A Comparison of 90° Bending for Foldable Electronics

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

- The introduction should be revised to fully reflect the title of the manuscript. Especially the first paragraph should be expanded. Namely, first of all, examples or usage areas for flexible and foldable electronics should be mentioned. Secondly, in addition to bending, flexing, rolling and folding applications applied to these materials should be mentioned.

- There is no reference to literature in the comments for Figs.5 and 6. However, there are similar studies. In this context, evaluations should be reviewed by taking into account the studies in the literature. Additionally, the results for Mo and Al/Mo added electronics are interpreted in terms of the bending method and related parameters. How correct is it to ignore the mechanical properties of film materials when evaluating the results?

- In Section 4.1, paragraph 2, it was mentioned that springback will decrease with decreasing substrate thickness. However, as it is known, springback in metal bending operations decreases as the plate thickness increases. Therefore, the relevant explanation should be checked and the necessary improvement should be made.

 

- Recommendations for foldable and rollable systems are given in the 4th paragraph of Section 4.3. However, Rolling mechanics are different. Therefore, it is not correct to give recommendations about rollable based on the results obtained by bending. This section should be improved.

 

Author Response

- The introduction should be revised to fully reflect the title of the manuscript. Especially the first paragraph should be expanded. Namely, first of all, examples or usage areas for flexible and foldable electronics should be mentioned. Secondly, in addition to bending, flexing, rolling and folding applications applied to these materials should be mentioned.

Examples and usage areas for flexible and foldable electronics have been added to the first paragraph of the introduction.

The final sentence of the Reviewer is not clear and we do not know how to address. Is the Reviewer asking to add what material systems are generally investigated? If we have interpreted correctly, a statement has been added to the revised manuscript.

The goal of the manuscript was to illustrate that bending and rolling are not the same, but are being treated the same, and to focus on the simplest bending geometry, 90-degree bending. We thought that a short review of the possible bending options available and how they differ was necessary. Additionally, we only have access to 90-degree bending and have no first-hand experience with the other bending geometries. The rest of the manuscript only considers the 90-degree bending tests.

- There is no reference to literature in the comments for Figs.5 and 6. However, there are similar studies. In this context, evaluations should be reviewed by taking into account the studies in the literature. Additionally, the results for Mo and Al/Mo added electronics are interpreted in terms of the bending method and related parameters. How correct is it to ignore the mechanical properties of film materials when evaluating the results?

The authors do understand why references to the literature need to be given for the new data presented in Figures 5 and 6. During our literature review no other papers were found, other than our own which has now been added, that presented the mechanical damage and change in electrical resistance in such a way. Please direct us towards possible literature that we have missed as we would be happy to review it and include if relevant. We have attempted to provide another way to examine and present the data in case one does not have in-situ electrical capabilities and to investigate the evolution of the mechanical damage with the electrical resistance response. This can only be completed using intermittent or multiple tests to different numbers of cycles. Additionally, when the mechanical damage is presented in papers, is it generally after testing with no information of the evolution. Many, if not most, literature focus on the electrical resistance and not the combination of mechanical damage and electrical resistance. Therefore, we feel as though we could not make any good comparisons to the literature other than our own work.

The Reviewer is correct that the results were interpreted in terms of the 90-degree bending parameters. However, the other film properties were not ignored. In section 2.2 the monotonic tensile testing was presented to provide the crack onset strains for each film system as well as the crack density at maximum applied strain to compare to the bending damage. The section has been highlighted. Other mechanical properties, such as hardness, was not evaluated since it is impossible to interpret this property on a polymer substrate. It can be said that the apparent fracture toughness of Mo could be higher in the Al/Mo bilayer compared to the single Mo film from some recent work [1].

1. Cordill, M.J.; Kreiml, P.; Jörg, T.; Zak, S.; Mitterer, C. Parameters Influencing the Fracture of Mo Films and Their Wider Significance. MRS Adv 2023, doi:10.1557/s43580-023-00612-3.

 

- In Section 4.1, paragraph 2, it was mentioned that springback will decrease with decreasing substrate thickness. However, as it is known, springback in metal bending operations decreases as the plate thickness increases. Therefore, the relevant explanation should be checked and the necessary improvement should be made.

The Reviewer is correct that for metal substrates the spring back would decrease with increasing thickness. The spring back is more of a combination of different factors. First is the stiffness (or Youngs modulus), it is smaller for the polymer substrate and it leads to an inherently lower spring back in comparison to metals. There is also the density (weight) and also damping. Damping can be neglected since it should not change much with the substrate thickness (in the range being considered for flexible electronics). Thus, the main effects will come from stiffness and density. Stiffness provides the "force" for spring back (similar to a spring constant) and the weight influences how much force (or energy) is needed to "spring it" as well as how long sample may “flop” around. If the stiffness would be high, such as for a metal, one would have a high force to "flop" and if density and weight increase there is a need for a higher force. Which is the case for metals with increasing substrate thickness. The overall increasing mass and bending stiffness (with higher thickness) are acting against the spring back in metal substrates.

For the polymer substrates, the same parameters are present, but these effects are simply too small to observe and there could be an increase in spring back as the polymer substrate decreases. However, the spring back is also counteracted by air friction (air resistance). The "springiness" of the polymer substrate is so small that it is much more affected by the air resistance than a metal substrate and the thinner it is, the higher the impact of air will be. For a low modulus, thin and light material, the effect of change in thickness of the substrate is negligible in comparison to simple air resistance, and the air resistance will, in the case of thin polymer substrates, make the polymer spring back less if the sample is lighter, which is equivalent to a smaller volume (thickness).

We have recently created a finite element model of the FLEX-E-TEST bending which includes spring back and it will be presented in a future manuscript. A short explanation has been added to the revised manuscript to briefly explain the concept.

- Recommendations for foldable and rollable systems are given in the 4th paragraph of Section 4.3. However, Rolling mechanics are different. Therefore, it is not correct to give recommendations about rollable based on the results obtained by bending. This section should be improved.

Rolling mechanisms are different as pointed out by the Reviewer. Section 4.3 is more focused on the 90-degree bending geometry and the discussion points remain fixed to bending throughout the section. Rolling is only mentioned in the section to: 1) not leave out the rolling, and 2) illustrate that rolling tests also require the same four aspects in order for comparisons to be made between research groups or other rolling instrumentation. The four aspects presented are universal no matter the damage mechanisms and would allow for a qualitative comparison. The message is to point researchers in that direction so that the data can be better compared among different groups to increase the speed of new knowledge to apply to real-world applications.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript entitled A comparison of 90° bending for foldable electronics, reports the assessment of foldable electronics, including folding or bending tests on model systems. The authors report that the current bending tests are not equal in their measure as they differ in configurations which can lead to different mechanical damage, as well as electrical response. The 90° bending instruments were compared using metallic thin film systems (300 nm Mo and 130 nm Al on 50 nm Mo) on polyimide substrates to establish if the two seemingly similar bending tests yield the same results. These two film systems were tested in the custom-built FLEX-E-TEST and the commercially available YUASA test which is capable of in-situ resistance measurements under folding or bending. For statistics, Several (10-12) samples were tested to gain statistics of each film on a folding device using the same applied bending strain and number of cycles. Samples were intermittently characterized with confocal laser scanning microscopy and electrical resistance to correlate the amount of mechanical damage (crack density) and the normalized electrical resistance or ratio of the damaged area. The test results showed that with the same bending radius, similar but not the same amount of mechanical damage forms for both bending devices. Additionally, the resistance as a function of cycles also differs after 10,000 cycles. A closer examination of the damage, especially in the Al/Mo film indicates that the speed of the bending (FLEX-E-TEST) and if the samples experience springback (YUASA test), can alter the received mechanical damage. The in-situ resistance data of the YUASA test was further examined and a standardization of folding or bending test results should be reported is discussed. Despite this interesting discussion and report and timely research on flexible and wearable electronics relevant to biomedicine among others, there are some minor, but important points that must be addressed:

1. I don't see why authors have chosen only this commercial system for testing these bending results. Typically people design their system to evaluate such results just like the authors did. Highlight the differences a well as the advantages and disadvantages of each of them with some relevant refs.

2. Some minor English deficiency needs to be addressed, so I recommend reading the entire manuscript and taking care of those as there are several throughout the manuscript.

3. The Figs. 8. 9 and 12 need to be presented in a consolidated and representative manner as there are too many and do not convey much to the reader. The same goes for Fig. 5 and 6. They can be moved to the Supplementary Information.

4. Finally, it is an interesting study as mentioned above, BUT this study is far more relevant and for more visibility in the Electronics Journals.

Once these revisions are taken care of, it can be reconsidered for publication.

 

Comments on the Quality of English Language

Minor Editorial English revision is required.

Author Response

Despite this interesting discussion and report and timely research on flexible and wearable electronics relevant to biomedicine among others, there are some minor, but important points that must be addressed:

1. I don't see why authors have chosen only this commercial system for testing these bending results. Typically people design their system to evaluate such results just like the authors did. Highlight the differences a well as the advantages and disadvantages of each of them with some relevant refs.

This commercial system was used because we had access to it. We noticed that we were getting different results between the commercial YUASA and our custom designed FLEX-E-TEST and thought a full study was needed. Everything about the samples was kept as similar as possible so that only the machine differences would stand out. As mentioned in the manuscript, the YUASA system was further modified to load exactly like the FLEX-E-TEST. Thus, one outcome of the study was to demonstrate that not all 90-degree bending tests ARE the same. One should not simply design their own system. The results will not be comparable to other systems unless all factors are considered. The advantages and disadvantages are highlighted in section 2.3.

2. Some minor English deficiency needs to be addressed, so I recommend reading the entire manuscript and taking care of those as there are several throughout the manuscript.

The manuscript was written and revised by native English speakers. We hope to corrected any grammatical mistakes in the revised manuscript.

3. The Figs. 8. 9 and 12 need to be presented in a consolidated and representative manner as there are too many and do not convey much to the reader. The same goes for Fig. 5 and 6. They can be moved to the Supplementary Information.

The point of Figures 8 and 9 are to demonstrate that more than 1 sample should be evaluated. Each sample had a different electrical response to the bending even though the samples were made to be as similar as possible (pre-cut substrates, same deposition run, same bending parameters, etc.). It is difficult to consolidate the graphs and still have them be readable to a large audience. For example, colorblind readers would have great difficulty seeing the individual samples if all were plotted together. Even the data shown in Figures 5 and 6 illustrate that each sample has its own mechanical and electrical response which needs to be rectified before understanding how to define failure or design criteria. From our literature review, connecting the evolution of the mechanical damage density to the change in electrical resistance is hardly, if ever, shown. Most publications only focus on the electrical resistance aspect and ignore the mechanical damage or only example the mechanical damage the end of the testing. We have shown that the two are highly connected and need to be considered together.

Figure 12 illustrates that one should look at electrical data in detail and also capture it fully. Presenting the delta R (assuming it is the Rmax-Rmin, most publications do not fully clarify how delta R is defined) is only a small part and not the main “character” of the story. The cracks open at Rmax and close at Rmin. These are the parameters that relate directly to the mechanical damage and not the average (the delta R) which is somewhere halfway through the bending cycle. The critical points of folding are at the maximum fold and the unfolded device. The authors are suggesting that more can be learned by examining the full electrical resistance curves, such as the morphology of the formed cracks.

We wish to have all of the figures remain in the manuscript because otherwise, readers may not download the supplementary material to see the important effect that the figures demonstrate. These figures are key to the manuscript and removing them would completely reduce the worth and visibility of the manuscript. Several key points are highlighted in the revised version.

4. Finally, it is an interesting study as mentioned above, BUT this study is far more relevant and for more visibility in the Electronics Journals.

Thank you for this comment. Electronics Journals did not see its merit and why we chose Coatings to publish the manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

In the work two types of films (Mo films and Al/Mo/bilayer) on polyimide (PI) substrates were deposited by magnetron sputtering technique. Obtained film systems were tested by cycle bending. Tests were performed using 90^o bending. 

The authors have made the necessary corrections in accordance with the reviewers^' recommendations. In this form the manuscript seems to be suitable for publication.