Solvent Effects in the Regioselective N-Functionalization of Tautomerizable Heterocycles Catalyzed by Methyl Trifluoromethanesulfonate: A Density Functional Theory Study with Implicit Solvent Model
Congjie Zhang
Imre Bako
Leonardo Bernasconi
Round 1
Reviewer 1 Report
The reactions of the regioselective functionalization of N-center of tautomerizable benzoxazole, benzothiazole are very important, which can produce a new class of antibiotics. Interestingly, the yield of the functionalization of such reactions depends upon the solvent. In this contribution, the authors have investigated the reaction mechanism, the relationship between the yield and dieletric constant the using DFT calculation, This article could be published. However, the following things need to be addressed before it finally gets accepted for publication.
1. The data given in Table 1are not clear. The reactions should be described as the format of A®B. In addition, the transition structures associated with these reactions were necessary to be given. Why were the TS in solvent CAN and NIT not obtained?
2. All of molecules in Figures are not clear. These molecules can be depicted by Chemdraw or Gview software.
3. The reaction of 1a+2a have three intermediates. Can all intermediates produce products via TS? If not, only the intermediate involved the reaction was given.
4. Figures 5 and 6 were suggested to be amalgamated into single one.
Minor mistakes:
1. Yeld should be yield.
2. In table 1, the imaginary frequencies should be written as 127.27i, as so on.
3. In Figure 2, 1a and 2a should be in bold.
Author Response
Point 1: All of the calculation is performed carefully. Unfortunately the experimental evidence and the calculation with SMD correction can not correlated. I know that the negative results is important, but I do not think that in this case is enough. The barrier is larger for the system when the yield is larger. In this case something is not appropriate in the modell (reaction path, method, correction, solvent is incorporated the reaction... )
Response 1: The reviewer is correct, but the calculations performed in the present work can show the limitations to use the SMD model with implicit solvent molecules. In the conclusions of the corrected manuscript we present some sugestions to account for the problem, and indicate the more appropriate methods to that in account the solvente effects in the chemical reaction studied. The present work is the beginning of a more elaborate work in progress concerning the use of another methodology.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The manuscript seems to me rather interesting. The authors have done laborious computational work. Using the density functional theory, the authors presented a detailed study of the reaction mechanism of regioselective N-functionalization of tautomerizable heterocycles catalyzed by MeOTf in different solvents. In general, the manuscript presents a solid theoretical study. It does not contain serious drawbacks. In my opinion, the manuscript is well written; the methods are described in detail, which allows the reader to repeat all the results, if necessary. I would like to recommend to the authors add a more detailed description of quantum chemical descriptors, such as chemical potential, electrophilicity index, etc., to the "Computational Methodology" section. In my opinion, it would be helpful for the reader. Presented in the manuscript results seem solid, discussion, and main conclusions sound reasonable. On the presentation of the results, I have no critical remarks. Therefore, I think the manuscript can be accepted for publication in Computation.
Author Response
Point 1: In this submission to Computation, the authors use quantum calculations to explore catalysis reactions of methyl trifluoromethanesulfonate in the nucleophilic substitution of hydroxyl group of alcohols. Specifically, the authors use the B3LYP functional with Grimme's D3 dispersion to optimize the molecular geometries (further comments on their use of functional are given below). The authors calculate harmonic vibrational frequencies to characterize the stationary points on the potential energy surface. To ensure that all the stationary points were smoothly connected to each other, the authors use intrinsic reaction coordinate (IRC) analyses.
I find this manuscript to be of interest to computational chemists as well as readers of this journal. As such, I am generally in favor of publication with a few required revisions. In particular, it is well known that B3LYP (even with dispersion corrections) generally underestimates barrier heights compared to more accurate wavefunction-based CCSD(T) methods, which should be noted:
Phys. Chem. Chem. Phys., 2017, 19, 32184-32215
Nat. Commun. 2018, 9, 4733
Response 1: We acknowledge the rewiewer fot the comments.
Point 2: In particular, these prior works showed that B3LYP still underestimates barrier heights when compared to more accurate CCSD(T) methods. I am not necessarily asking the authors to carry out CCSD(T) calculations, but the authors should note these effects since they are well known. With these minor revisions, I would be willing to re-review this manuscript for possible publication in Computation.
Response 2: The reviewer comments was add in the conclusion of the manuscript were we present the limitations of the B3LYP functional and propose as a future work the use of more accurate ones, like the CCSD(T).
Author Response File: Author Response.pdf
Reviewer 3 Report
All of the calculation is performed carefully. Unfortunately the experimental evidence and the calculation with SMD correction can not correlated. I know that the negative results is important, but I do not think that in this case is enough. The barrier is larger for the system when the yield is larger. In this case something is not appropriate in the modell (reaction path, method, correction, solvent is incorporated the reaction... )
Author Response
Point 1: In this submission to Computation, the authors use quantum calculations to explore catalysis reactions of methyl trifluoromethanesulfonate in the nucleophilic substitution of hydroxyl group of alcohols. Specifically, the authors use the B3LYP functional with Grimme's D3 dispersion to optimize the molecular geometries (further comments on their use of functional are given below). The authors calculate harmonic vibrational frequencies to characterize the stationary points on the potential energy surface. To ensure that all the stationary points were smoothly connected to each other, the authors use intrinsic reaction coordinate (IRC) analyses.
I find this manuscript to be of interest to computational chemists as well as readers of this journal. As such, I am generally in favor of publication with a few required revisions. In particular, it is well known that B3LYP (even with dispersion corrections) generally underestimates barrier heights compared to more accurate wavefunction-based CCSD(T) methods, which should be noted:
Phys. Chem. Chem. Phys., 2017, 19, 32184-32215
Nat. Commun. 2018, 9, 4733
Response 1: We acknowledge the rewiewer fot the comments.
Point 2: In particular, these prior works showed that B3LYP still underestimates barrier heights when compared to more accurate CCSD(T) methods. I am not necessarily asking the authors to carry out CCSD(T) calculations, but the authors should note these effects since they are well known. With these minor revisions, I would be willing to re-review this manuscript for possible publication in Computation.
Response 2: The reviewer comments was add in the conclusion of the manuscript were we present the limitations of the B3LYP functional and propose as a future work the use of more accurate ones, like the CCSD(T).
Author Response File: Author Response.pdf
Reviewer 4 Report
The manuscript describes a computational study of the mechanism of N-functionalization of heterocycle molecules in the presence of methyl trifluoromethanesulfonate as a catalyst in a series of solvents. The authors use hybrid (B3LYP) density functional theory calculations to study the structure and energetics of the reactants, transition states and products. The presence of the solvents is accounted for using an implicit solvation model. The results provide insight into the mechanism and free-energy profiles of the reactions. At the same time, they highlight some of the limitations of the implicit treatment of solvation adopted in the work, which are particularly evident when the parameters of the solvation models are similar for different solvents. This shows that implicit solvation does not represent with sufficient accuracy the interaction between the solutes, solvent environment and the internal microscopic dynamics of the solvent.
Despite this "negative" result, the manuscript makes a valuable contribution to the study of this class of catalytic reactions in solvents from a computational point of view and may serve as a starting point for future and more elaborate treatments of solvent effects. The work is described clearly and the results are well supported by the calculations. I recommend publication subject to the minor points listed below.
1) Since the topic of implicit solvation is central in the work, I suggest adding it to the title and replace "A Density Functional Theory Study" with "A Density Functional Theory Study with Implicit Solvation".
2) On page 1, line 7 from bottom, replace "the reaction rate increase" with "the reaction rate increases".
3) On page 2, line 12 from bottom, replace "Beck" with "Becke" before reference 12.
4) On page 3, line 5 from the end of Section 2, replace "To characterization of the reaction mechanism were calculated the following electronic properties" with "To characterize the reaction mechanism, the following electronic properties were calculated".
5) Page 3, line 4 of Section 3, replace "And they associated to the corresponding transition states are given in Table 1" with "The corresponding transition state properties are given in Table 1".
6) On page 4 and elsewhere in the paper replace "yeld" with "yield".
7) At the end of the Conclusion, it should be mentioned that, in addition to hybrid models of solvation, in which a small number of solvent molecules are treated explicitly while the solvent long-range effects are implicit, fully explicit simulation of solvent effects is nowadays possible using density functional based ab initio molecular dynamics techniques.
Author Response
Point 1: In this submission to Computation, the authors use quantum calculations to explore catalysis reactions of methyl trifluoromethanesulfonate in the nucleophilic substitution of hydroxyl group of alcohols. Specifically, the authors use the B3LYP functional with Grimme's D3 dispersion to optimize the molecular geometries (further comments on their use of functional are given below). The authors calculate harmonic vibrational frequencies to characterize the stationary points on the potential energy surface. To ensure that all the stationary points were smoothly connected to each other, the authors use intrinsic reaction coordinate (IRC) analyses.
I find this manuscript to be of interest to computational chemists as well as readers of this journal. As such, I am generally in favor of publication with a few required revisions. In particular, it is well known that B3LYP (even with dispersion corrections) generally underestimates barrier heights compared to more accurate wavefunction-based CCSD(T) methods, which should be noted:
Phys. Chem. Chem. Phys., 2017, 19, 32184-32215
Nat. Commun. 2018, 9, 4733
Response 1: We acknowledge the rewiewer fot the comments.
Point 2: In particular, these prior works showed that B3LYP still underestimates barrier heights when compared to more accurate CCSD(T) methods. I am not necessarily asking the authors to carry out CCSD(T) calculations, but the authors should note these effects since they are well known. With these minor revisions, I would be willing to re-review this manuscript for possible publication in Computation.
Response 2: The reviewer comments was add in the conclusion of the manuscript were we present the limitations of the B3LYP functional and propose as a future work the use of more accurate ones, like the CCSD(T).
Author Response File: Author Response.pdf
Reviewer 5 Report
After carefully reading the manuscript I found that the research design and findings are original and important.
However, the following are a few suggestions that may be useful to improve the reader's experience of the manuscript.
1) Figure axis names seem to be too small and it may be better to increase the text size.
2) Details inside Figures 1 and 7 are not clear enough to read. Consider updating the figures in a way the details inside can be read easily.
Author Response
Point 1: In this submission to Computation, the authors use quantum calculations to explore catalysis reactions of methyl trifluoromethanesulfonate in the nucleophilic substitution of hydroxyl group of alcohols. Specifically, the authors use the B3LYP functional with Grimme's D3 dispersion to optimize the molecular geometries (further comments on their use of functional are given below). The authors calculate harmonic vibrational frequencies to characterize the stationary points on the potential energy surface. To ensure that all the stationary points were smoothly connected to each other, the authors use intrinsic reaction coordinate (IRC) analyses.
I find this manuscript to be of interest to computational chemists as well as readers of this journal. As such, I am generally in favor of publication with a few required revisions. In particular, it is well known that B3LYP (even with dispersion corrections) generally underestimates barrier heights compared to more accurate wavefunction-based CCSD(T) methods, which should be noted:
Phys. Chem. Chem. Phys., 2017, 19, 32184-32215
Nat. Commun. 2018, 9, 4733
Response 1: We acknowledge the rewiewer fot the comments.
Point 2: In particular, these prior works showed that B3LYP still underestimates barrier heights when compared to more accurate CCSD(T) methods. I am not necessarily asking the authors to carry out CCSD(T) calculations, but the authors should note these effects since they are well known. With these minor revisions, I would be willing to re-review this manuscript for possible publication in Computation.
Response 2: The reviewer comments was add in the conclusion of the manuscript were we present the limitations of the B3LYP functional and propose as a future work the use of more accurate ones, like the CCSD(T).
Author Response File: Author Response.pdf
Reviewer 6 Report
In this submission to Computation, the authors use quantum calculations to explore catalysis reactions of methyl trifluoromethanesulfonate in the nucleophilic substitution of hydroxyl group of alcohols. Specifically, the authors use the B3LYP functional with Grimme's D3 dispersion to optimize the molecular geometries (further comments on their use of functional are given below). The authors calculate harmonic vibrational frequencies to characterize the stationary points on the potential energy surface. To ensure that all the stationary points were smoothly connected to each other, the authors use intrinsic reaction coordinate (IRC) analyses.
I find this manuscript to be of interest to computational chemists as well as readers of this journal. As such, I am generally in favor of publication with a few required revisions. In particular, it is well known that B3LYP (even with dispersion corrections) generally underestimates barrier heights compared to more accurate wavefunction-based CCSD(T) methods, which should be noted:
Phys. Chem. Chem. Phys., 2017, 19, 32184-32215
Nat. Commun. 2018, 9, 4733
In particular, these prior works showed that B3LYP still underestimates barrier heights when compared to more accurate CCSD(T) methods. I am not necessarily asking the authors to carry out CCSD(T) calculations, but the authors should note these effects since they are well known. With these minor revisions, I would be willing to re-review this manuscript for possible publication in Computation.
Author Response
Point 1: In this submission to Computation, the authors use quantum calculations to explore catalysis reactions of methyl trifluoromethanesulfonate in the nucleophilic substitution of hydroxyl group of alcohols. Specifically, the authors use the B3LYP functional with Grimme's D3 dispersion to optimize the molecular geometries (further comments on their use of functional are given below). The authors calculate harmonic vibrational frequencies to characterize the stationary points on the potential energy surface. To ensure that all the stationary points were smoothly connected to each other, the authors use intrinsic reaction coordinate (IRC) analyses.
I find this manuscript to be of interest to computational chemists as well as readers of this journal. As such, I am generally in favor of publication with a few required revisions. In particular, it is well known that B3LYP (even with dispersion corrections) generally underestimates barrier heights compared to more accurate wavefunction-based CCSD(T) methods, which should be noted:
Phys. Chem. Chem. Phys., 2017, 19, 32184-32215
Nat. Commun. 2018, 9, 4733
Response 1: We acknowledge the rewiewer fot the comments.
Point 2: In particular, these prior works showed that B3LYP still underestimates barrier heights when compared to more accurate CCSD(T) methods. I am not necessarily asking the authors to carry out CCSD(T) calculations, but the authors should note these effects since they are well known. With these minor revisions, I would be willing to re-review this manuscript for possible publication in Computation.
Response 2: The reviewer comments was add in the conclusion of the manuscript were we present the limitations of the B3LYP functional and propose as a future work the use of more accurate ones, like the CCSD(T).
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors are commended for addressing the suggestions. The revised manuscript addresses all my queries.Reviewer 3 Report
I think my point concerning to the manuscript was correct, but I look the other two comments and answers I accept this paper to publication. Especially important to do in the future some high level CCSD(T) caclulation (ORCA or MRCC is avialble) to check the accuracy of B3LYP.
