Impact of Corrosion in Simulated Body Fluid on Fatigue Characteristics of 3D-Printed Polylactic Acid-Coated AM60 Magnesium Alloys

Round 1

Reviewer 1 Report

This manuscript is well organized, details are senfitically discribed, I think this manuscript can be accepted. I have a few questions:

1. Is the thickness of the glue uniform? How to maitain a uniform thickness?

2. In Figure 4, it seems that after the corrosion test, some sample has a incomplete coating, what is the reason?

3. Why there are sphere like particles in the shrinkage cavity?

4. How to avoid gas cavities in the coating in the future?

Author Response

Reviewer #1

This manuscript is well organized, details are scientifically described, I think this manuscript can be accepted. I have a few questions:

Answer: The authors would like to thank the respected reviewer for his/her positive opinion. Then, we have tried our bests to addressed all comments in the revised manuscript.

1. Is the thickness of the glue uniform? How to maintain a uniform thickness?

Answer: The mentioned thickness was not exactly measured and controlled but try to use a uniform optically on the whole area. The quality control could be also seen from the FESEM images. However, since the concern of the respected reviewer is correct, the following sentences are added to the revised article (in two parts of the research method and the results), to address this comment,

It was optically tried to use a uniform thickness of glue in the whole areas in the gauge length of the sample (Figure 1). However, this issue was not measured and controlled and therefore, it could be a root cause of the uncertainty. Since different samples were used for testing as the repeatability of experiments, under the same condition, this issue could be ignored as a concern for affecting on the fatigue lifetime. The quality and the thickness of glue was also evaluated by the images of the field-emission scanning electron microscopy (FESEM) from the fracture surface of samples.

From the FESEM images in Figures 9-11, it could be understood that the glue thick-ness was between 50-150 microns. As mentioned before, it could be an uncertainty but by considering the repeatability of testing under the same condition, there is no concern about this variety. Moreover, this issue can be considered in the further investigations to find the effect of the glue thickness on the strength. 

2. In Figure 4, it seems that after the corrosion test, some sample has an incomplete coating, what is the reason?

Answer: When we used the 1X simulated body fluid (SBF), no corrosion occurred in the sample. However, in the 10X SBF, all the samples were degraded due to the high corrosion rate. There was no sample without coating in Figure 4, but in some cases, the coating layers were also removed from the specimen surface due to high corrosion rate. In order to address the mention comment, the following paragraph is added to the revised article,   

As it can be seen, in some cases, the coating layers removed from the sample surface, due to a high corrosion rate. Notably, all samples had complete coating layers on the gauge length. After the corrosion tests, in some specimens, a part of coatings and in some others, the whole coating layers detached from the surface. Therefore, the corrosion phenomenon could penetrate into the base metal and a high corrosion rate occurred due to the low strength of AM60 magnesium alloys under the SBF condition or other mediums [7-15].

3. Why there are sphere like particles in the shrinkage cavity?

Answer: Based on the EDS analysis (Figure 12), these particles are Mg, which was created due to the casting process. The following sentences are added to the revised manuscript to address this nice comment from the respected reviewer,

Besides, the shrinkage holes with spheres like particles inside the holes are also shown in Figures 11 and 12. The magnesium alloy is produced by the casting method, which is the cause of this type of shrinkage cavities. The EDS analysis demonstrated that these particles were Mg. Hu et al. [66] reported that the having no porosity in AM50 magnesium alloy was resulted from partial-solidification of the primary α-Mg phase due to proper casting parameters such as the cooling rate during solidification. Therefore, Mg particles could be created due to the casting process. Chen et al. [67] presented such droplets in AM60 magnesium alloys. Increasing the holding time during remelting, the intragranular liquid droplets coalesced forming some liquid pools, and migrated to the liquid matrix. These intragranular liquid droplets had two types including the chemical constitution segregation and entrapping during coalescence. When the magnesium alloy is reheated above the solidus, it is possible to have several intragranular liquid formed in the α-Mg matrix [67].      

4. How to avoid gas cavities in the coating in the future?

Answer: Thanks for the nice comment. The following sentences are added to the revised manuscript for the solutions that could eliminate gas cavities in 3D printing,

Such defects may occur during the filament fabrication process, through extrusion during 3D-printing, or with inappropriate speed and precision in the nozzle movement or temperature. Kuznetsov et al. [60] showed that the increase in the interlayer cohesion with enhancing the nozzle diameter can be explained this issue. Gonabadi et al. [61] mentioned that in 3D-printed parts, the filament gaps (voids) between deposited layers were due to the variation in the layer-to-layer adhesion quality and the shrinkage during cooling. Therefore, these gas cavities or voids could be eliminated using proper parameters during 3D-printing. Working on this issue to eliminate or reduce these defects could be considered for the future works.  

Author Response File: Author Response.pdf

Reviewer 2 Report

The current paper reports the investigation on the effect of 3D-printed polylactic acid coating layers on high cycle fatigue behaviors of AM60 magnesium alloys for additive-manufactured and degraded specimens in simulated body fluid. The major criticism of this paper is the lack of discussion without which it is just a technical paper not a research paper. There are also some issues regarding experimental details and reported data, as mentioned in the detailed comments. Based on my assessment, I suggest major revision of this paper.

The detail comments are as follows:

1.      The title must be revised. The title should be concise that reflect the content of the manuscript.

2.      The abstract must be re-written in a focused way. Refereeing of the previous work should be avoided in the abstract, like lines 10-12. In abstract, there is no need for sample/experimental details. It will contain the essence of the current work with major data/findings.

3.      The aim of the present work is missing, which usually comes as a last paragraph of the introduction section.

4.      Try to avoid group citation (e.g., line 80, 106) to make the claims more credible.

5.      Line 116: Specify the ‘quantometric’ test.

6.      It was not clear enough how the polymer coatings were attached to the samples with glue. More elaborate description is required. Was the coating applied all over the sample or only in the notch area?

7.      How it was ensured that the thickness of the glue is consistence in each cases?

8.      What standard was followed for the fatigue test samples dimensions? Please include that is the manuscript.

9.      The data regarding the adhesion test is missing.

10.  I am missing a comprehensive discussion of the reported results, without which it is just a technical paper not a research paper.

11.  It is highly recommended to include a dissuasion section, just before the conclusion section.

Minor editing of English language required

Author Response

Reviewer #2

The current paper reports the investigation on the effect of 3D-printed polylactic acid coating layers on high cycle fatigue behaviors of AM60 magnesium alloys for additive-manufactured and degraded specimens in simulated body fluid. The major criticism of this paper is the lack of discussion without which it is just a technical paper not a research paper. There are also some issues regarding experimental details and reported data, as mentioned in the detailed comments. Based on my assessment, I suggest major revision of this paper.

Answer: The authors would like to thank the respected reviewer for his/her positive opinion. Then, we have tried our bests to addressed all comments in the revised manuscript.

The detail comments are as follows:

1) The title must be revised. The title should be concise that reflect the content of the manuscript.

Answer: The article title is changed into the following one,

The corrosion impact of simulated body fluid on 3D-printed polylactic acid coated AM60 magnesium alloys through high-cycle fatigue regimes

2) The abstract must be re-written in a focused way. Refereeing of the previous work should be avoided in the abstract, like lines 10-12. In abstract, there is no need for sample/experimental details. It will contain the essence of the current work with major data/findings.

Answer: The abstract is rewritten based on the mentioned issues by the respected reviewer. The information about the previous work is eliminated and highlights of obtained results are mentioned more. The research method is also shortened.

3) The aim of the present work is missing, which usually comes as a last paragraph of the introduction section.

Answer: The mentioned paragraph is extended in the revised article, as follows,

Therefore, the innovation of this research is to compare the fatigue lifetime of PLA coated Mg after the corrosion in the simulated body fluid (SBF) with uncoated Mg samples. For this objective, after machining of the initial AM60 samples, the 3D-printed PLA coating layers were added with a glue on the gauge length of the specimen. Several samples were corroded in the SBF and then, fatigue testing was performed under bending loads. Finally, the fracture surface of samples was evaluated in a microscopic state to find the damage mechanisms. 

4) Try to avoid group citation (e.g., line 80, 106) to make the claims more credible.

Answer: The respected reviewer is generally correct. The first issue (line 80) is changed in the revised article by mentioning more details of references. However, the second one (line 106) is a conclusion on the literature review. Such that mentioning is required to show the novelty. Moreover, details of these references were mentioned in the introduction and in that paragraph, they just mentioned to have a conclusion on the literature review. To address this comment, the following changes are made,

For example, Uematsu et al. [20] improved the corrosion fatigue strength of magnesium alloy by multilayer diamond-like carbon coatings. AlamKhan et al. [21] compared the corrosion fatigue behavior of die-cast and shot-blasted AM60 magnesium alloy. Ishihara et al. [22] investigated the corrosion fatigue resistance of an electrolytically-plated magnesium alloy.  

Again, if the respected reviewer think they must be removed, we could eliminate them.

5) Line 116: Specify the ‘quantometric’ test.

Answer: The following descriptions are added to the revised article,

Then, the chemical composition of this alloy was determined using a quantometric test (based on a chemical analysis with the RMRC-WI-520-100-06 standard in Razi Metallur-gical Research Center, Tehran, Iran), where its results (the weight percentage of elements) are shown in Table 1.

6) It was not clear enough how the polymer coatings were attached to the samples with glue. More elaborate description is required. Was the coating applied all over the sample or only in the notch area?

Answer: The coating layers were attached to the gauge length (or the notch area) of the samples. For this part of the comment, the following paragraph is added to the revised article,

It was optically tried to use a uniform thickness of glue in the whole areas in the gauge length of the sample (Figure 1). However, this issue was not measured and con-trolled and therefore, it could be a root cause of the uncertainty. Since different samples were used for testing as the repeatability of experiments, under the same condition, this issue could be ignored as a concern for affecting on the fatigue lifetime. The quality and the thickness of glue was also evaluated by the images of the field-emission scanning electron microscopy (FESEM) from the fracture surface of samples.  

Then, for the corrosion tests, the other surfaces of a sample were covered with the bee wax to prevent the corrosion of the base metal, where there were no coating layers. In order to address this comment, the following sentences are added to the revised article,

In this image, as it can be seen, the coating layers were attached to the gauge length of the sample (after using a glue on the base metal). Then, the other surfaces (the dumb-bell-shaped parts) of the sample were also covered by the bee wax to prevent any corrosion of the base metal (where there was no coatings).    

Moreover, Figure 2 is also modified in the revised article.

7) How it was ensured that the thickness of the glue is consistence in each case?

Answer: This issue is correctly mentioned by the respected reviewer. The mentioned thickness was not exactly measured and controlled but try to use a uniform optically on the whole area. The quality control could be also seen from the FESEM images. However, since the concern of the respected reviewer is correct, the following sentences are added to the revised article (in two parts of the research method and the results), to address this comment,

It was optically tried to use a uniform thickness of glue in the whole areas in the gauge length of the sample (Figure 1). However, this issue was not measured and controlled and therefore, it could be a root cause of the uncertainty. Since different samples were used for testing as the repeatability of experiments, under the same condition, this issue could be ignored as a concern for affecting on the fatigue lifetime. The quality and the thickness of glue was also evaluated by the images of the field-emission scanning electron microscopy (FESEM) from the fracture surface of samples.

From the FESEM images in Figures 9-11, it could be understood that the glue thick-ness was between 50-150 microns. As mentioned before, it could be an uncertainty but by considering the repeatability of testing under the same condition, there is no concern about this variety. Moreover, this issue can be considered in the further investigations to find the effect of the glue thickness on the strength. 

8) What standard was followed for the fatigue test samples dimensions? Please include that is the manuscript.

Answer: The following sentences are added to the revised article,

The high-cycle rotary bending fatigue test (R=-1) was performed based on the ISO-1143 and DIN-EN-50113 standards [44,45]. Although they are only related to metals but also, in this research that the fatigue properties of magnesium alloys (with and with-out polymeric coatings) were studied, again, the mentioned standard was utilized.

9) The data regarding the adhesion test is missing.

Answer: The respected reviewer is correct. However, these data were published before in the previous work and just one data was reported here. To address this comment, the following sentences are added to the revised article.

The average value and the standard deviation of the obtained results, which shows the amount of adhesion between the base metal and the coating, was measured 4.29±0.71. In another research [39], the amount of adhesion between PLA and Mg, which were connected to each other by electrolytic plasma oxidation method, was also measured nearly. More details of these adhesion testing results could be found in the previous work [36].

10) I am missing a comprehensive discussion of the reported results, without which it is just a technical paper not a research paper.

Answer: The respected reviewer is correct and therefore, we extended the discussion on the obtained results, by comparing them with other results of other articles, in different parts of the revised article, as follows,

The average value and the standard deviation of the obtained results, which shows the amount of adhesion between the base metal and the coating, was measured 4.29±0.71. In another research [39], the amount of adhesion between PLA and Mg, which were connected to each other by electrolytic plasma oxidation method, was also measured nearly. More details of these adhesion testing results could be found in the previous work [36].

As it can be seen, in some cases, the coating layers removed from the sample surface, due to a high corrosion rate. Notably, all samples had complete coating layers on the gauge length. After the corrosion tests, in some specimens, a part of coatings and in some others, the whole coating layers detached from the surface. Therefore, the corrosion phenomenon could penetrate into the base metal and a high corrosion rate occurred due to the low strength of AM60 magnesium alloys under the SBF condition or other mediums [7-15].

This is due to water absorption by the polymer coating [46]. Hasanpour et al. [47] investigated water absorption and corrosion of pure magnesium and magnesium with PLA. The specimens were immersed in the SBF solution for 30 days. It was concluded that the corrosion rate in samples with PLA is increased due to more water absorption. Balogova et al. [48] for PLA samples showed that the mass of samples increases with increasing water absorption. Redondo et al. [49] also reported water absorption in corrosion tests for PLA samples. Alksne et al. [50] reported swelling and water absorption over time for Compo-site samples made of PLA+hydroxyapatite (HA) and PLA+bioglass (BG).

Chor et al. [51] also reported a decrease in the corrosion rate for a sample made of PLA materials over the time. Voicu et al. [52] combined MgZ31 with PLA nanofibers and the corrosion of samples in SBF solution was investigated. The results illustrated that the rate of corrosion decreased with the help of PLA coating. Shi et al. [53] fabricated a PLA layer on the sample by placing AZ31 in the PLA-chloroform solution. The corrosion results demonstrated that the coating layers decreased the corrosion rate.

Based on the experimental data in Figure 8, several comparisons of obtained results could be done with the literature for the root cause of fatigue failure. Under CF conditions, Nan et al. [55] presented that the 70-80% duration of the fatigue lifetime was occupied due to the corrosion crack initiation. In other words, the corrosion fatigue lifetime under lower stresses was determined by the growth behavior of corrosion pits [55]. As a claim by Bhuiyan et al. [56], this ratio of the pit growth lifetime to the fatigue crack nucleation and total fatigue lifetime was almost 30%. Notably, this crack initiation lifetime ratio must be considered with the effect of the ap-plied stress. They claimed that most of the initiation lifetime was spent on the corrosion pit generation [56]. It must be noted that these issues may occur when fatigue testing was done inside the corrosive medium. However, in this study, fatigue tests were performed on the pre-corroded specimens. When emerging the samples in the SBF, the produced defects and pits on the surface of the specimens will act as the stress concentration that boost the crack initiation stage [57]. More transgranular or intergranular cracks can propagate un-der the CF condition [58].

Such defects may occur during the filament fabrication process, through extrusion during 3D-printing, or with inappropriate speed and precision in the nozzle movement or temperature. Kuznetsov et al. [60] showed that the increase in the interlayer cohesion with enhancing the nozzle diameter can be explained this issue. Gonabadi et al. [61] mentioned that in 3D-printed parts, the filament gaps (voids) between deposited layers were due to the variation in the layer-to-layer adhesion quality and the shrinkage during cooling. Therefore, these gas cavities or voids could be eliminated using proper parameters during 3D-printing. Working on this issue to eliminate or reduce these defects could be considered for the future works.

Besides, the shrinkage holes with spheres like particles inside the holes are also shown in Figures 11 and 12. The magnesium alloy is produced by the casting method, which is the cause of this type of shrinkage cavities. The EDS analysis demonstrated that these particles were Mg. Hu et al. [66] reported that the having no porosity in AM50 magnesium alloy was resulted from partial-solidification of the primary α-Mg phase due to proper casting parameters such as the cooling rate during solidification. Therefore, Mg particles could be created due to the casting process. Chen et al. [67] presented such droplets in AM60 magnesium alloys. Increasing the holding time during remelting, the intragranular liquid droplets coalesced forming some liquid pools, and migrated to the liquid matrix. These intragranular liquid droplets had two types including the chemical constitution segregation and entrapping during coalescence. When the magnesium alloy is reheated above the solidus, it is possible to have several intragranular liquid formed in the α-Mg matrix [67].

From the FESEM images in Figures 9-11, it could be understood that the glue thick-ness was between 50-150 microns. As mentioned before, it could be an uncertainty but by considering the repeatability of testing under the same condition, there is no concern about this variety. Moreover, this issue can be considered in the further investigations to find the effect of the glue thickness on the strength.

Such this improvement of the corrosion resistance with coatings was reported by Kumar [11], who used hydroxyapatite (HA) on different magnesium alloys. In another work, Uematsu et al. [20] used multilayer diamond-like carbon coatings to increase the corrosion strength of magnesium alloy under fatigue. Although these materials and their processes were different from this research (considering PLA with 3D-printing), but coatings on magnesium alloys have similar results in various applications, especially for biomedical engineering. 

If the respected reviewer think we should do more, we eagerly do it in the next stage.

11) It is highly recommended to include a dissuasion section, just before the conclusion section.

Answer: The mentioned comment is so nice. However, we changed the name of Part 3 from "Results" into "Results and Discussion". Then also, the obtained results in each part were compared to other results of other articles. We wish to address the concern of the respected reviewer. If it is not okay, we could separate them in two separated parts.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

All the comments were addressed accordingly and the paper may be accepted in current form.

Minor editing of English language required

Author Response

Thank you for your positive response. Then, the whole manuscript is again read and checked for any mistakes in English writing. Moreover, the text is checked by the software of Grammarly with a score of 86.