Effect of Internal Hydrogen on the Fatigue Crack Growth Rate in the Coarse-Grain Heat-Affected Zone of a CrMo Steel

Round 1

Reviewer 1 Report

L218: I cannot see any proof that the failure in BLD. Please reword as a “possible” failure or “assumed#2 failure

L222: to me the facets have a much higher grain size (>> 200 µm) than the PAGS (117 µm) so this statement needs more explanation

L235: this sentence is confusing to me because IG and BLD are not the same. Please explain in more detail and differentiate between IG and BLD

Table 2: what is MLD?

L256: the plural of index is indices

Author Response

First of all, we want to thank the work performed by all reviewers as they have helped us to improve our document for a better dissemination of our research results.

We have marked in red letter all corrections in the revised manuscript. Here we are now answering all reviewer´s questions (the number of the lines, references, images or tables are referring to the reviewed manuscript).

L128 (now L245-246): We have included new text to evidence the relationship between failure units in Fig. 5a) with microstructural units.

L222 (now L249-250): PAGS of the simulated CG-HAZ was not 117 mm, but 242±101 mm (see line 123) and this size corresponds quite well with grain sizes observed on the failure surfaces of the fatigued samples.

L235 (now L243-250): IG and BLD are not the same. IG is due to decohesion of prior austenite grain boundaries and BLD corresponds to decohesion of bainite lath interfaces.

Table 2: It was a mistake. It is BLD. Table was corrected.

L256: Another mistake. Text was corrected in the new version of our manuscript.

Reviewer 2 Report

The article highlights peculiarities of the effect of internal hydrogen on the fatigue crack growth rate of the coarse grain region of a 2.25Cr1Mo steel welded joint. The fatigue crack growth rate was measured under standard laboratory conditions using compact tensile (CT) specimens that were uncharged and hydrogen pre-charged in a hydrogen pressure reactor. The authors performed fatigue tests at different frequencies (f = 10, 0.1 and 0.05) and different asymmetries of the loading cycle (R = 0.1 and 0.5).

The article is interesting, but a number of shortcomings need to be corrected:

1. In Fig.1, a scale bar cannot be recognized.
2. The font size in legend in Fig.3 should be increased.
3. It is difficult to estimate hydrogen effect (Lines 196-197) using data given in Fig.3, since the data correspond to different frequencies of the loading cycle for hydrogen pre-charged (f=0.05 and f=0.1 Hz) and uncharged samples (f=10 Hz). According to Fig.3, it can only be confirmed that the effect of hydrogen is only noticeable at low frequencies [4,10]. Please explain why Fig. 3b contains the data obtained at the asymmetry of the loading cycle R = 0.1.
4. It is known that the effect of hydrogen is manifested at the threshold fatigue crack growth rates (https://doi.org/10.1023/A:1013262826275, https://doi.org/10.1007/s11003-021-00489-3, https://doi.org/10.1016/j.prostr.2020.06.052, https://doi.org/10.3390/met11111776), while the authors present data on fatigue crack growth rates obtained in the Paris region. Please comment on this fact.
5. In F7, the data obtained according to formula (1) are given (Lines 251-255) “In order to have a quantitative comparison of the effect of the different test variables on the fatigue crack growth behaviour, the hydrogen embrittlement index defined in expression (1), that compares the fatigue crack growth rate of uncharged and hydrogen precharged specimens, was used throughout the tests carried out under the different conditions. The obtained results are presented in Figure 7.”. According to formula (1), the authors determine the hydrogen embrittlement index (HEI), whereas in Fig. 7, HEF parameter is given. If it is the same parameter, the same abbreviation should be used.
6. To determine the hydrogen embrittlement index (HEI) using formula (1), the fatigue crack growth rate values measured with hydrogen pre-charged specimens (da/dN)_H and with non-charged ones (da/dN)_air are required. Please explain where the fatigue crack growth rate values measured with non-charged specimens (da/dN)_air at load frequencies f=0.05 and f=0.1 Hz come from.
7. More new References (2018-2022) should be added. For example, adding the following links will improve the quality of the Introduction (Line 32): “Also, the use of the current natural gas infrastructure to transport and storage hydrogen is being studied by different researches [https://doi.org/10.3390/pr9071219, https://doi.org/10.3221/IGF-ESIS.59.26]”.

Author Response

1: The scale bars were changed and they are now more visible in the lower part of the images.

2: The legend font size in Fig 3. has been increased.

3: As it is said in lines 169-171, hydrogen embrittlement in fatigue tests is only noticeable under low frequencies. Indeed, we have performed fatigue crack growth tests under 10 and 1 Hz with this base metal and also with other quenched and tempered steels and any hydrogen effect was detected. We may use low frequencies (0.1 Hz or below) to give time to hydrogen atoms to move and to attain the process zone to promote the embrittlement reaction. We have also now added that we have only performed tests without hydrogen under R = 0.1 (line 204).

4: We cannot determine the DK_th using hydrogen pre-charged specimens, because quite long tests would be required and most hydrogen would be lost on the course of the test. We have added this point in the revised manuscript (lines 173-176).

5: Thank you. It was a mistake. We are using now hydrogen embrittlement index (HEI) in all the revised manuscript.

6: Fatigue crack growth rates in air at room temperature are not affected by frequency. This point was added in the revised version of our manuscript (line 168-169).

7: Introduction was extended and more references added in the revised version of our manuscript

Reviewer 3 Report

Review report on the topic ‘Effect of internal hydrogen on the fatigue crack growth rate in the coarse-grain heat-affected zone of a CrMo steel’. Comments are listed below:

1. Strengthen the abstract section. Add the key conclusion of the works in the last two lines of the abstract section.
2. Discuss the novelty of the work in respect of the application.
3. There are numerous spelling and grammatical errors. Please revise the manuscript thoroughly. Sentences are also not complete and references are also cited in a rough manner.
4. Try to make a bridge between current and previously published work and specify the gap area and objective of the work. Discuss the major problem associated with diffusible hydrogen content, their role on mechanical performance and their measurement. Refer to some published work: https://doi.org/10.1016/j.ijhydene.2017.07.225; https://doi.org/10.1016/j.ijhydene.2016.07.202; https://doi.org/10.1080/09507116.2010.540828; https://doi.org/10.1115/1.4035764.
5. Provide the image of the experimental setup with good quality.
6. The SE image of base metal needs more discussion and, if possible, add the white particles' EDS spectra.
7. Discussion related to hydrogen diffusion is not clear in section 2.2. How was the diffusion measured?
8. Provide clear detail about the fatigue test experiment and also add the image of the specimen before and after a fracture.
9. The fracture surface needs more study and if possible, add the EDS spectra: https://doi.org/10.1016/j.ijhydene.2017.05.214; https://doi.org/10.1016/j.jmapro.2017.06.009.
10. The relation between diffusible hydrogen content and fatigue properties is not discussed properly. Strengthen the discussion section with more references.
11. Shorten the length of the conclusion section. Keep only key points.

The work is good, but the technical discussion and introduction section needs improvement. Paper can be accepted after following minor corrections.  

Author Response

1: We have included the main conclusions of the work in the last lines of the Abstract section.

2: The main novelty of the work is now included: definition of testing conditions necessary to limit hydrogen egress in the course of fatigue crack growth tests when hydrogen pre-charged specimens are used.

3: All document was revised.

4: We have enlarged the introduction section and more references were included in the revised version of our manuscript.

5: We have not included the experimental setup we have used because conventional fatigue crack propagation testing was used in accordance to the corresponding standard.

6: We have now better explained microstructures presented in Figure 1. This 2.25Cr1Mo steel is very well known. We have not used EDS but it is known these carbides are mixed iron-chromium carbides, as we have written now in the revised text.

7: The diffusion coefficient was previously measured, so we only want to mention here the general testing procedure. Details can be found on reference [32] and on the ASTM G148 standard, as it is said in the text. Anyway, new information was now included in the revised text (lines 138-140).

8: The dimensions of the CT specimens were calculated according to the standard, now it is mentioned in the text (line 159-160). There is nothing important to show after fracture, but failure mechanisms, which are described in figures 5, 6 and 7.

9: Fatigue failure surfaces are described in lines 211-218, 239-257 and 262-272. We have not performed EDS analysis in these surfaces. Only iron is usually detected in failure surfaces of this steel.

10: We have now related hydrogen contents on the simulated CG-HAZ microstructure with the values measured in the permeation test (0.45 ppm, Table 2). Two more references were also included in the discussion section.

11: Conclusions were shortened.

Round 2

Reviewer 2 Report

The authors took into account the comments of the reviewer and made appropriate corrections to the manuscript.