Self-Healing Coatings Consisting of an Outer Electrodeposited Epoxy Resin Layer and an Inner Porous Anodic Oxide Layer with Healing Agents for the Corrosion Protection of Al Alloys
Round 1
Reviewer 1 Report
In regard to citations, some are not very much relevant (as such 1, 3, 13-15, 25 and 26), whereas many other notable citations that describe the advancements of anodized layer treatment are missing.
Nevertheless, the greatest weaknesses of the article are a) a very small number of methods used (only SEM and EIS) and tests made; b) lack of data analysis.
In particular, the cross-section view of the coating before and after scratching (perpendicular to the scratch) would be very useful and are strongly recommended. It is also advised to increase the contrast of the existing SEM images, as they appear quite flat in the printer version. If the samples of the various scratched depth were prepared, it would greatly enrich the paper.
Electrochemical Impedance Spectroscopy was used as a major method. Foremost, Nyquist impedance plots, in my mind, are not much representative because of a wide impedance range of the EIS data, and I think that Bode phase plots alone would be sufficient. If decided to be shown, Nyquist plots must be plotted with isotropic scales, otherwise EIS data is distorted. I would also advice to avoid kOhm·cm2 and use Ohm·cm2, as 1·106 kOhm, which is 1 GOhm·cm2, complicates interpretation of the magnitude for a reader, especially considering the logarithmic scale of Bode impedance plots. I would also suggest to plot Bode phase angle plot between 0 and -90°, as the values shall not be below -90°. These are merely recommendations, but on the other hand there are several more serious issues. Figure 5, high-to-medium frequency impedance Nyquist plots have invisible but printable lines.
The major issue is the choice of experimental parameters for EIS. First of all, I really doubt that all the samples would have close open circuit potentials (OCP) of -550 mV (± 50 mV). Thus, using identical DC voltage for all the samples would at least drive some samples of their steady state condition, and at most cause corrosion due to substantial polarization. Constant DC vs RE can be only justified for the samples of very high impedance, where there are no stable values of OCP. Thus, I would like to ask to provide an OCP curve for all the samples to justify such choice. Secondly, the voltage perturbation of 50 mV should be justified only for the samples without faradaic processes, whereas if corrosion initiated, 50 mV would be too high for most of the cases to keep linear response. Hence, I would like to ask to provide Lissajous plots for all the samples at any of low frequency (between 50 mHz and 10 mHz).
Another issue with all EIS data plots is that they do not show complete data: though in the Experimental part it is mentioned that the EIS was made between 100 kHz and 10 mHz, the authors decided to show only a part of the spectra that they thought would not cause doubts. Such decision can be understood taking into consideration EIS data for anodized Al with scratched organic coating is complex, but the cut away data at low frequencies gives the evidence and quantification of corrosion activity, if present. The data at high frequencies was discarded for some reason. In the case authors prefer to fit only part of the EIS spectra, using a simplified equivalent electrical circuit of the Figure 6, a completely valid approach, it is strongly recommended to show the fitting curve (within selected range of frequencies) and the actual experimental data in the whole tested range. In such case, the reader would a) clearly see how close the fitting curve to real data is; and b) have no doubts in regard the evolution of the EIS (and the state of the sample) at low frequencies. Having fitting curve in selected region would also be helpful for normal and self-healing coating with freezing on Figure 9, where spectra are too complex to be presented by the Figure 6.
I would like also to point out the necessity to explain the difference in impedance modulus values at high frequencies between the samples of the Figure 5 and 9. On the one hand, it can be attributed to different depth of the scratch: deeper scratch for the reference sample, and shallower scratch for self-healing coating, providing false-positive results for the latter specimen. On the other hand, it can be due to the fact that Al-oxide layer can become filled with aqueous solution for the reference sample, while being blocked with IPDI and its hydration product for the self-healing coating, having equal depth of the scratch for all the samples, actually proving the better corrosion protection of the latter coating. Hence, clear evidence of the scratch depths is necessary.
Comments for author File: 
Comments.pdf
Generally, the article is well-written, yet there are many mistypes scattered throughout the manuscript. Example 1 is Line 159: “In a previous study …” followed by multiple citations should be “In previous studies …”; example 2 is line 272: “micro-cocks” instead of “micro-cracks”; etc. Besides, in several places there are multiple spaces instead of single ones. Concerning the abbreviations, one should better avoid such if they are used only once in the entire article (e.g. SMP). However, the biggest criticism is caused by Figures numbering that does not correspond the text of the manuscript: they are either missing (Figure 10, line 239; Figure 14-b, line 262) or misplaced or referred incorrectly (line 243, etc.)
Author Response
Reviewer 1
Thank you very much for your kind consideration and treatments in your review with careful reading. The authors read the comments and questions of the reviewer, and revised the manuscript, according to them. The replies to them are as follows.
[Comment]
1) In regard to citations, some are not very much relevant (as such 1, 3, 13-15, 25 and 26), whereas many other notable citations that describe the advancements of anodized layer treatment are missing.
[Reply]
Thank you for your comment. 4 references (Ref. 24, 26, 27 and 28) on anodized of aluminum are added.
[Comment]
2) Nevertheless, the greatest weaknesses of the article are a) a very small number of methods used (only SEM and EIS) and tests made; b) lack of data analysis.
[Reply]
Thank you for your comment. As you suggested, this study may not be perfect yet. On the other hand, this paper is a large challenging study, and we consider that it is a very useful achievement, because it clearly shows that this coating has self-healing properties and maintains high corrosion protection after physical damaging of this coating.
[Comment]
3) In particular, the cross-section view of the coating before and after scratching (perpendicular to the scratch) would be very useful and are strongly recommended. It is also advised to increase the contrast of the existing SEM images, as they appear quite flat in the printer version. If the samples of the various scratched depth were prepared, it would greatly enrich the paper.
[Reply]
Thank you for your comment. Cross-sectional images of specimens damaged by 50˚ and 120˚ tip-angle indenter is added as Fig. A.
[Comment]
4) Electrochemical Impedance Spectroscopy was used as a major method. Foremost, Nyquist impedance plots, in my mind, are not much representative because of a wide impedance range of the EIS data, and I think that Bode phase plots alone would be sufficient. If decided to be shown, Nyquist plots must be plotted with isotropic scales, otherwise EIS data is distorted.
I would also advice to avoid kOhm·cm2 and use Ohm·cm2, as 1·106 kOhm, which is 1 GOhm·cm2, complicates interpretation of the magnitude for a reader, especially considering the logarithmic scale of Bode impedance plots. I would also suggest to plot Bode phase angle plot between 0 and -90°, as the values shall not be below -90°. These are merely recommendations, but on the other hand there are several more serious issues. Figure 5, high-to-medium frequency impedance Nyquist plots have invisible but printable lines. I would also suggest to plot Bode phase angle plot between 0 and -90°, as the values shall not be below -90°.
[Reply]
Thank you for your comment. As you suggest, these figures are revised.
[Comment]
5) The major issue is the choice of experimental parameters for EIS. First of all, I really doubt that all the samples would have close open circuit potentials (OCP) of -550 mV (± 50 mV). Thus, using identical DC voltage for all the samples would at least drive some samples of their steady state condition, and at most cause corrosion due to substantial polarization. Constant DC vs RE can be only justified for the samples of very high impedance, where there are no stable values of OCP. Thus, I would like to ask to provide an OCP curve for all the samples to justify such choice. Secondly, the voltage perturbation of 50 mV should be justified only for the samples without faradaic processes, whereas if corrosion initiated, 50 mV would be too high for most of the cases to keep linear response. Hence, I would like to ask to provide Lissajous plots for all the samples at any of low frequency (between 50 mHz and 10 mHz).
[Reply]
Thank you for your comment. The impedance values of all the specimens discussed in this paper, are bigger than 10-6 ohms, as shown in Fig. 9. Therefore, the current density is very small when an amplitude of 50 mV is applied, and little change is expected due to this.
[Comment]
6) Another issue with all EIS data plots is that they do not show complete data: though in the Experimental part it is mentioned that the EIS was made between 100 kHz and 10 mHz, the authors decided to show only a part of the spectra that they thought would not cause doubts. Such decision can be understood taking into consideration EIS data for anodized Al with scratched organic coating is complex, but the cut away data at low frequencies gives the evidence and quantification of corrosion activity, if present. The data at high frequencies was discarded for some reason. In the case authors prefer to fit only part of the EIS spectra, using a simplified equivalent electrical circuit of the Figure 6, a completely valid approach, it is strongly recommended to show the fitting curve (within selected range of frequencies) and the actual experimental data in the whole tested range. In such case, the reader would a) clearly see how close the fitting curve to real data is; and b) have no doubts in regard the evolution of the EIS (and the state of the sample) at low frequencies. Having fitting curve in selected region would also be helpful for normal and self-healing coating with freezing on Figure 9, where spectra are too complex to be presented by the Figure 6.
[Reply]
Thank you for your comment. As you suggest, these figures are revised.
7) I would like also to point out the necessity to explain the difference in impedance modulus values at high frequencies between the samples of the Figure 5 and 9. On the one hand, it can be attributed to different depth of the scratch: deeper scratch for the reference sample, and shallower scratch for self-healing coating, providing false- positive results for the latter specimen. On the other hand, it can be due to the fact that Al-oxide layer can become filled with aqueous solution for the reference sample, while being blocked with IPDI and its hydration product for the self-healing coating, having equal depth of the scratch for all the samples, actually proving the better corrosion protection of the latter coating. Hence, clear evidence of the scratch depths is necessary.
[Reply]
Thank you for your comment. Cross-sectional images of specimens damaged by 50˚ and 120˚ tip-angle indenters are added as Fig. A.
8) Comments on the Quality of English Language
Generally, the article is well-written, yet there are many mistypes scattered throughout the manuscript. Example 1 is Line 159: “In a previous study ...” followed by multiple citations should be “In previous studies ...”; example 2 is line 272: “micro-cocks” instead of “micro-cracks”; etc. Besides, in several places there are multiple spaces instead of single ones. Concerning the abbreviations, one should better avoid such if they are used only once in the entire article (e.g. SMP). However, the biggest criticism is caused by Figures numbering that does not correspond the text of the manuscript: they are either missing (Figure 10, line 239; Figure 14-b, line 262) or misplaced or referred incorrectly (line 243, etc.)
[Reply]
Thank you for your comment. The authors revise “In a previous study,“ in line 167 to “in previous studies,” and “micro-crocks” in line 312 to “micro-cracks”. And these figures have been reorganized and renumbered.
9) In particular, the cross-section view of the coating before and after scratching (perpendicular to the scratch) would be very useful and are strongly recommended. It is also advised to increase the contrast of the existing SEM images, as they appear quite flat in the printer version. If the samples of the various scratched depth were prepared, it would greatly enrich the paper.
[Reply]
Thank you for your comment. Cross-sectional images of specimens damaged by 50˚ and 120˚ tip-angle indenters are added as Fig. A. The contrast and brightness of Figs. 4 and 6 are adjusted.
10) Electrochemical Impedance Spectroscopy was used as a major method. Foremost, Nyquist impedance plots, in my mind, are not much representative because of a wide impedance range of the EIS data, and I think that Bode phase plots alone would be sufficient. If decided to be shown, Nyquist plots must be plotted with isotropic scales, otherwise EIS data is distorted. I would also advice to avoid kOhm·cm2 and use Ohm·cm2, as 1·106 kOhm, which is 1 GOhm·cm2, complicates interpretation of the magnitude for a reader, especially considering the logarithmic scale of Bode impedance plots. I would also suggest to plot Bode phase angle plot between 0 and -90°, as the values shall not be below -90°. These are merely recommendations, but on the other hand there are several more serious issues. Figure 5, high-to-medium frequency impedance Nyquist plots have invisible but printable lines.
[Reply]
Thank you for your comment. Nyquist plots in Figs.5 and 7 were deleted. And bode plots are revised, as you suggest.
Reviewer 2 Report
The manuscript is devoted to the self-healing coating, which consists of an outer electrodeposited epoxy resin layer and an inner anodic oxide layer with healing agents. The authors studied effects of the tip-angle of indenters and freezing treatments on self-healing properties for corrosion protection of Al alloy.
The topic is valuable for technical and industrial practice like automobile body components, so the paper may be of interest to scientific community. I believe the work is worth publishing, but the manuscript needs major revision.
My specific comments and suggestions in the order in which they appear in the text of the manuscript:
1) line 81
The correct name of the EIS technique is:
Electrochemical Impedance Spectroscopy, not Spectrometry
2) line 82
The title of section 2 (Experimental) would be better called Materials and methods, so familiar to readers of the journal Corrosion and Materials Degradation.
3) line 133
The phrase "Potentials of 50 mV to a rest potential of -550 mV (vs. R.E.) were applied" is incorrect. In the description of measurements using the EIS technique, it is commonly accepted that the a.c. amplitude of the perturbation signal is given (50 mV), not potential of 50 mV.
4) line 239
"are summarized in Fig. 10"
There is no Figure 10 in the manuscript.
5) section 4. Discussion
This section should be corrected and rewritten. The reason is errors in the description of test results and incorrect references to figures, e.g.:
- in Fig.6 there is no part a), and in line 259 there is a reference to Fig.6a.,
- in the text of the section there are references to Fig.10, Fig.11, Fig.14 - and in the manuscript there are only 9 figures.
Minor editing of English language required.
Author Response
Reviewer 2
Thank you very much for your kind consideration and treatments in your review with careful reading. The authors read the comments and questions of the reviewer, and revised the manuscript, according to them. The replies to them are as follows.
[Comment]
1) line 81
The correct name of the EIS technique is: Electrochemical Impedance Spectroscopy, not Spectrometry
[Reply]
Thank you for your comment. The authors revised “electrochemical impedance spectrometry“ to “electrochemical impedance spectroscopy” in line 82.
[Comment]
2) line 82
The title of section 2 (Experimental) would be better called Materials and methods, so familiar to readers of the journal Corrosion and Materials Degradation.
[Reply]
Thank you for your comment. The authors revised “Experimental“to “Materials and methods” in line 85.
[Comment]
3) line 133
The phrase "Potentials of 50 mV to a rest potential of -550 mV (vs. R.E.) were applied" is incorrect. In the description of measurements using the EIS technique, it is commonly accepted that the a.c. amplitude of the perturbation signal is given (50 mV), not potential of 50 mV.
[Reply]
Thank you for your comment. The authors revised the sentence to “A potential amplitude of 50 mV was applied to -550 mV (vs. R. E.) in the range of frequency between 10-1 and 5 x 104 Hz on all the specimens. The value of -550 mV (vs. R. E.) is the rest potential of electropolished specimen.” in line 137.
[Comment]
4) line 239"are summarized in Fig. 10"
There is no Figure 10 in the manuscript.
[Reply]
Thank you for your comment. This is a mistake. This part was deleted.
[Comment]
5) section 4. Discussion
This section should be corrected and rewritten. The reason is errors in the description of test results and incorrect references to figures, e.g.:
- in Fig.6 there is no part a), and in line 259 there is a reference to Fig.6a.,
- in the text of the section there are references to Fig.10, Fig.11, Fig.14 - and in the manuscript there are only 9 figures.
[Reply]
Thank you for your comment. Discussion section was corrected and rewritten.
Reviewer 3 Report
The Authors have presented fresh and important results on the corrosion and scratch resistance of their novel self-healing type surface treatment procedure. The paper's length is just fine and it covers all the necessary data for reaching their conclusions.
The topic as well is an important one as being related to aluminium car body systems.
I would propose chacking the text/misprints in the following lines: 82, 132, 157, 191, 192.
Author Response
Reviewer 3
Thank you very much for your kind consideration and treatments in your review with careful reading. The authors read the comments and questions of the reviewer, and revised the manuscript, according to them. The replies to them are as follows.
[Comment]
I would propose checking the text/misprints in the following lines: 82, 132, 157, 191, 192.
[Reply]
Thank you for your comment. The authors revise “electrochemical impedance spectrometry“ in line 82 to “electrochemical impedance spectroscopy”, “Ag/sat-AgCl. electrodes” in line 136 to “Ag/ AgCl /saturated KCl electrode electrodes”, “Cu2+/Cl- solution” in line 165 to “Cu2+/Cl- solution”, “” in line 199 and 201 to “Dq”.
Author Response File: Author Response.pdf
Reviewer 4 Report
There are no serious comments on the manuscript.
1. It is necessary to correct the design of the names of section 2 (line 82), section 2.2 (line 105).
2. In section 2.1. the chemical composition of the Al alloy and the composition of the cationic electrodeposition coating solution should be given.
These comments do not reduce the overall positive impression of the work.
Author Response
Reviewer 4
Thank you very much for your kind consideration and treatments in your review with careful reading. The authors read the comments and questions of the reviewer, and revised the manuscript, according to them. The replies to them are as follows.
[Comment]
- It is necessary to correct the design of the names of section 2 (line 82), section 2.2 (line 105).
[Reply]
Thank you for your comment. The designs of line 85 and line 109 were corrected.
[Comment]
- In section 2.1. the chemical composition of the Al alloy and the composition of the cationic electrodeposition coating solution should be given.
[Reply]
Thank you for your comment. The composition of the Al alloy is added in line 87. However, the composition of this coating cannot be described here, due to confidentiality obligations with the suppliers. Instead, the authors described in line 104, that this coating is used in automotive coatings.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
In the revised manuscript the authors addressed my comments and recommendations. I believe the manuscript has been sufficiently improved and it can be accepted in present form.
Author Response
I felt so happy, when I read your comment. Thank you so much.