Supramolecular copolymerization driven by integrative self-sorting of hydrogen-bonded rosettes

Molecular recognition to preorganize noncovalently polymerizable supramolecular complexes is a characteristic process of natural supramolecular polymers, and such recognition processes allow for dynamic self-alteration, yielding complex polymer systems with extraordinarily high efficiency in their targeted function. We herein show an example of such molecular recognition-controlled kinetic assembly/disassembly processes within artificial supramolecular polymer systems using six-membered hydrogen-bonded supramolecular complexes (rosettes). Electron-rich and poor monomers are prepared that kinetically coassemble through a temperature-controlled protocol into amorphous coaggregates comprising a diverse mixture of rosettes. Over days, the electrostatic interaction between two monomers induces an integrative self-sorting of rosettes. While the electron-rich monomer inherently forms toroidal homopolymers, the additional electrostatic interaction that can also guide rosette association allows helicoidal growth of supramolecular copolymers that are comprised of an alternating array of two monomers. Upon heating, the helicoidal copolymers undergo a catastrophic transition into amorphous coaggregates via entropy-driven randomization of the monomers in the rosette.

In this manuscript, Professor Yagai et al. report their research on a bioinspired supramolecular copolymerization system based on preorganized rosette complexes. The co-assembly of the electronically complementary monomers exhibits attractive autonomous copolymerization behavior controlled by the kinetic formation of various rosette complexes with different conformations. Both the structure of the copolymers and the mechanism of supramolecular copolymerization are supported substantially by experimental characterization as well as theoretical analysis. The copolymerization into hierarchical helicoids and the catastrophic transition into amorphous aggregates present a very interesting example for biomimic, molecular recognition-controlled kinetic self-assembly and disassembly processes. Overall this excellent work is of high importance for advancement of supramolecular chemistry and the manuscript is written very well.
Minor revision is needed for publication: (1) In page 5 and 6, the delta H of 1, 2 and the coassembly was reported respectively as 63, 54, 57 KJmol-1, and in page 13, delta H value of 56.9 KJmol-1 was reported for coassembly. Here, the number of significant digit for the delta H should be unified. In addition, I suggest to report the delta S values together with the delta H values in the van't Hoff analysis results in page 5 and 6.
(2) In Page 13, 3rd paragraph, it is described that "we diluted the solution to ct = 10 µM, at which the onset of aggregation (33°C) is below the Tc". However, the value of Tc was obtained only after the measurement of delta H and delta S in page 14. The assertion of "below the Tc" should not appear too early.
(3) In Figure S28 in supporting information, for the inset, it is described "using the natural logarithm of the reciprocal ct as a function of the reciprocal temperature" and in the figure the vertical coordinate is label as "ln K". This should be corrected.
Reviewer #2 (Remarks to the Author): This manuscript by Yagai et al. describes the supramolecular copolymerization driven by integrative self-sorting of hydrogen-bonded rosettes. The method is very attractive and elegant since it uses a very simple approach to realize the molecular recognition in supramolecular polymers. I think the result is of sufficiently broad interest for Nature Communications. The manuscript was also a pleasure to read, with concepts and data explained clearly and in depth along with impressive figures. This paper has been among the best written ones during my recent reviews.
I have some comments which I believe are useful to improve the quality of the manuscript.
1. The area of supramolecular organic self-assembly has been developed for years. I acknowledge the contribution from this area to promote the understanding in molecular sciences. Of course, there are a great number of beautiful structures generated from supramolecular self-assembly. However, the eventual applications of this class of materials essentially rely on the bulk properties or the scalable fabrications. One may challenge the significance and the motivation of the current results by asking questions like: Why do we need this kind of research except that we can get pretty microscopy images? I suggest that the authors improve the introduction to justify themselves better.
3. I would ask the authors whether they can control the assembly in a scalable way, since they got the materials in a little amount only for characterizations. And if they can prepare the bulk materials of, for instance, toroids or fibrils, can they align the assemblies into macroscopic ordered structures. If you just get some nice morphologies from AFM or TEM, the aspect of applications will be weakened. Or can you really incorporate the toroids or fibrils into some other mediums to build composites? If you said your systems have mimicked the biological systems, you should not forget that in reality they are related to very complicated environments.
4. Can the authors discuss what factors determine the diameter of the helicoid? Are there any elastic energy relationships responsible for the structure? 5. The plot of Figure 4a is poorly presented. Please change the colors and the style of the lines to make sure that the data are clear to understand. ¬Also Figure 5c and d are too small to see.
Reviewer #3 (Remarks to the Author): The manuscript by Aratsu et al. reports on the exciting and really interesting co-assembly features of two complementary naphthalene-based barbiturates endowed with electron-donating (ether) and electron-withdrawing (ester) functional groups. The manuscript collects a very complete set of experiments, microscopy images and spectroscopic measurements, to justify the experimental evidences found out in the course of such investigations. As in many other publications, as a typical feature of the research group, it is worthy to mention the quality of the AFM images. At the same time, I have to remark the amount of experimental evidences provided to demonstrate the hypothesis of the manuscript. I consider that the manuscript meets the quality criteria to be published in a reputed journal like Nature Comm. However, there are some points that should be addressed prior to its publication. 1) As stated before, the authors naphthalene-based barbiturates endowed with electron-donating (ether) and electron-withdrawing (ester) functional groups. It is obvious that the electronic nature of the ether or ester functional groups is opposite. However, I am very curious about the electronic complementarity of these two functional groups. It would be very useful if the authors could provide any experimental or bibliographic reference about the redox potentials of such compounds that could demonstrate the potential charge transfer effect between these moieties, as it is mentioned in page 6 of the manuscript.
2) In a manuscript dealing with co-assembly, it is very important to derive the thermodynamic parameters associated to the supramolecular polymerization of both the pristine components and also to the investigated mixture of components. The authors have made such calculations and provide the corresponding parameters. However, there are two circumstances that make these values not very accurate. The first one is the fact the authors have utilized the model described by Meijer et al. in 2006 (reference S9 in the Supporting Information). At the same time, and more importantly, the authors have utilized only three curves to derive these parameters. Of course, if you have three point the R2 value for the linear fitting is 1. The authors should: a) make the calculations with, at least, four cooling curves or b) utilize the most recent model published by ten Eikelder and coworkers (J. Phys. Chem. B, 2012, 116, 5291) that allows a global fitting of the data and, consequently, a more accurate determination of the parameters.
3) The authors report on a very complex system composed by three components, is certain cases one of them is chiral. It is absolutely necessary to perform the corresponding studies with binary systems. For instance, how are the binary 1/2 or 2/3 systems behaving? More interesting is the ternary systems incorporating the chiral congener. In this case, the mixtures 1+1S or 2+2S would be examples of sergeant-and-soldiers experiments that provide very useful information about the required ratio of the chiral sergeant, and also the energetics derived of this chiral amplification phenomenon, to further optimize the investigation of the ternary systems.
Overall, I recommend the publication of this timely work in Nature Comm. after major revision.
In this manuscript, Professor Yagai et al. report their research on a bioinspired supramolecular copolymerization system based on preorganized rosette complexes. The co-assembly of the electronically complementary monomers exhibits attractive autonomous copolymerization behavior controlled by the kinetic formation of various rosette complexes with different conformations. Both the structure of the copolymers and the mechanism of supramolecular copolymerization are supported substantially by experimental characterization as well as theoretical analysis. The copolymerization into hierarchical helicoids and the catastrophic transition into amorphous aggregates present a very interesting example for biomimic, molecular recognition-controlled kinetic self-assembly and disassembly processes.
Overall this excellent work is of high importance for advancement of supramolecular chemistry and the manuscript is written very well. Minor revision is needed for publication: Specific Comments (1): In page 5 and 6, the delta H of 1, 2 and the coassembly was reported respectively as 63, 54, 57 KJmol-1, and in page 13, delta H value of 56.9 KJmol -1 was reported for coassembly. Here, the number of significant digit for the delta H should be unified. In addition, I suggest to report the delta S values together with the delta H values in the van't Hoff analysis results in page 5 and 6.

Specific Comments (1):
The area of supramolecular organic self-assembly has been developed for years. I acknowledge the contribution from this area to promote the understanding in molecular sciences. Of course, there are a great number of beautiful structures generated from supramolecular self-assembly.
However, the eventual applications of this class of materials essentially rely on the bulk properties or the scalable fabrications. One may challenge the significance and the motivation of the current results by asking questions like: Why do we need this kind of research except that we can get pretty microscopy images? I suggest that the authors improve the introduction to justify themselves better.

Specific Comments (2):
If the authors acknowledge that their materials are termed as "polymers", they should realize that how polymers can perform superior over other materials. As the monomers are polymerized, the collective behavior shown in the polymers render many useful properties.

Specific Comments (3):
(i)I would ask the authors whether they can control the assembly in a scalable way, since they got the materials in a little amount only for characterizations. And if they can prepare the bulk materials of, for instance, toroids or fibrils, can they align the assemblies into macroscopic ordered structures. If you just get some nice morphologies from AFM or TEM, the aspect of applications will be weakened. (ii)Or can you really incorporate the toroids or fibrils into some other mediums to build composites? (iii)If you said your systems have mimicked the biological Can the authors discuss what factors determine the diameter of the helicoid? Are there any elastic energy relationships responsible for the structure?