Two-dimensional amine and hydroxy functionalized fused aromatic covalent organic framework

Ordered two-dimensional covalent organic frameworks (COFs) have generally been synthesized using reversible reactions. It has been difficult to synthesize a similar degree of ordered COFs using irreversible reactions. Developing COFs with a fused aromatic ring system via an irreversible reaction is highly desirable but has remained a significant challenge. Here we demonstrate a COF that can be synthesized from organic building blocks via irreversible condensation (aromatization). The as-synthesized robust fused aromatic COF (F-COF) exhibits high crystallinity. Its lattice structure is characterized by scanning tunneling microscopy and X-ray diffraction pattern. Because of its fused aromatic ring system, the F-COF structure possesses high physiochemical stability, due to the absence of hydrolysable weak covalent bonds.

4. The pXRD spectrum of the framework material shown in Figure 2 exhibits very broad peaks, where relative peak heights between experimental and simulated spectra are not in full agreement. The authors should take care to rationalize deviations in relative peak heights obtained experimentally from the simulated spectrum. The authors should also comment to acknowledge the reasons behind the broadness of peaks in the spectrum (poor crystallinity, small crystal size, etc.) in more detail. In particular, it is unclear what the authors mean by "extremely large molecular size of the F-COF" as rationale for the broadness in peaks in the context of the Scherrer equation relating crystallite size to peak broadening. 5. Did the authors undertake any reaction optimization for obtained the fused COF material? It is quite common to implement extensive reaction optimization to achieve and optimize COF synthesis. Any reaction optimization work should be reported, and comments should be made about important experimental steps, additives, etc that are critical to obtaining crystalline material. 6. Figure 1 seems to suggest an alternating arrangement of -OH and -NH2 moieties within each pore of the framework. The authors should comment what evidence (if any) this structural model is based on and if other arrangements are possible. Furthermore, some rationale for the molecular designed of these functional group moieties should be provided. Is -OH substitution expected to provide certain synthetic or functional advantages over hexamine substitution? 7. The authors should take care to describe which instruments (brand, make, model) of each instrument were used to obtain various characterizations. All key settings of the experimental conditions and sample preparation for analysis should be given. This information is essential for ensuring the reproducibility of the work and the utility of this work to others.
8. Characterization of the material by SEM and TEM would be helpful in understanding the bulk morphology and long range order of the framework. STM only show small fragments and do not help in ascertaining long-range order or bulk morphology of the material.
9. The text in the experimental section of SI "Important Note regarding STM" should be refined to make it more clear and improve English language usage. How are the samples prepared for STM? Is there sonication and dropcasting involved? What are the key instrument settings for obtaining the images?
10. What is the red trace in Figure S1? Please label.
Reviewer #2 (Remarks to the Author): This work describes the formation of pyrazine consists of a reversible amine/ketone addition and an irreversible ring closure. The value of the work is that the product is much more conjugated than normal Schiff-base COF, so this represents a significant advancement.
One point that needs attenton is that the crystallinity is more like a polymer from powder XRD. Can the authors comment on this? compared to the usual COF made the XRD peaks of this is broad, should it be closer to a polymer when crystallized in bulk form.
However on-surface synthesis shows that 2-D COF can be formed locally. The STM result is reasonably good among all published 2D polymer grown in STM, in terms of uniformity and size. Such dehydration reaction is usually not suitable for on-surface growth, so this result is excellent and the work can be published. At least, it provides a design guiding principle for future researchers.
Reviewer #3 (Remarks to the Author): In this manuscript, the authors synthesize a covalent organic framework featuring fused aromatic rings via irreversible condensation. This pyrazine-based COF has been characterized with different technique (PXRD, TGA, XPS, FT-IR, STM etc.) and exhibits a high physiochemical stability owing its fused aromatic ring system. After reading through the whole manuscript carefully, I think it is not suitable to publish in Communications Chemistry as it does not meet the novelty criteria and the studies are not sufficient. 1. Compared with other pyrazine-based COFs (J. Am. Chem. Soc. 2019, 141, 16810., ref. 42), the quality of PXRD pattern is poor, and the values of Rwp and Rp are not reasonable. 2. Compared with the paper that the authors published in Nat. Commun. previously, this manuscript does not have too much advantage and novelty. 3. COF is crystalline polymer and not easy to sublimate. For the STM experiment, the authors have claimed that "They rigorously evaporate and sublimate.". Why? Are they oligomers? Or they are the mixture of COFs and oligomers? 4. The Solid state NMR of the COF should be provided to analyze the structure of F-COF. 5. There are some mistakes in the manuscript and SI, such as Supplementary Fig. 1, weight loss is wrong.
Reviewers' comments: ******************************************* Reviewer #1 (Remarks to the Author): ****************************************************** This paper describes an interesting structural analog of an aza-fused covalent organic framework with good crystallinity derived from the condensation pentaaminophenol and hexaketocyclohexane. The manuscript represents an important advance in aza-fused framework syntheisis, but has several shortcomings that should be carefully addressed by the authors: Comment 1.1. The statement put forward by the authors "Thus far, only two pyrazine based crystalline COF structures have been reported: one was prepared by solution process11 and the other employed solvothermal synthesis.42" is incorrect as it overlooks several significant developments in this class of materials listed below. The authors should be careful to conduct a thorough literature review to ensure that their claims are correct and carefully account for and acknowledge the recent developments in the field. Response 1.1. We highly appreciate the reviewer #1 for thoughtful evaluation and valuable suggestions to improve the literature survey and quality of the manuscript. We are sorry for missing some important references related to this work. As a matter of fact, the manuscript was written before the recent pointed references were published online. We have now added the underlined references in the revised manuscript with corrections of the related text. Comment 1.2. In light of the previous comment, the title of the work is too generic, given that there are several other reports of aza-fused frameworks available. The authors should make their title more specific and descriptive to reflect their specific contribution to this field. Response 1.2. We agree with the reviewer and based on the reviewer's suggestion; we have modified the title (Two-dimensional amine and hydroxy functionalized fused aromatic covalent organic framework) to make it more understandable for the readers. Comment 1.3. The synthetic scheme presented in Figure 1 should list all of the reagents, additives, and solvents and key reaction conditions (such as temperature, pressure, reaction time, etc) for the optimized chemical transformation shown to quickly highlight these aspects for the reader. Furthermore, a detailed procedure for the COF synthesis should be included into the SI. As it stands, the article currently does not give an experimental procedure for the synthesis of the COF material that would be sufficient for reproducing the work. Response 1.3. Thanks to the reviewer for the constructive suggestion. We made necessary changes in Figure 1 and added the reaction conditions. Detailed experimental procedure is given in the Methods section. The synthesis procedures related to monomer are given in the SI along with characterizations.
Comment 1.4. The pXRD spectrum of the framework material shown in Figure 2 exhibits very broad peaks, where relative peak heights between experimental and simulated spectra are not in full agreement. The authors should take care to rationalize deviations in relative peak heights obtained experimentally from the simulated spectrum. The authors should also comment to acknowledge the reasons behind the broadness of peaks in the spectrum (poor crystallinity, small crystal size, etc.) in more detail. In particular, it is unclear what the authors mean by "extremely large molecular size of the F-COF" as rationale for the broadness in peaks in the context of the Scherrer equation relating crystallite size to peak broadening.
Response 1.4. We agree with the reviewer that the XRD spectrum reveals broad peaks. The broad XRD peaks are very common in the irreversible reaction driven synthesis of organic framework. With fast growing molecular weight in solution because of large energy gain (aromatization), it is difficult to regularly stack in a long-range planarly ordered manner due to difficulty in the movement (kinetics) of the large flakes with ripples and wrinkles. We also agree with the reviewer the broad spectrum could be related to poor crystallinity and small crystal size. However, in this case, from the SEM, TEM and STM (only small size was delaminated and observed images), we can see the size of the flakes formed large enough due to huge thermodynamic energy gain (aromatization).
Straight forward application of Scherrer equation, which is limited for nano-scale crystallites. For polymer structure, it is not very simple due to polycrystallinity. Because the shape of crystallites is usually irregular, most of the applications of the Scherrer analysis assume spherical crystallite shapes. The peak broadening can result from the non-uniform lattice distortions, dislocations, mixture of crystalline phases and grain boundaries. It is also suggested that the internal pressure exerted by the surface tension on the nanomaterial will create a stress field to trigger lattice strain, in case of small crystallite size (Journal of Theoretical and Applied Physics 2014, 8, 123-134). Comment 1.5. Did the authors undertake any reaction optimization for obtained the fused COF material? It is quite common to implement extensive reaction optimization to achieve and optimize COF synthesis. Any reaction optimization work should be reported, and comments should be made about important experimental steps, additives, etc that are critical to obtaining crystalline material. Response 1.5. Capitalizing on the experience we already had in the synthesis of irreversible reaction based organic frameworks, we were able to synthesize this material. One thing we found was the addition of second monomer at low temperature to control the very quick reaction between the diketone and diamine to form pyrazine type rings. In this case, TFMSA (melting point: −40 °C) was used as a solvent and cooling bath of −40 °C was employed to increase the crystallinity by controlling the temperature. No other optimizations were conducted in this regard. Comment 1.6. Figure 1 seems to suggest an alternating arrangement of -OH and -NH2 moieties within each pore of the framework. The authors should comment what evidence (if any) this structural model is based on and if other arrangements are possible. Furthermore, some rationale for the molecular designed of these functional group moieties should be provided. Is -OH substitution expected to provide certain synthetic or functional advantages over hexamine substitution? Response 1.6. Thanks to the reviewer for the insightful comment and pointing out very important issue. It was really hard to differentiate the (−NH 2 and −OH) functional groups by any means including STM, as their van der Waals radii are very similar each other. Occurrence of (−NH 2 and −OH) functional groups is important for selective functionalization of the material for further applications. Like attaching long alkyl chain for better solubility in different organic solvents and to modulate the electronic properties of the material ( Figure  R1). Figure R1│Possibility of the functionalization of (−NH 2 and −OH) functional groups. Comment 1.7. The authors should take care to describe which instruments (brand, make, model) of each instrument were used to obtain various characterizations. All key settings of the experimental conditions and sample preparation for analysis should be given. This information is essential for ensuring the reproducibility of the work and the utility of this work to others. Response 1.8. Thanks to the reviewer for the constructive suggestion. We have given the SEM and TEM analysis for the reviewer's consideration. The SEM analysis of the F-COF displayed the 2D structural arrangements (Figure R2), showing the stacking of the layers. As the reviewer may understand, it is very difficult to check the crystallinity of porous COFs by TEM, in which the electron beam quickly deforms their structures (beam damages) and the images look like less crystalline ( Figure R3). Moreover, due to poor stacking because of the irreversibility of the reaction, it was not possible to resolve its structure using TEM (Nat. Commun. 2015, 6, 6486;Chem. Mater. 2019, 31, 819−825;J. Am. Chem. Soc. 2019, 141, 11929−11937;J. Am. Chem. Soc. 2019, 141, 16810−16816). That is why, we resorted to scanning tunneling microscopy (STM) for atomic resolution of the structure, which was much more difficult and time consuming.  Comment 1.9. The text in the experimental section of SI "Important Note regarding STM" should be refined to make it more clear and improve English language usage. How are the samples prepared for STM? Is there sonication and dropcasting involved? What are the key instrument settings for obtaining the images?
Response 1.9. Thanks to the reviewer for highlighting the poor writing issue. Now, we have modified the text to make it more understandable for the readers. The sample preparation procedure is already given in the 'Methods section' of the manuscript. In short, the powder sample was directly loaded on the sample stage of the STM and sublimized in situ inside the chamber and characterized.
Comment 1.10. What is the red trace in Figure S1? Please label.
Response 1.10. Sorry for the confusion. The red trace was related to heat flow changes with respect to temperature. To avoid confusion, we have removed the red trace in the TGA thermogram.
Comment 2.0. One point that needs attenton is that the crystallinity is more like a polymer from powder XRD. Can the authors comment on this? compared to the usual COF made the XRD peaks of this is broad, should it be closer to a polymer when crystallized in bulk form.
However on-surface synthesis shows that 2-D COF can be formed locally. The STM result is reasonably good among all published 2D polymer grown in STM, in terms of uniformity and size. Such dehydration reaction is usually not suitable for on-surface growth, so this result is excellent and the work can be published. At least, it provides a design guiding principle for future researchers.
Response 2.0. We greatly appreciate the reviewer #2 for his/her thoughtful evaluation and recognition of the importance of this work. As a matter of fact, we had the same opinion with the reviewer, because of the presence of broad XRD peaks. However, after resolving structure using STM, we concluded that we were able to produce a right material and the XRD peak broadening should be associated with not poor ordering but poor packing. Sorry for the confusion, we did not prepare the structure by reacting the monomers the on the surface. Instead, we prepared the structure in solution and later resolved the structure on the Cu(111) substrate by subliming the prepared material under ultrahigh vacuum chamber.