Recycling of Titanium Alloy Powders and Swarf through Continuous Extrusion (ConformTM) into Affordable Wire for Additive Manufacturing
Round 1
Reviewer 1 Report
- Please add detailed information on the swarf/powder cleaning procedures in the experiment section;
- Please add oxygen analysis results for all starting materials and fabricated materials;
- Please discuss the differences between the fabricated CP-Ti, swarf Ti-6Al-4V alloy with their counterparts fabricated by other methods, in terms of microstructure and mechanical properties.
- Authors claim that the process cost is about 40% reduction, please add a section to discuss the economic figures of the studied process.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Review of paper entitled: “Recycling of titanium alloy powders and swarf through continuous extrusion (Conform) into affordable wire for additive manufacturing”
The focus of this paper was to explore the validity of using GA Ti64 powders and Ti64 machining swarf as feed products to produce Ti64 wire. I do not recommend publication of this paper.
Major Issues:
I understand that a lot of work has been done here and it does provide a proof of concept that wire can be produced using this process for Ti64. However, I believe this paper requires additional experimentation and therefore the option given by the journal is to "reject". This does not mean that this work is worthless, it just requires additional work to be recommended for publication as a full paper. Hopefully, this review can aid in that process.
The process is described as “continuous” in the title, yet steady state is never reached in the production of the wire. Only two samples of each Ti64 wire are taken, each with different characteristics since they were taken at different locations in an unsteady process. Each of these samples were taken at points in the process where it was an undetermined blend of CPti and Ti64. This means that the properties of the wire characterized in this work are not the true characteristics of the continuous Ti64 wire produced if the process reached steady state. Additionally, the characterization of CPTi wire produced from HDH powder is not novel (published previously by authors) it reduces the novel result to exactly 4 non-repeated samples, which have unknown chemical composition and location (not repeatable). I think this a serious issue for publication as I believe that this significantly reduces the impact and importance of the results.
I recommend that the authors either repeat experimentation, allowing the system to reach steady state, ensure pure Ti64 and take multiple samples at the same condition to improve the robustness of the results. If this is not possible, potentially a restructuring of the paper to change the focus to be on this unsteady component of the Conform process would be appropriate. In this case you would need to properly identify and explore the “blend” of CPti and Ti-64 that was produced.
General comments:
This paper is titled with the words “additive manufacturing”, however at no point in the work is the viability of this wire for AM validated or even discussed. I recommend either adding further experimentation or discussion on AM impact or to remove it from the title and introduction.
Since only 4 samples of interest were actually produced, I am not sure why direct comparisons were not completed using all testing methods at your disposal. Figure 8 and 9 show polarized light micrographs only of the GA powder wire, why was this not done for the swaft wire? Figure 11 has BSE only for the swarf wire not the GA. Why was heat treatment only done on the swarf? The authors should consider a clearer comparison for all tests like what was done in figure 13. The lack of consistency in result comparisons demonstrates major flaws in the experimental design.
Why was the wire drawing even done? It was only done for the CPTi wire and is completely off topic for the paper. I recommend that the entire wire drawing section be removed or include the Ti64 wires. This seems completely out of scope.
Plots and Tables are poorly presented throughout the work, making it very difficult to understand the results. Also many are previous published, I am not sure how stringent Metals is on copyright but I believe all figures should be originals or explicitly referenced as such with permissions.
Figure 1: I am not sure if this figure adds anything and is barely discussed.
Figure 2: While this figure does help explain the process, it does appear to be previously published.
Figure 3: The text in this figure cannot be ready. This figure adds nothing to the paper and should be removed. Although I could not find this figure quickly, I assume it is also previous published.
Figure 4: is unnecessarily messy. In the particle size analysis part of the figure, the colours in the legend are difficult to ready. Really the particle size analysis can be its own figure, I am unsure why it is in this table.
In the figure 4 table portion, initially colums are headed as the 3 types of powder in the study, however a few lines down in the table the column titles are changed to properties and the different rows are now the 3 powder types. This is very odd and sloppy and really looks like 3 separate figures were haphazardly stuck together.
Lastly in figure 4, the rendering looks very poor on the text areas. Not sure why but it should be addressed.
Figure 5: is previously published and adds nothing to the paper. Should be removed.
Figure 6: What material is this wire?
Figure 7: Is the top image the same scale as the bottom image?
Figure 12: This is a mess. Separate this into multiple plots. Individual points should not be connected with lines in a scatter plot. The convention you used in figure 10, while not perfect, was much better. Sample 2 and 4 values for the swarf appear to be missing with no explanation.
Figure 14 should probably be earlier in the results with way more discussion.
Figure 15 and 16: caption is missing what material this is.
Caption in figure 6 uses ”as-conformed” and figure 7 uses “as-extruded”. Are these the same thing?
Specific comments:
Line 33-34: Would powder that has excessive interstitial elements not also be unsuitable for the wire AM process?
Sentence at like 39 should not be a stand alone paragraph.
Like 47 – WAAM is introduced as an acronym, line 65 introduces it again for a different acronym.
Why is section 5.1 in discussion when its clearly new results
There are many more comments, but I think this is enough.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper by S.A. Smythe et al. reported an extrusion method of converting Ti-6Al-4V scrap (powders and swarf) to wire products. The reviewer recommends publication on Metals after minor revisions.
- The authors mentione that more work needs to be done to ensure the chemical purity of the final product in Conclusions. The reviewer understood this process is still under development. However, the reviewer does think the O/C/N contents are a critical piece for this research. The discussion of mechanical properties is meaningless without knowing the contents of interstitial impurities in the starting materials and extruded wires.
- In Introduction, the authors state “…demonstrated successful Conform extrusion without the need for feedstock preheating or an inert atmosphere”. It would be beneficial to include the information of processing atmosphere and feedstock temperature in Experimental as well.
- Figure 14. The pictures of microstructures are too small. These pictures can provide not much information. The reviewer suggests replacing them with pictures at high magnification and better quality.
- Page 13, the second paragraph. Are the “larger grains” possibly caused by dissolving oxygen during the process? Alpha case?
- Page 11, the first paragraph. The authors think the low elastic modulus is caused by porosity. However, according to Figure 10, the average porosity is very low ( 0.02-0.14 vol%?). It seems not high enough to change much of the elastic modulus.
- It would be beneficial to explain how the grain size and the amount of porosity in Experimental.
- Is it possible to use Ti64 powder to preheat the wheel and coat the groove instead of CP-Ti?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
1. The authors need to provide the oxygen analysis results to indicate the oxygen pick-up during the process, to convince the merit of the processing method.
2. Better to provide detailed swarf clean procedure to make sure other research can repeat the experiments.
3. Better to compare/contrast the results with Ti-64 alloy produced by other methods from either powder or machining chips to demonstrate the advantages of the process used in current research, and/or clarify the realted fundamental mechannism(s).
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
I will go into this review a 2nd time with fresh eyes. I may have been a bit critical in my first review. I understand that this is a proof of concept and maybe that point should be emphasized more in the introduction. I really would still have liked to see an actual Ti64 wire be produced to show the possibility prior to publication, but I understand the experimental limitations and timelines.
I apologize for thinking the images were previously published. Some looked very close to previously published images from the authors.
In point 2, the authors state “Previous experimental work using Ti-6Al-4V powders to “warm up” the processes resulting in significant tool abrasion and premature tool failures” as the reason the authors required the CP-Ti for warm up. Do the authors believe that that if after warm up with CPTi, production of pure Ti64 wire would eventually be possible without contamination of CPTi? Or will the system always require some CPTi in the blend to reduce tool failure? This issue is touched upon in the discussion, but still needs further clarification because as a proof of concept it is still unclear if a pure Ti64 wire could be ever be produced.
In point 3, the authors state “It is strongly argued that the work is repeatable”. However, the locations of the samples, the composition of the samples, and the warm up time are all unknown. Yes it is possible that the authors may be able to replicate the study in their lab under the same conditions, but not nearly enough information is available to the reader. This reduces the significance of any of the mechanical properties – to the reader; they are just the properties of any random blend under any conditions. Because we don't know the composition, the actual numerical values could be anything and it does not change the outcome of this paper. The authors should attempt to clarify the composition, or failing that, the location of where the samples were taken.
If I understand correctly, each data point for stress and elongation is only from 1 sample at each of the 5 (or 3) locations (n=1). It is unclear how the authors added the error bars to figure 14. If the authors did do multiple tests per location point please make this clear, or make it clear what the error bars represent. More testing would have made the results more robust, but again I understand the limitations of experiments, just make the text clear that the results are only of a single sample and why this is the case.
In section 2.3 and 3.1 it is unclear which CP-Ti wire the authors are wiredrawing (which tooling diameter and abutment material). This needs to be clarified (it is clarified in the conclusions on line 648, but until then I was unsure).
In line 644 the authors state “mechanical properties of the as-conformed wire are good enough for use in wire deposition AM processes”. At no point were the requirements for AM processes discussed nor was the wire qualified. If the ability for the wire to be cold drawn is the qualification, then only the one CP-ti wire underwent this qualification. This must be clarified.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 3
Reviewer 1 Report
If the authors can not provide a detailed swarf cleaning process, I would suggest provide chemical compositions of the staring swarf materials and solid materials made from swarf in the manuscript.
Author Response
If the authors can not provide a detailed swarf cleaning process, I would suggest provide chemical compositions of the staring swarf materials and solid materials made from swarf in the manuscript.
The material is Ti-6Al-4V Grade 5, with a chemical composition of 5.5 – 6.5 wt. % Al and 3.5 – 4.5 wt. % V (see line 124). Chemical analysis was not conducted on the material produced after the trial.
Reviewer 2 Report
Point 5: It still needs to be clarified in the text what qualifies the wire in this statement: "mechanical properties of the as-conformed wire are good enough for use in wire deposition AM processes”.
Author Response
It still needs to be clarified in the text what qualifies the wire in this statement: "mechanical properties of the as-conformed wire are good enough for use in wire deposition AM processes”.
The key qualifications for wire AM are a material that can be handled during the process and a diameter that will fit. Other than those qualifications, there are no rigorous requirements for the mechanical properties of AM feedstocks but a good ductility and a reasonable yield strength is sufficient. The materials produced in this work successfully achieve those mechanical properties, and the key qualifications mentioned are achieved by the cold drawn CP-Ti.
The statement mentioned in the review has been amended to reiterate that the work is not industrially ready but the current mechanical performance makes the product a strong potential for use in wire deposition processing (see lines 432 – 435).
