Bulk Tungsten Fiber-Reinforced Tungsten (W_f/W) Composites Using Yarn-Based Textile Preforms

Round 1

Reviewer 1 Report

The authors have made a significant optimization compared to the previous study by using a self-made material with a smaller diameter, which I think is of some significance for practical applications.

1. The abstract is too brief, and it is suggested to add the significance of the authors' research to practical applications.

2. On page 2, line 46, the author has an "Error! The authors should do a lot of checking before submission to reduce this problem.

Reference source not found", I suggest the author to make a lot of changes.

4. What does the author want to express in Figure 14b? Is it the way of surface breakage?

Where is the data in 5.2.2.2? The authors need to list the experimental data. Not two pictures.

What is the test method for fatigue behavior in 6.2.3.3?

7. Why is the citation even needed in the conclusion?

Overall, I think the authors need to make a lot of changes.

Author Response

Response to Reviewer 1 Comments

Dear Reviewer 1, first hand, I would like to thank you very much for your time and effort. I tried to include your suggestions into the new file attached.

Point 1 / 2 / 3:  The abstract is too brief, and it is suggested to add the significance of the authors' research to practical applications. / On page 2, line 46, the author has an "Error! The authors should do a lot of checking before submission to reduce this problem. / reference source not found", I suggest the author to make a lot of changes.

It seems, that during the transfer from my own layout to JNE’s layout, the links and references weren’t transferred correctly. I think, I fixed the layout now with the new attached file. If there are any further adjustments you would like to have, please do not hesitate to suggest this to JNE and me. (e.g. I would decrease the 4,6 cm left boundary, but this seems to be the standard format). At least for me, all crosslinks and references are shown correctly now as in my original file. The abstract is adjusted in the new version.

Point 4: What does the author want to express in Figure 14b? Is it the way of surface breakage?

I wanted to point out, that ductile necking only occurs when there is an effective transfer of the mechanical load to the fiber. This makes the presence of ductile necking a reliable indicator of the occurrence of pseudo-ductile mechanisms of W_f/W, even when a force-displacement curve is unavailable. It should be noted that the samples presented in this work lack an yttria-interface, resulting in exceptional strength between the W-fiber and the CVD-W matrix. However, this has also resulted in some fibers exhibiting brittle fracture and, consequently, a suboptimal performance of the material. Nonetheless, due to the multiple mechanisms contributed by the fibers, a significant improvement in the fracture behavior can be observed even in the absence of an interface, when compared to standard Tungsten. It is worth mentioning that the type of stress applied to the sample also plays a significant role in determining the material's performance, as it can exhibit different behaviours in tensile tests or e.g. 3-point bending tests. In particular, the thin yarns are weaker in 3-point bending tests due to the low stiffness and therefore exhibit lower fracture toughness. As a result, the idea was to highlight the improvement in the material's performance even in worst-case stress states. The pictures show the fracture images of the broken samples and the different fracture behavior of the fibers. I tried to include this information briefly in the new version but I hope I could answer your question.

 

Point 5: Where are the data in 5.2.2.2? The authors need to list the experimental data. Not two pictures.

I regret to inform you that I am uncertain about the exact contents of 5.2.2.2 as it does not exist in my file. Please understand that due to the sheer volume of experimental data, it is not possible to include all of it in the manuscript. In particular, for cyclic loading, measurements were taken every 0.05 seconds over a duration of more than 10000 seconds. Hence, publishing all information in detail would exceed the limits of this manuscript. If you are interested in reviewing the original files, I would be more than happy to provide you with the csv/Excel or OPJU (Origin) file. But the diagrams I showed show exact these results summarized in the manuscript.

Point 6: What is the test method for fatigue behavior in 6.2.3.3?

The fatigue behavior of the material was tested using the same test machine and KLST-type geometries as for classic three-point bending tests. However, instead of continuously increasing the applied force until the material is destroyed as in classic three-point bending tests, a predetermined range below the maximum load was cyclically applied to the material. The material is considered to be exhibiting signs of fatigue, if the machine displacement varies within the set load range over an increasing number of cycles. To determine the point of complete failure, the maximum of the sinoids was trendlined over the number of cycles. As the critical displacement of the machine was previously established through classic three-point bending tests including a complete fracture of the samples, we were able to determine the fatigue behavior without actually breaking the samples completely. It has to be mentioned, that this method can only be used, if there are enough specimen, that can be tested first hand to determine the critical machine displacement.

Also, a high reproducibility of the material properties is required to enable this method. But for our new composite, this has been the case, showing the significant improvement the composite development has made. If you require a better understanding of the cyclic test method, I would be happy to share some video files with you upon request.

 

Point 7: Why is the citation even needed in the conclusion?

One of the main purposes of this work is to show the improvement of the new material to the results presented in previous works with 150 µm fiber based base fabrics. Since these results were demonstrated in a different publication, I was forced to refer to the data published even in the conclusion. I could take out the second reference since I cited it on previous pages but I was asked by my co-authors to include this references again to reduce the effort to access the necessary information to those, who are interested in some background information.

Overall, I hope I could answer your questions with the new version of the manuscript. Please do not hesitate to make further suggestions. The authors will try to fit your expectations as good as possible.

 

Reviewer 2 Report

This is a very important and interesting research result that has been carefully investigated. Please consider the following points.

1) It seems that the linking to the figure in the text is not good throughout, so please improve it.

2) L.210: ”Although not all fibers exhibit ductile necking” Why does necking occur in only one part of the same tungsten fiber? Additional explanation is required.

3)  L.213-215: Which experimental results show the relationship between the number of fibers and reproducibility? If it is past experimental results, please cite references.

4) L.216-218: Please explain why type 1 is superior to type 2 in some respects.

 

Author Response

Dear Reviewer 2, first hand, I would like to thank you very much for your time and the effort.

Point 1:  It seems that the linking to the figure in the text is not good throughout, so please improve it.

Response: It seems, that during the transfer from my own layout to JNE’s layout, the links couldn’t be transferred as intended. I think, I fixed the layout now with the new attached file. If there are any further adjustments you would like to have, please do not hesitate to suggest this to JNE and me. (e.g. I would decrease the 4,6 cm left boundary, but this seems to be the standard format).

Point 2: L.210: ”Although not all fibers exhibit ductile necking” Why does necking occur in only one part of the same tungsten fiber? Additional explanation is required.

I would like to point out, that ductile necking only occurs when there is an effective transfer of the mechanical load to the fiber. This makes the presence of ductile necking a reliable indicator of the occurrence of pseudo-ductile mechanisms in composites, even when a force-displacement curve is unavailable. It should be noted that the samples presented in this work lack an yttria-interface, resulting in exceptional strength between the W-fiber and the CVD-W matrix. However, this has also resulted in some fibers exhibiting brittle fracture and, consequently, a suboptimal performance of the material. Nonetheless, due to the multiple mechanisms contributed by the fibers, a significant improvement in the fracture behavior can be observed even in the absence of an interface, when compared to standard Tungsten. It is worth mentioning that the type of stress applied to the sample also plays a significant role in determining the material's performance, as it can exhibit different behaviours in tensile tests or e.g. 3-point bending tests. In particular, the thin yarns are weaker in 3-point bending tests due to the low stiffness and therefore exhibit lower fracture toughness. As a result, the idea was to highlight the improvement in the material's performance even in worst-case stress states. I tried to include this information briefly in the new version.

Point 3: L.213-215: Which experimental results show the relationship between the number of fibers and reproducibility? If it is past experimental results, please cite references.

Unfortunately, there are no existing references that show a relationship between the number of fibers and the reproducibility, since nobody else uses such a high number of fibers in a metal matrix composite. But this is a logical consequence, that has been made by comparison with the experience and the results presented for the material based on 150 µm fibers. I agree, it is interesting to see e.g. a minimum number of homogeneously distributed fibers, that result in homogenous mechanical stress test results, but this has to be investigated in future experiments. For a material  fitting to industrial requirements, a minimum fibers volume fraction and number of fibers is going to be necessary. But this is always dependent on the stress type and the application purpose of the material. The main goal I am currently working on is a higher design freedom of the WfW composites to make necessary adjustments, optimized for each stress state.

Point 4: L.213-215: Please explain why type 1 is superior to type 2 in some respects.

I tried to extend the abstract in the new version and to include this information here. Short answer: The type 2 still has gaps in the matrix, which has to be prevented to provide a material fitting to industrial requirements. Type 2 cannot fulfil this requirement due to the high layer-spacing.

Round 2

Reviewer 1 Report

Agree to accept.