Effect of Microcracks on the Tensile Properties of 3D Woven Composites

Round 1

Reviewer 1 Report

The contribution of Huang et al proposes a four-step experimental procedure that combines three non-destructive evaluation techniques as X-ray microtomography, acoustic emission and digital image correlation to study the effect of microcracks on the tensile properties of 3D woven composites.

The work is well described and apparently quite innovative in terms of the aforementioned methodology adopted to investigate the damage mechanisms induced in carbon-epoxy specimens by the applied external stress.

The results obtained with appropriate methodologies are clearly described and undoubtedly allow a valuable advancement of knowledge on the topics covered.

Although the scientific relevance of the contribution is appreciable, few minor considerations deserve further attention from the authors:

in the text authors often speak about “nondestructive sample” and “nondestructive tensile tests”. What do the authors mean by these concepts? For example, by default, tensile tests are destructive. Please clarify.

Line 88: for a better understanding of the sentence it is suggested to insert the conjunction “and” between “… AE signal analysis” and “DIC full shield…”.

Line 355: it is recommended to replace “VS” with “vs”.

Author Response

Please see the attachment! Thanks a lot for the valuable feedbacks!

Author Response File: Author Response.pdf

Reviewer 2 Report

I've reviewed the manuscript coatings-1260433 titled "Effect of microcracks on the tensile properties of 3D woven composites" with interest. Few of my comments and suggestions are below:

1. Please check "2" affiliation, looks like a typo there.... "Indiana"
2. Abstract: It would be good is authors can provide results/conclusions  briefly for better understanding of the outcome of this research.
3. Line 34: Application of 3D woven composites, "been widely used in the aerospace......", authors may consider biomedical/dental applications of these as reported in literature "Petersen R, Liu PR. 3D-WOVEN FIBER-REINFORCED COMPOSITE FOR CAD/CAM DENTAL APPLICATION. Sampe J. 2016 May;2016:LB15--0138."
4. Line 180 and Fig.5: Please provide a clear description of Fig. 5 " different strain levels...." for better understanding
5. Line 184: please explain more about "satisfactory coefficient of variation (CoV)...." 
6. Line 192: Provide a short details of "slight increase of the normalized stiffness...." related to fracture mechanism in this section.
7. What is reason of regular decrease in residual modulus along with the damage increases and a slight increase of the normalized stiffness after the plateau at Stage 4 ?

Author Response

Please see the attachment! Thanks a lot for the valuable feedbacks!

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors presented a detailed study of the damage identification, initiation and propagation process and the evaluation of the damage mechanisms in 3D woven composites under tensile loading by using three NDT methods, namely Acoustic Emission, micro-XCT and digital image correlation. The manuscript is in general well written and very interesting for the scientific community. However, I firmly believe it would be much better suited to another MDPI journal, i.e. Journal of Composite Science (https://www.mdpi.com/journal/jcs/about ). After reading the manuscript in detail and cross checking with the “Coatings” Journal’s scope (FYI: The journal covers coatings, surfaces and interfaces in the broader sense.), I propose to transfer (if possible) or resubmit the manuscript to the aforementioned journal. In any case, you can find my review which solely aims to the enhancement of the manuscript for publication below.

Introduction

lines 44-47: Include some references on the mentioned studies.

lines 88-90: What is meant by nondestructive tensile tests? It is not clear, as tensile tests are by their nature destructive. Please elaborate or consider revising.

Experimentation

line 104: Please give some more details on the material used for manufacturing of the tabs and on the adhesive used to attach them on the specimens.

Results and discussion

Figure 5: A colour scale shall be added in those strain maps.

line 274: Consider rewriting to make clear what you mean by "a good completion of the plased scheme".

lines 342-345: The strength increment that is observed for specimens 6, 8 and 10 is marginally significant based on what will be the selected confidence level. If a confidence level of 2 standard deviations from the mean strength of specimens 1-5 is selected, which is usually used in such cases, then all the residual strength results of specimens 6-10 lie in between this confidence level, indicating that the observed increase is not significant and might be caused by other factors such as homogeneity of specimen manufacturing quality. This would require a C-scan ultrasonic imaging of the plate before specimen extraction to identify if the manufacturing quality of all the tested specimens (before testing) is consistent and free of randomly located defects or porosity that could affect the strength of the specimens.

Some comments could be inserted into the text in relation to the above in order be clear that this increase in strength requires further examination to be safely attributed to a cause.

Figure 12: Please add an explanation to the plots that the first column corresponds to the initial test until failure of specimens 1-5. It can be confusing as the values are not residual values but results from the initial tests.

Figure 13: I think there is a mistake in the normalised residual strength plot of figure 13, at least for some of the values. Considering equation 1 that you provide and also the plot in Figure 12 b which shows the residual strength values, I suspect that Specimen 6 corresponds to the 0% crack volume fraction (first point in blue plot of Ns),specimen 7 to the second point of the plot and so on. If this is true then the 1st, the 3rd and the 5th points should have positive values and the 2nd and the 4th points should be almost zero. Consider revision.

Consclusions

line 387:Consider replacing the word “reflection” by the word “deflection”.

line 390: You mention “However, Stage 4 shows a slight increase of the normalized stiffness after the plateau.” This could be mentioned as strain hardening. Both the resin material and the fibers are so tensed in the loading direction at this stage that the molecular orientation changes so that a theoretical maximum stiffness is approached to a degree for the composite. This may cause the strain hardening phenomena that you observe in the plots before final failure.

Author Response

Please see the attachment! Thanks a lot for the valuable feedbacks!

Author Response File: Author Response.pdf

Reviewer 4 Report

The manuscript is well written and describes a method combining 3 types of NDTs in order to elucidate the microcracks origin and evolution. 

Author Response

Thanks  a lot for your valuable feedbacks.

Round 2

Reviewer 3 Report

Most of the reviewer’s comments from the first revision were successfully addressed. However, there are two points that require further consideration. Please see my comments below.

Point 1: OK. Excuse me for the inconvenience then, I did not understand that the manuscript was being submitted to this special issue.

Point 2: OK.

Point 3: Although I understand your reasoning, I disagree. The only occasion and I am commenting on this with many reservations, where the tensile tests that you have performed can possibly be considered non-destructive are the first two stress levels, as you have quantified in your results that the damage is non-existent or minimum. For the rest of the stress levels, although you stop the test before the final failure, the applied stress level was already sufficient to cause internal damage to the specimens as also quantified in your study. Therefore, I propose to remove the word “non-destructive” from all the places in the manuscript where you use it to characterize the tensile tests because it is misleading and does not truly offer any originality or added value to the manuscript. There are several other aspects that make your research valuable for publishing.

Point 4: OK

Point 5: OK

Point 6: OK

Point 7: OK

Point 8: OK

Point 9: It was clear from the beginning (previous version of Figure 13) that the plot is a double Y plot. However, it seems not logical if you consider equation 1 for the normalized residual strength and the plot in Figure 12b. The trend of the normalized residual modulus is ok in Fig. 13, as it follows the values in Figure 12 a. On the other hand, the behaviour of the normalized residual strength (Figure 13 – blue curve - right axis) seems different than the plotted residual strengths in Figure 12 b. Please read again my current and previous comments regarding this point and check the values in both plots (Figure 12b and 13-blue curve). Is the first point in Figure 13 equivalent to the second column in Figure 12b? If yes, there is some mistake in the calculations that shall be corrected.

Point 10: OK

Point 11: OK

Author Response

Thanks a lot for your valuable commments. Please see the attached response letter.

Author Response File: Author Response.docx