Well, given the increased attention and articles about our recent
paper, I decided to write a short explanation of what it actually says and the implications for those who are interested but can't read the paper behind the paywall or not interested in wading through it.
First, a short description of what was done. We used electrospinning to produce continuous fibers with diameters between ~100 nanometers and a few microns (for comparison, human hair is about 100 microns in diameter). For those unfamiliar with the process, electrospinning, similar to electrospraying, applies high voltage to a solution of polymer (you can also use a polymer melt, but it's less versatile and more complicated). Above a certain polymer concentration this process produces fibers that are deposited on a collector (in regular cases it produces a random mat and you need a special electrode to produce oriented or individual fibers). The diameter of the fibers is controlled by different process parameters and can be as low as few nanometers for some polymer systems (in our case we tested fibers as thin as ~100 nanometers).
After making the fibers we mechanically tested them. Other people previously tested such nanofibers, but usually on very short samples (several tens of microns)and the examined diameter range was much narrower. What we saw was that the fibers became much much stronger as they got thinner. This is not entirely unexpected (there are theories that predict this effect), though the magnitude of this size effect was beyond what we've expected. What was unexpected that the deformation to failure did not decrease with the increase in strength, which led to smaller fibers being able to absorb significantly more energy (expressed as toughness) than the large ones (toughness is the area under the stress/strain curve).
Now, optimally, one would have liked to examine the structure of individual nanofibers, but examining crystal structure of such small samples is rather difficult (especially for polymers). We examined crystallinity of nanofiber mats (with distribution of diameters) and found that it decreased slightly for mats with thinner average diameter. This led us to hypothesize that our increase in toughness is associated with the low and decreasing crystallinity. In order to validate the hypothesis we intentionally increased crystallinity in our fibers by heating them up. After this process, we found that the deformation the fibers were able to take before breaking decreased significantly (and thus the toughness of the fibers decreased as well). We also made an effort in the other direction (to decrease crystallinity), and found, as expected, that the toughness increased, but there were confounding effects and we decided to leave this part of the experiments for future papers.
Now, for the claims out there and for the implications of our study. As it stands, taking the fibers and using them is still a long way off. Controlling the diameter and scaling things up is not trivial. Furthermore, though the strength of our fibers is relatively high for the thinnest filaments, it is just starting to push the advanced fibers which are used in composites, armor etc. The bigger problem though is that the large toughness comes from large deformations, which is not very useful for many applications (in ballistics, for example, you are interested in the first 10% deformation and how much energy you absorb there). So, while the results are promising, it is waaay too early to talk about shirts that can stop bullets :-) The study does offer some possible paths forward. I don't want to go into too much details here, but there are possibilities to try and tailor the properties, by sacrificing some of the toughness at high deformations to get higher strength etc. In addition, the discovery that the low crystallinity might be a good thing is very unusual and goes against what people regularly do to produce advanced fibers. I hope this helps to clear some of the clutter out there. Or maybe it just confuses people who ventured here to read this post even more :-)