bringing insights into the impact of microstructure on their passivity, Corrosion Science, Volume 176, November 2020, 108914
2. Microstructure and mechanical properties of the austenitic stainless steel 316L fabricated by gas metal arc additive manufacturing, Materials Science and Engineering: A, Volume 703, 4 August 2017, Pages 567-577

Author Response

Unfortunately, the microstructures of the analyzed steel states in Fig. 5, Fig. 10 and Fig. 11 are still unclear, of insufficient quality and contain artifacts. In metallographic preparation of steel samples, it is necessary to use the correct grinding, polishing and etching procedures and suitable metallographic preparations and etchant.

Thank you for pointing this out to us. While we agree the figures could be improved slightly, we believe the figures support the conclusions made from the work and cannot be replaced entirely in a timely manner. We have updated Fig 5. with new etched images to be clearer than before. Fig 10 has had the images post-processing improved to make the different phases and grain boundaries clearer. Fig 11 has had some artifacts cropped out, better post processing to highlight the defects (nonmetallic inclusions and voids) and increased scale bar size to enhance the figure.