Close-packed polymer crystals from two-monomer-connected precursors

The design of crystalline polymers is intellectually stimulating and synthetically challenging, especially when the polymerization of any monomer occurs in a linear dimension. Such linear growth often leads to entropically driven chain entanglements and thus is detrimental to attempts to realize the full potential of conjugated molecular structures. Here we report the polymerization of two-monomer-connected precursors (TMCPs) in which two pyrrole units are linked through a connector, yielding highly crystalline polymers. The simultaneous growth of the TMCP results in a close-packed crystal in polypyrrole (PPy) at the molecular scale with either a hexagonal close-packed or face-centred cubic structure, as confirmed by high-voltage electron microscopy, and the structure that formed could be controlled by simply changing the connector. The electrical conductivity of the TMCP-based PPy is almost 35 times that of single-monomer-based PPy, demonstrating its promise for application in diverse fields.

The manuscript "3-dimensional close-packed polymer crystals from two-monomer-connected precursors" shows a new method for the preparation of highly crystalline conjugated polymer films using precursors in which 2 pyrrole units are linked through a connector. The connectors are the acids MSA, EDSA, BDSA and BPDSA. The polymerization is performed in the presence of ammonium persulfate. Overall, I find the ideas of the manuscript highly interesting and novel and I think it would be suitable for Nature Communications. The manuscript is well-written and clear. I would however ask the authors to address the following issues before the manuscript can be considered for publication: How exactly is the polymerization performed on the substrate? Is polymerization only happening at the substrates or also within and at the walls of the vials? The authors should cite relevant literature on this interfacial process. 2) The authors should provide data on the pure pyrrole polymerization using the oxidantwithout the acid. 3) The authors write "..... exhibited redshifted absorption spectra.... because of the regular growth of the polymers from the TMCPs..." on page 4. This is an interesting statement since the polymerization will lead -under the conditions used -in charged polypyrrole backbones. To solve this issue the authors should prepare their films on conducting substrates. Cyclic voltammetry and In-situ spectroelectrochemistry will give information about the neutral and the charged species. Only based on these data the observed red-shifts can be properly understood and explained. The spectra given in Figure S4 are not very clear and meaningful. This will also help the authors to be able to make statements about the polymerization mechanism and the nature of the crystalline films. It would be interesting to have statements about the degree of polymerization.

4)
The reviewer is not convinced that "the DSA links caused the Ppy to be conductive" (page 4 last line) Ppy becomes conducting due to the oxidative polymerization used. 5) The exact conditions of the van der Pauw method should be given. Errors of the measurements are needed as well.
[redacted] 7) Figure S2: It is really hard to distinguish the bands in the FT-IR spectra.

Reviewer #2 (Remarks to the Author):
This Group reported successful preparation of 3-dimmensional polymer crystals by connecting two monomers with different disulfuric acid linkers and polymerizing the precursors. The formation of 3-D polymer crystals were confirmed by various analytical techniques of XRD, HRTEM, and FFT images.
The film of resulting polymers demonstrated flexibility on the various substrates. The work is well executed and the manuscript is also written very well, which makes this work above average standards.
However, the present work is very similar to work reported by author himself in the "Advanced Materials, 2012, 24, 3253-3257". Author has earlier reported that polymers containing pyridine can be cross-linked by dibromoalkane which forms 3-D molecular level ordering. In the current manuscript, author has reported formation of crystal using same strategy, and the previous work has not been cited. Author has not discussed any performance improvement over the previous work. It suggests that this manuscript lags significantly in originality. In addition, manuscript has some technical flaws, which needed to be addressed before the work is submitted to any journal.
(1) The synthesis of P(Py:MSA) and P(Py:DSA:Py) films by in-situ chemical oxidative polymerization has not been described completely as well as purification and characterizations especially. In detail, after polymerization, they characterized the polymers by only FT-IR, UV-Vis, and XPS. However, for investigation of polymer characteristics, other characterization such as NMR spectra, DSC, and TGA, etc are very important and needed to be reported. (2) In Figure 4, depending on linker structure, conducting was influenced highly. However, if the polyaniline homopolymers and their conductivity were measured together as a control sample, the conductivity of the resulting polymer films in this study could be understandable more.

Responses to the comments and suggestions from the reviewers (Manuscript ID: NCOMMS-16-07034-T)
Reviewer #1 (Remarks to the Author): We greatly appreciate the valuable comments (shown in italic font) and have provided responses (the highlighted sentences can be found in the revised manuscript).

Response:
We thank the reviewer for the valuable time spent in reviewing the manuscript. The polymerization simultaneously occurred 1) on the substrates, 2) in the solution (within the vials) and 3) on the walls of the vials (film state). However, it was not feasible to characterize the polymer precipitates that formed during polymerization (in the solution) or the product that formed on the walls of the vials. This type of polymerization on various substrates has been previously reported by others [1][2][3].
A new sentence has been added on page 4, and the relevant literature has been cited; these new citations can be found in the revised manuscript as references #21-23 in "References" section.   We observed that the resultant polymer possessed similar optical and electrical properties to those that have been reported by others in the case of pure polypyrrole (PPyoxidant) synthesized via oxidative polymerization [4,5]. Figure 1a shows the UV-Vis spectrum of PPy-oxidant, which indicates that the oxidant causes the polymer to become partially doped [5] with sulfate counter anions, and the XRD analysis results (Figure 1b) show the amorphous nature of the material. We also compared the electrical conductivity of PPy-oxidant with those of P(Py:MSA) and P(Py:EDSA:Py) and observed that it was dramatically reduced because of the absence of the protonic acid dopant ( Table 1).

Comment 3.
The To solve this issue the authors should prepare their films on conducting substrates.
Cyclic voltammetry and In-situ spectroelectrochemistry will give information about the neutral and the charged species. Only based on these data the observed red-shifts can be properly understood and explained. The spectra given in Figure S4 are not very clear and meaningful. This will also help the authors to be able to make statements about the polymerization mechanism and the nature of the crystalline films. It would be interesting to have statements about the degree of polymerization.

Response:
We thank the reviewer for this valuable suggestion. Regular growth in crystalline polymers such as ours can result in redshifts in the absorption spectra, as previously reported by others [6][7][8][9][10]. The electronic interaction between the stacked polymer chains in a crystalline conductive polymer causes its energy band gap to be narrower compared with that of an amorphous polymer, although the polymer backbone structure is the same in both cases.
Moreover, if the polymer forms a crystalline structure through the interdigitation of alkyl side chains or a more planar conformation of the backbone, redshifted absorption can occur even if the polymer has a charged structure [11]. Therefore, the P(Py:DSA:Py)s exhibited redshifted absorption spectra compared with that of the reference polymer, i.e., P(Py:MSA). We performed cyclic voltammetry on the ITO surface, as shown in Figure 2, to confirm the neutral or charged nature of P(Py:MSA) and P(Py:EDSA:Py). Oxidation and reduction peaks were observed for dedoped PPy because it is neutral. However, in the cases of P(Py:MSA) and P(Py:EDSA:Py), these peaks were absent and the current densities were higher because of their polaron and bipolaron states, thereby indicating that P(Py:MSA) and P(Py:EDSA:Py) have doped (oxidized or charged) structures. This behaviour has been previously reported [12]. We also analysed the UV-Vis spectra of the polymers (dedoped PPy, P(Py:MSA) and P(Py:EDSA:Py)) synthesized on glass substrates (Figure 3a), and the results further confirmed the neutral and charged natures of the films. The neutral PPy exhibited an absorption maximum at ~350 nm, which is related to the π-π* transition. By contrast,

Comment 4.
The reviewer is not convinced that "the DSA links caused the Ppy to be conductive" (page 4 last line) Ppy becomes conducting due to the oxidative polymerization used.

Response:
We thank the reviewer for pointing out this error. This statement was made in the context that the conductivities of the doped polymers (doped with protonic acids such as sulfonic acids) were higher than that of the polymer synthesized via oxidative polymerization using only the oxidant [4,14]. Therefore, the relevant sentence has been modified in the revised manuscript.

Original manuscript: (page 4, line 24)
"This finding confirmed that all Py units were connected and that the DSA links caused the PPy to be conductive."

Revised manuscript: (page 5, line 3)
"This finding confirmed that the Py units were linked by DSA connectors."

Comment 5.
The exact conditions of the van der Pauw method should be given. Errors of the measurements are needed as well.

Response:
The van der Pauw (VDP) method is a conventional method for evaluating the electrical properties of semiconductor materials, such as resistivity, carrier density, and mobility [15,16]. However, we measured only the sheet resistances of our samples using this technique.
The VDP method can be used to measure samples of arbitrary shape, although several basic sample conditions must be satisfied to obtain accurate measurements and avoid errors; for example, the thickness of the sample must be constant, the measurements must be performed by means of point contacts placed at the edges of the sample (as shown in Figure 4), and the sample quality must be homogeneous. Because the polymer chains in a conducting polymer are randomly distributed, the VDP method is most suitable in such cases and may provide more reliable data than any other methods, such as the four point-probe method and the four line-probe method, in measurements of the sheet resistance. Response: [redacted] [redacted] Comment 7. Figure S2: It is really hard to distinguish the bands in the FT-IR spectra.

Response:
We thank the reviewer for this observation. In the revised manuscript, Supplementary ordering. In the current manuscript, author has reported formation of crystal using same strategy, and the previous work has not been cited. Author has not discussed any performance improvement over the previous work. It suggests that this manuscript lags significantly in originality. In addition, manuscript has some technical flaws, which needed to be addressed before the work is submitted to any journal.
As a result, this study has no priority so that the contents are not suitable for publication in Nature Communications.

Response:
The reviewer's concerns are understandable. There is no doubt that the current work was inspired by our previous publication (Advanced Materials, 2012, 24, 3253-3257), in which we showed that an amorphous polymer, poly(2-vinylpyridine) (P2VP), when cross-linked with 1,4-dibromobutane, influences the crystallinity of P2VP, leading to sub-nanometre-level molecular ordering in the P2VP polymer chain; we have, indeed, cited this work (reference #17 in the "References" section in the original manuscript).

Figure 1 | Strategy difference between previous research and current research.
However, in the present work, the strategy is not similar as previous work, which is related to the crosslinking of P2VP with dibromobutane. In this approach, we have used the connected monomers as a precursor for polymerization rather than a single monomer ( Figure   1). When the single monomer is polymerized to form a linear chain, the chain direction is very likely to be randomized at each connection site due to an entropic effect, and therefore resulting structures are immensely entangled. When polymerization proceeds with a twomonomer-connected precursor (TMCP), which is not with a single monomer, four reactive sites of a TMCP (two from each monomer) introduces a constraint on the propagation direction, and prevents randomly oriented growth in monomeric or oligomeric state during the polymerization unlike single monomer case (Figure 1). Therefore, we produced highly crystalline conjugated polymers using TMCPs, yielding crystalline structures while inhibiting the entropically favourable chain entanglements. Our results regarding this novel polymerization strategy and morphological analysis in 3 dimensions can serve as the basis for new approaches to polymer crystallography.
On the basis of these merits compared with our previous work, we strongly believe that the current work has significant novelty and deserves publication.

Response:
We thank the reviewer for this suggestion. The synthesis and purification of the P(Py:MSA) and P(Py:DSA:Py) films via in situ chemical oxidative polymerization has already been described in the methods section of the manuscript in full detail on page 18.
The polymerization simultaneously occurred 1) on the substrate, 2) in the solution (within the vials) and 3) on the walls of the vials. To enable 1 H NMR measurements, we attempted to dissolve the polymer precipitates that formed in the solution, but as is known, polypyrrole does not easily dissolve in any organic solvent, especially in our case, in which it had a connected (cross-linked) structure; therefore, we were not able to perform this characterization. However, the valuable structural information of our polymers (which NMR can provide) has been discussed in the manuscript by FT-IR, UV-Vis and XPS analyses. We investigated the thermal properties of polymer samples (P(Py:MSA), P(Py:EDSA:Py) and P(Py:BPDSA:Py)) that had been freshly synthesized using the same methodology described in the manuscript. The powder samples were obtained through the purification of the precipitates that remained in the vials after polymerization. Figure 2a shows the thermal gravimetric analysis (TGA) results for all polymers, which began to degrade above 200 °C, consistent with the values previously reported in the literature for polypyrrole [1]. As seen from the thermograms, the degradation temperature (T d ) showed no change among the samples, but as the ordering of the molecular structures of the connectors increased from EDSA to BPDSA, they showed relatively higher thermal stability compared with MSA (the control sample). In the case of P(Py:BPDSA:Py), the thermal stability was the highest among the polymers above ~500 °C; this is because of the aromatic rings in the BPDSA connector. Figure 2b shows the differential scanning calorimetry (DSC) results for the polymers; here, little difference is evident among the thermograms. Broad endothermic bands were observed in all samples. Such bands have been regarded as a glass transition temperature (T g ) in some studies [2], although their nature is still not clear. However, they may be accompanied by the expulsion of residual water (<150 °C).

Comment 2.
In Figure 4, depending on linker structure, conducting was influenced highly. However, if the polyaniline homopolymers and their conductivity were measured together as a control sample, the conductivity of the resulting polymer films in this study could be understandable more.

Response:
As the reviewer suggests, it is very important to compare the electrical conductivities of our P(Py:DSA:Py)s with that of the homopolymer (control sample). Because this work is based on polypyrrole, we believe that it is not appropriate to compare the electrical conductivities of the P(Py:DSA:Py)s with those of polyaniline homopolymers. However, we have already discussed the differences in conductivity between the non-connected homopolymer (P(Py:MSA)) and the connected (cross-linked) polymers (P(Py:DSA:Py)s) in the original manuscript on page 7. 3.04 Table 1 shows the conductivity of P(Py:MSA), P(Py:DSA:Py)s and PPy-oxidant (from the polymerization of pure pyrrole using oxidant without any acid). The electrical conductivities of P(Py:DSA:Py)s were higher than that of P(Py:MSA) due to their crystallinity. The conductivity of PPy-oxidant was dramatically reduced compared to other polymers because of the absence of the protonic acid dopant.