Mechanistic Model with Empirical Pitting Onset Approach for Detailed and Efficient Virtual Analysis of Atmospheric Bimetallic Corrosion
, Claes Taxén 1, Bo Rendahl 1, Joakim Voltaire 3, Katarina Eriksson 4, Dominique Thierry 1 and Johan Tidblad 1
Round 1
Reviewer 1 Report
Please find my comments in the attached file.
Comments for author File: 
Comments.pdf
Author Response
Theory and model development
1.“The corrosion product activity, as, is used as the criteria for corrosion product precipitation. The precipitation solubility product should be used as the criteria, not the activity.”
Answer: We realize that the descriptive text was a bit unclear, but the precipitation solubility product is used. The activity of a solid phase (or in this case a corrosion product) can either be 0 or 1 and that is true also in this model. The degree of saturation is computed from the solubility product and super-saturation is used as criterium for precipitation. Text related to Eqs. 13 and 14 have now been updated.
2.“As authors mentioned, activity was considered in the mass transport equation due to the high concentration nature of the liquid thin film. However, for the diffusivity of different ions, did the authors consider the viscosity effect?”
Answer: The viscosity effect is an interesting aspect that we did not account for. In general, it is difficult to find good corrections of the mass transport properties in highly concentrated solutions and we have therefore also been forced to make other simplifications, e.g. no salt-concentration dependency in the diffusivity of ions. However, we think the addition of the non-ideality using activities into the model formulation is a step towards a more accurate description than simply using infinitely dilute solution theory which is quite often the case.
3.“Page 8 line 299, the conical shape pits are considered for the modelling. Can the authors provide more evidence or explanations why conical shape are selected?”
Answer: It was trial and error with close consideration to the references. The conical shape seemed to give a relatively good fit. However, it is a simplification. Due to the many parameters making up the model, we decided to leave this property unchanged throughout the manuscript. For clarity, some rephrasing of the text related to Figure 4 has been added (page 9, lines 293-296). We have added some text in the results section (page 14, lines 483-487) to highlight that the model is restricted to the choice of conical pits and that further adjustments perhaps with the input from microstructural experimental method could be the next step in improving this model.
Results and discussion
“Figure 10a, the experimental pit depth was measured after 5h exposure. The line of experimental data is confusing. I will recommend to use symbols at 5h to show the experimental pit depth.”
Answer: Figure 10a and figure text have been updated for clarity.
Please note that abstract and conclusions have been rephrased. This to meet the comments of the other reviewers.
Reviewer 2 Report
1.The author has taken lot of pain in writing the manuscript and it seems that in order to justify the mechanistic model of atmospheric bimetallic corrosion with simplified empirical approach of the onset of localized corrosion attack, the author has overwritten some of the content and at one stage one might loose the interest in the paper.
2. The author has chosen the study related to bimetallic sample made up of aluminium alloy 1050 and stainless steel 316L sheets, but the motive for choosing this particular combination is missing.
3. In order to understand the corrosion of the alloy-the very basic method of immersion technique is very much needed!!!!!The author could have gone through the weight-loss studies and simulation regarding the same could also be carried out or mentioned in the text !!!!!!
4. Pitting is one of the most important corrosion parameter that one should take care of in the proper evaluation. As author has done simulation in regards to pitting using the average pit depth based on the deepest pits. So, it would really be a little bit challenging in getting the exact calculation and reproducibility in the study. How much result is near to the exactness is questionable????
5. Methods to find the efficient assessment of pitting damage.??needs to be explained..
6. The conclusion must be written in more precise manner-specifically in the 2nd point "A combination of post-exposure and in-operando data were required to validate the model. The latter was achieved with an experimental bimetallic calibration" needs to be pinpointed .....
Author Response
1.“Abstract is poor. It needs to be expanded with more obtained relevant conclusions.”
Answer: We have rephrased parts of the abstract and the conclusions section. Much of this is related to your comments below as well. For details, see below.
2.“Related to comment 1, the underlying microstructure mechanisms that come into picture during pitting owing to atmospheric bimetallic corrosion are absolutely missing for the studied alloys i.e., aluminum alloy 1050 and stainless steel 316L. Simulations are often misleading.”
Answer: The microstructure is simplified and that was also the intention with this Al1050 vs 316L model. We claim that at least for the 5 hours we simulate at controlled RH and salt load, the galvanic coupling dictates the electrochemical behavior locally and that the model can capture more chemical properties than what is usually accounted for. The geometry of the microstructure is simplified and empirical in terms of pit location and projected area, and the impact/limitations of this is also investigated in the manuscript. We are aware that if more advanced exposures, e.g. dry/wet cycling, is to be modelled this will require the model to be updated and there is always room for improvement. However, we consider this work to provide a valuable method using the mechanistic modelling approach for straightforward corrosion and material damage analysis of the material combination under controlled atmospheric conditions. We have rephrased both the abstract and the conclusions to better stress limitations and opportunities using the model.
3.“Related to comment 2, at least few microstructure based studies that demonstrate the validity of the developed mechanistic model need to be added.
*Author(s) should highlight all the modifications carried out in the paper.”
Answer: We agree that more experiments for model calibration can increase the validity of the model. However, we believe that the experiments presented here (and the supplementary material) are relevant and that we offer a degree of model calibration that is extensive. We realize that microstructure experiments are valuable and have added some text on this to ensure that the reader is not mislead in our model description and analysis (page 8 (lines 294-297) and page 14 (lines 483-487)).
Reviewer 3 Report
The submitted manuscript on “Mechanistic model with empirical pitting-onset approach for detailed and efficient virtual analysis of atmospheric bimetallic corrosion” appears to an interesting work. However, there are few issues as listed below that need to be revised by the author(s) before any further recommendation –
1. Abstract is poor. It needs to be expanded with more obtained relevant conclusions.
2. Related to comment 1, the underlying microstructure mechanisms that come into picture during pitting owing to atmospheric bimetallic corrosion are absolutely missing for the studied alloys i.e., aluminum alloy 1050 and stainless steel 316L. Simulations are often misleading.
3. Related to comment 2, atleast few microstructure based studies that demonstrate the validity of the developed mechanistic model need to be added.
*Author(s) should highlight all the modifications carried out in the paper.
Comments for author File: Comments.pdf
Author Response
1.“The author has taken lot of pain in writing the manuscript and it seems that in order to justify the mechanistic model of atmospheric bimetallic corrosion with simplified empirical approach of the onset of localized corrosion attack, the author has overwritten some of the content and at one stage one might loose the interest in the paper.”
Answer: We agree that the manuscript is quite long and we have tried to divide the text into chapters were the reader depending on his or her interest easily can select to read about either the experimental or the modelling work.
2.“The author has chosen the study related to bimetallic sample made up of aluminium alloy 1050 and stainless steel 316L sheets, but the motive for choosing this particular combination is missing.”
Answer: The lightweight aspect is a motivation for the choice of this combination, which is mentioned. 1050 was chosen by the project partners due to its high purity to reduce the system complexity. 316L was chosen due to its extensive use. A note on the choice has been added on page 2 lines 68-70.
3.“In order to understand the corrosion of the alloy-the very basic method of immersion technique is very much needed!!!!!The author could have gone through the weight-loss studies and simulation regarding the same could also be carried out or mentioned in the text !!!!!!”
Answer: We agree that the immersion method is valuable, but our focus was on atmospheric corrosion and developing a model for that. That is why we try to highlight the use of the atmospheric calibration cell. Immersion tests cannot capture all the complexities of the atmospheric corrosion. Carbon dioxide, oxygen dissolution, and more into/within the thin highly concentrated electrolyte film are properties difficult to capture with the immersion method. We actually chose methods comparable to weight-loss measurements in the work. The geometrical mapping experiments performed (see also the supplementary material) gave us information on the mass dissolution in terms of volume. Those data were critical for the pit growth approximation as described by equations 21 and 22.
4.”Pitting is one of the most important corrosion parameter that one should take care of in the proper evaluation. As author has done simulation in regards to pitting using the average pit depth based on the deepest pits. So, it would really be a little bit challenging in getting the exact calculation and reproducibility in the study. How much result is near to the exactness is questionable????”
Answer: The results point to difficulties related to the reproducibility of atmospheric exposures and the relevance of using an averaged/simplified approach for describing several pitting aspects. It is extremely difficult say how near but we believe the model to be useful for analysis at least for controlled atmospheric exposures (constant RH and salt load). Much of this is already discussed in the model results section (especially sections 5.2.1 and 5.2.2). However, we think that these things can be stressed more and therefore the conclusions have now been rephrased to better highlight limitations and opportunities with this model.
- “Methods to find the efficient assessment of pitting damage.??needs to be explained..”
Answer: To clarify this the abstract has been updated. See second last sentence.
- “The conclusion must be written in more precise manner-specifically in the 2nd point "A combination of post-exposure and in-operando data were required to validate the model. The latter was achieved with an experimental bimetallic calibration" needs to be pinpointed .....”
Answer: We agree that some of the points can be clarified. See updated conclusions section.
Round 2
Reviewer 3 Report
Author(s) have done well. In my opinion, the paper is now ready for publication.
Comments for author File: Comments.pdf
