Ultrarobust, tough and highly stretchable self-healing materials based on cartilage-inspired noncovalent assembly nanostructure

Self-healing materials integrated with excellent mechanical strength and simultaneously high healing efficiency would be of great use in many fields, however their fabrication has been proven extremely challenging. Here, inspired by biological cartilage, we present an ultrarobust self-healing material by incorporating high density noncovalent bonds at the interfaces between the dentritic tannic acid-modified tungsten disulfide nanosheets and polyurethane matrix to collectively produce a strong interfacial interaction. The resultant nanocomposite material with interwoven network shows excellent tensile strength (52.3 MPa), high toughness (282.7 MJ m‒3, which is 1.6 times higher than spider silk and 9.4 times higher than metallic aluminum), high stretchability (1020.8%) and excellent healing efficiency (80–100%), which overturns the previous understanding of traditional noncovalent bonding self-healing materials where high mechanical robustness and healing ability are mutually exclusive. Moreover, the interfacical supramolecular crosslinking structure enables the functional-healing ability of the resultant flexible smart actuation devices. This work opens an avenue toward the development of ultrarobust self-healing materials for various flexible functional devices.

line 70: make cartilage mechanical strong, line 138: the verb is missing, line 176: the deformation mechanical of the composite...and there are many others).
-The abstract should indicate what materials have been used, instead of the vague: nanosheets and polymer matrix. -in the intro, around lines 47-48 the schematic of the cartilage structure should already be presented, to better illustrate the work.
-Line 64 to 77 does not belong to the section on Results. Instead, this is background which should be in the introduction.
-line 86-87 and later...it is not clear at all why you should get an interwoven network of WS2 at this stage and later in the document. Certainly, this is a key aspect of the work and a main result of this work, as illustrated in the figure 2e and f; the WS2 structure assembles into a hexagonal pattern, which creates this network that may be crucial for the mechanical behavior. However, I do not see any discussion on this point. For me, looking into the experimental procedures, this is a direct consequence of the latex route used to process the samples, but there may be other physical effects linked to the processing route, drying method, etc. Or is the binding energy between PU and the WS-TA sufficient to explain the whole range of properties. Can the authors comment and evaluate the role of this structure? Have they tried different processing routes? -in all results presented in the main document, there is no precision of the volume content of WS2 in the samples. Are these all the same composition? A range of compositions was investigated, as explained in the experimental section and in the supplementary information, what the samples are should be precisely indicated in the figures or the text.
-the experimental section is not complete and there is no mention of the measurement of mechanical properties, size of samples, machine used, how are the properties measured (in particular the modulus), and how many repeats of samples? How was the self-healing assessed (beyond a basic observation as reported in the main document showing repair by holding a weight (another typo in the text, by the way, it is weight not weigh)?). How many repeats, and what was the time left for healing, at what temperature? -these materials are all very sensitive to humidity, due to the strong role of H bonds. Was the RH controlled in the sample storage, in the experiments? I could not find any mention to this in the document.
-line 337, what is a vacuum slam? Please explain, as the reader may not understand what this refers to.
- Table 1 in the supplementary section, please remove the % that appear in some of the boxes in the healing efficiency section.
Reviewer #3 (Remarks to the Author): In short, mechanical properties of nanocomposite hydrogels are very impressive. Several minor comments should be addressed before publication. 1. First, writing needs to be improved. In the first several pages, particularly, in the abstract and conclusion, nothing is mentioned for this hydrogel system. It makes very difficult to understand what is the novelty of current study. 2. Simulation modeling in Figure 2i is poorly described, what methods are used to calculate the binding energy? It is a common flaw that experiments applied an inaccurate modeling with many artifacts to quickly demonstrate the experimental observation. 3. what are water contents for different hydrogel systems? Water content is a critical parameter to control mechanical and actuation performance of hydrogels. 4. Where are the self-healing data? What are the tensile properties of self-healed hydrogels? Selfhealing data should be presented in main text.
Dec. 09, 2020 Dear Editors, Thank you very much for your efficient work in processing our manuscript entitled "Ultrarobust, tough and highly stretchable self-healing materials based on cartilage-inspired noncovalent assembly nanostructure" (Manuscript ID: NCOMMS-20-40345). We have carefully read the comments of the reviewers. Based on these comments and suggestions, we made careful modifications on the original manuscript through conducting a series of additional experiments.
We feel that the revised manuscript is a great improvement on the original. All the revisions have been highlighted in blue color in the revised manuscript. We hope that the modified manuscript will meet the standard of the journal Nature Communications. Our point-by-point responses to the reviewers' comments/suggestions are presented as follows: To the reviewer 1: This interesting manuscript reported a bioinspired nanocomposite assembled by inorganic nanosheets and polyurethane matrix with robust mechanical strength and high self-healing efficiency. The method used to fabricate the materials with high performance is simple and effective. In addition, the mechanical properties and healing efficiency are impressive.
However, the crucial characterizations of these materials are not sufficient. Overall, I believe this work has the potential to be appreciated by the readership of Nature Communications after major revisions. Some comments are listed below: We acknowledge your positive comments and suggestions. We have revised our manuscript in accordance with your instructive guidance. Hopefully we have addressed your concerns.
Thank you very much. 3) The elasticity of the hybrid materials after adding the 2D nanosheets and the healing process should be carefully studied by cycling stretching tests.
√ Based on the reviewer's concern, we evaluated the elasticity and healing process of the TA-WS 2 /PU nanocomposites. The results were added in the Revised Supporting Information 5) The caption of Fig. 3j is "Comparison of Young's modulus, elongation at break, ultimate tensile strength, toughness, self-healing efficiency, and functional healing ability of our nanocomposite with other self-healing materials". However, the Young's modulus is missing.
Therefore, the comparison of Young's modulus should be added in Fig. 3j.

To the reviewer 2:
This article presents a new self-healing elastomer nanocomposite based on 2D Tungsten disulphide nanosheets, functionalized with Tannic Acid, and a PU matrix. The results are of interest to the community, demonstrating a good combination of high toughness and elongation to break, but yet with a reasonably high tensile strength (50 MPa). Moreover, the material apparently is able to heal cuts, although precise details on this are lacking in the document. This combination of materials is novel, at least to the best of my knowledge, although inspiration from cartilage structure is not new in the field of novel bio-inspired materials, and deserves publication. However, the article in its present form lacks precision in the information and more insight in the mechanisms leading to these results. In particular, the role of the WS 2 content in the results in not analyzed (apart from 1 graph in the Supplementary materials, but would bring insight into the relative contribution of each physical parameters. More precise comments follow: First of all, we acknowledge your comments and suggestions, which are valuable in improving the quality of our manuscript. We have added the relevant operation procedures and details of self-healing experiments to Revised Manuscript. Furthermore, the influence of WS 2 content on the results and the assembly mechanism were explored. Hopefully we have addressed your concerns. We revised our manuscript in accordance with your instructive guidance and we feel that the revised manuscript is a significant improvement on the original one. 1) First of all, the language should be revised, there are many grammatical errors, and strangely built sentences that make the reading tedious. at this stage and later in the document. Certainly, this is a key aspect of the work and a main result of this work, as illustrated in the figure 2e and f; the WS2 structure assembles into a hexagonal pattern, which creates this network that may be crucial for the mechanical behavior. However, I do not see any discussion on this point. For me, looking into the experimental procedures, this is a direct consequence of the latex route used to process the samples, but there may be other physical effects linked to the processing route, drying method, etc. Or is the binding energy between PU and the WS-TA sufficient to explain the whole range of properties. Can the authors comment and evaluate the role of this structure?
Have they tried different processing routes?
√ For the WS 2 structure assembles, due to the strong electronegativity of TA-WS 2 nanosheets, the nanosheets are directionally distributed around the negatively charged PU microspheres through electrostatic repulsion. Furthermore, the branches of TA possess multiple hydrogen-bonding sites including multiple ester groups and phenolic hydroxyls. It is envisioned that such structure enables the TA molecule to entrap and bind PU chains through hydrogen bonds, thus maximizing the bonding chance between the hydrogen-bonding sites and easy to form high density hydrogen bonds. Therefore, the strong hydrogen bonding interaction induced the TA-WS 2 nanosheets to self-assemble into an interwoven network around the PU latex microsphere. As a control experiment, we dried the mixture of WS 2 and PU latex directly after vacuum filtrating. As shown in Supplementary Fig. 4a, the resulting elastomer without hydrogen bonding interaction is clearly stratified due to the nanosheets settled in latex in scanning electron microscope (SEM) observation. Also, the control sample has no desirable enhancement effect on mechanical properties ( Supplementary Fig. 4b).
Thank you for your pertinent advice. √ The vacuum slam here refers to the distance between WS 2 monolayers. According to the reviewer's suggestion, the vacuum slam has been specified in the Revised Manuscript. 10) Table 1 in the supplementary section, please remove the % that appear in some of the boxes in the healing efficiency section.
√ According to the reviewer's comments, we have removed the % that appear in some of the boxes in the healing efficiency section. Thanks for your helpful suggestion.

To the reviewer 3:
In short, mechanical properties of nanocomposite hydrogels are very impressive. Several minor comments should be addressed before publication.
We acknowledge your positive comments and suggestions. We have revised our manuscript in accordance with your instructive guidance. Hopefully we have addressed your concerns.
Thank you very much. 1) First, writing needs to be improved. In the first several pages, particularly, in the abstract and conclusion, nothing is mentioned for this hydrogel system. It makes very difficult to understand what is the novelty of current study.
√ We have improved the language in the Revised Manuscript. Based on the reviewer's suggestion, we have specified the materials system in the abstract and conclusion. Thanks for your helpful suggestion.
2) Simulation modeling in Figure 2i is poorly described, what methods are used to calculate the binding energy? It is a common flaw that experiments applied an inaccurate modeling with many artifacts to quickly demonstrate the experimental observation. Then, the binding energy can be calculated based on the final conformation model according to the following equation: E bind =E PU +E filler -E total , where E bind is the binding energy between PU and filler; E PU and E filler represent the corresponding energy of PU and filler, and E total is the total energy of the overall system.
3) what are water contents for different hydrogel systems? Water content is a critical parameter to control mechanical and actuation performance of hydrogels.
√ Since our material system is based on a rubbery elastomer, it is not sensitive to water.
According to the reviewer's comments, we performed the mechanical tests under different humidity (RH 50%, 60%, 70%, 80%). The results are shown in the Revised Manuscript (line 229-232, Supplementary Fig. 11). The results illustrated that water content has little effect on the mechanical performance. Similarly, there will be no change in actuation performance.
Thank you for your pertinent advice. 4) Where are the self-healing data? What are the tensile properties of self-healed hydrogels?
Self-healing data should be presented in main text.
√ Based on the reviewer's comments, we have provided the typical tensile curves of the hybrid materials and self-healing samples in Fig. 3 in the Revised Manuscript. Thank you very much for your suggestions.
Thank you very much for your assistance in this review process. The manuscript has been