From the journal Digital Discovery Peer review history

08-Apr-2024

Dear Dr Rzepa:

Manuscript ID: DD-ART-12-2023-000246
TITLE: Modelling kinetic isotope effects for SWERN Oxidation. DFT-based Transition State Theory is OK.

Thank you for your submission to Digital Discovery, published by the Royal Society of Chemistry. I sent your manuscript to reviewers and I have now received their reports which are copied below.

I have carefully evaluated your manuscript and the reviewers’ reports, and the reports indicate that major revisions are necessary.

Please submit a revised manuscript which addresses all of the reviewers’ comments. Further peer review of your revised manuscript may be needed. When you submit your revised manuscript please include a point by point response to the reviewers’ comments and highlight the changes you have made. Full details of the files you need to submit are listed at the end of this email.

Digital Discovery strongly encourages authors of research articles to include an ‘Author contributions’ section in their manuscript, for publication in the final article. This should appear immediately above the ‘Conflict of interest’ and ‘Acknowledgement’ sections. I strongly recommend you use CRediT (the Contributor Roles Taxonomy, https://credit.niso.org/) for standardised contribution descriptions. All authors should have agreed to their individual contributions ahead of submission and these should accurately reflect contributions to the work. Please refer to our general author guidelines https://www.rsc.org/journals-books-databases/author-and-reviewer-hub/authors-information/responsibilities/ for more information.

Please submit your revised manuscript as soon as possible using this link:

*** PLEASE NOTE: This is a two-step process. After clicking on the link, you will be directed to a webpage to confirm. ***

https://mc.manuscriptcentral.com/dd?link_removed

(This link goes straight to your account, without the need to log on to the system. For your account security you should not share this link with others.)

Alternatively, you can login to your account (https://mc.manuscriptcentral.com/dd) where you will need your case-sensitive USER ID and password.

You should submit your revised manuscript as soon as possible; please note you will receive a series of automatic reminders. If your revisions will take a significant length of time, please contact me. If I do not hear from you, I may withdraw your manuscript from consideration and you will have to resubmit. Any resubmission will receive a new submission date.

The Royal Society of Chemistry requires all submitting authors to provide their ORCID iD when they submit a revised manuscript. This is quick and easy to do as part of the revised manuscript submission process.   We will publish this information with the article, and you may choose to have your ORCID record updated automatically with details of the publication.

Please also encourage your co-authors to sign up for their own ORCID account and associate it with their account on our manuscript submission system. For further information see: https://www.rsc.org/journals-books-databases/journal-authors-reviewers/processes-policies/#attribution-id

I look forward to receiving your revised manuscript.

Yours sincerely,
César de la Fuente, Ph.D.
Associate Editor, Digital Discovery

************

Reviewer 1

The kinetic isotope effects for SWERN oxidation, studied in a previous paper (Ref. 9). are revisited in this article. Taking the previous study as a starting point, the calculation methods are improved to eliminate possible sources of error. In this way, they conclude that the discrepancies between theory and experiment are caused by the calculation levels rather than transition state theory, as claimed by the original paper's authors. In addition, supplementary information is provided according to the FAIR guidelines, making it easier to replicate these calculations in future work.
</p>
The manuscript is informative and interesting to a wide audience; therefore, I recommend its publication. However, some changes are recommended:
</p>
1. Tunneling. The authors mention that tunneling is computed using a very simple approximation, Bell’s approach for parabolic barriers. However, as they also mention, the curvature of Figure 7 seems to suggest that tunneling is significant, given the low temperatures. I believe that further discussion on this point would be insightful. Comparing Bell’s approximation to more sophisticated methods (even as simple as fitting the top of the barrier to Eckart’s potential) would evaluate the validity of the present results. This is the main drawback of the paper because other approximations made might compensate each other as computing kinetic isotope effects involves ratios, but tunneling does not cancel out.
</p>
2. Internal rotations. Harmonic approximations are not appropriate when dealing with hindered rotations and complex conformational changes. I advise the authors to further discuss this and, ideally, verify whether partition function ratios do not change significantly when moving from the simple harmonic partition function to more complicated anharmonic functions that take internal rotations into account. In addition, it is unclear why the scans in Figure 3 do not give the same value of the energy for -160 and 200, and Panel b shows a noisy behavior around 175 degrees.

Reviewer 2

review attached

Reviewer 3

Rzepa and coworkers reported a very systematic computational/DFT (B3LYP, B3LYP+D, wB97X-D, B2PLYP-D3 method) study to re-visit one previous experimental and computational KIE study on SWERN oxidation (ref. 9). The current systematic study attempts to understand and resolve the previously claimed discrepancy between the measured and computed KIEs in ref. 9 by examining effects on computational settings, solvent, basis sets, DFT functionals and dispersion correction on the computed KIE values. Notably and surprisingly, larger basis sets were found to be essential to match/resolve the measured and computed KIEs based on transition state theory. They also suggest that using a substituent of NMe2 can give a larger KIE value for the future experimental verification. Extensive data generated in this study are collected/stored in FAIR data mode, rather than conventional ESI file. Overall, this systematic study give interesting and less-common method comparison about KIE calculations, which should be useful in computational chemistry community. I would recommend this work after the authors address below minor issues. However, I am not sure whether this work fits themes of Digital Discovery, in which I found most papers relates to AI/ML. I let the editor to make a decision in this point.

1. If possible, add a specific base name in Scheme 1.

2. As to “Replication of Energies”, it should be helpful for the readers to follow if they put their obtained values using different setttings in a Table(s).

3. Fig. 3 (x/y-axis), it may be better to use element labels better than atom index? e.g., C1-O-S-C2? Moreover, “Species 2 (endo) has a C-O torsion angle of ~10° and species 3 (exo) ~165°” I suggest the authors to use labels or highlight the two values for 2 & 3 in Fig. 3.

4. wB97XD --> wB97X-D

5. What is the main problem for underestimating KIE values when a small basis set was used? A problem of weak non-covalent interactions? Could the authors add some discussion?

6. Accurate prediction or reproduction of a (intramolecular) primary (H/D) KIE value by computational methods is quite changeling. Tunnelling could have some influence. Instead, I would propose to perform parallel secondary (H/D) KIE using Ph-CH2-OS-R and Ph-CD2-OS-R substrate. Generally, secondary (H/D) KIE value is smaller and can be reliably evaluated by computational methods.

Referee: 1

The kinetic isotope effects for SWERN oxidation, studied in a previous paper (Ref. 9). are revisited in this article. Taking the previous study as a starting point, the calculation methods are improved to eliminate possible sources of error. In this way, they conclude that the discrepancies between theory and experiment are caused by the calculation levels rather than transition state theory, as claimed by the original paper's authors. In addition, supplementary information is provided according to the FAIR guidelines, making it easier to replicate these calculations in future work.

The manuscript is informative and interesting to a wide audience; therefore, I recommend its publication. However, some changes are recommended:
1. Tunneling. The authors mention that tunneling is computed using a very simple approximation, Bell’s approach for parabolic barriers. However, as they also mention, the curvature of Figure 7 seems to suggest that tunneling is significant, given the low temperatures. I believe that further discussion on this point would be insightful. Comparing Bell’s approximation to more sophisticated methods (even as simple as fitting the top of the barrier to Eckart’s potential) would evaluate the validity of the present results. This is the main drawback of the paper because other approximations made might compensate each other as computing kinetic isotope effects involves ratios, but tunneling does not cancel out.

Response:
We will incorporate further discussion of this aspect in a follow up article. As implied when citing Ref 57, we also intend to explore effects which include non-Born-Oppenheimer behaviour of the transferring proton. We have already discussed this with a collaborator, but it is not possible to start this project immediately within the timescale of the present article. We have expanded the statement associated with Ref 57 to indicate this.

2. Referee: 1 Internal rotations. Harmonic approximations are not appropriate when dealing with hindered rotations and complex conformational changes. I advise the authors to further discuss this and, ideally, verify whether partition function ratios do not change significantly when moving from the simple harmonic partition function to more complicated anharmonic functions that take internal rotations into account.

Response:
The low-frequency modes in the transition states are far from simple hindered rotors. Also the Bigeleisen equations are derived from partition function ratios for harmonic modes. We intend to include this aspect in the follow up as noted above exploring non-BO behaviour. The current focus of the present article was primarily to both address some of the issues revealed in replicating the original work and to identify the basis set quality as a clear contributor to the anomalous results previously reported. It was not focussed on exploring improved computational models for kinetic isotope effects using this system as an example, which we consider as being part of the scope of a follow up to the present article.

Referee: 1 In addition, it is unclear why the scans in Figure 3 do not give the same value of the energy for -160 and 200, and Panel b shows a noisy behaviour around 175 degrees.

Response:

The referee has missed the observation that the dihedral 16-15-12-13 is strongly correlated with the 12-15-16-17 dihedral. This correlation means that the scan of the latter does not return the system to the starting conformation, which is why the energies at -160 and +200 are not the same and which is why panel b was included to illustrate this. The noisy behaviour at around 175 again illustrates how strongly the two dihedral angles are correlated, resulting from effects introduced by round-off errors in the geometry optimisations. Discussion to make this point clear has been added.

Referee: 2

This useful paper investigates some apparent anomalies in an earlier computational study of the title reaction by Giagou & Meyer (GM), correcting previous deficiencies in the methodology, and highlighting the importance of better methods for data deposition to facilitate reproducibility of results in computational chemistry. However, in its present form it is not an easy read. The authors (BLR) rightly note a confusing transposition of labels by GM but it seems to me that they add further confusion in Scheme 1 of the present manuscript.
It is evident from inspection of the deposited data (yes, it works – thank you!) that species 2 (endo) and 3 (exo) are indeed conformers for rotational about the C-O, as stated in Section 3. However, they are not epimers differing in stereochemical configuration around the S atom, as they both have the (R) configuration: Scheme 1 is incorrectly labelled in this respect.

Response:
The origin of this confusion is not scheme 1 itself, but its application to exploring the rotamers of 2 and 3. The scheme refers to 2 and 3 as diastereomers resulting from isotopic substitution, whereas the conformational exploration is conducted without isotopic substitution, using the (R)-enantiomer for both conformations. The scheme has been corrected to indicate that substituent X can be either X=D (for isotope effect analysis) or X=H (for conformational analysis). The caption to Figure 3 now shows the configuration at the sulfur, clarifying that only the (R)-enantiomer is used for this analysis - the energy profiles for the (S)-enantiomer being of course identical to that of the (R)-enantiomer.

Referee: 2 The exo reactant is favoured (B3LYP/def2QZVPP, sum of electronic and thermal free energies G) by 0.35 kJ mol-1 (298 K) over the endo reactant, but this has nothing to do with any difference in energies of epimers caused by S/R inversion at S; Scheme 1 is also incorrect in this respect.
Products 4 and 5 are indicated as being obtained via endo steps k1 & k2 and exo steps k3 & k4, implying that there are only two distinct transition structures. Although there are indeed only two distinct first-order saddle points on the potential energy surface governing product formation from species 2 and 3, when isotopic substitution at the diastereotopic benzylic position is taken into account there are four distinct transition structures (an exo and an endo each leading to 4, and likewise leading to 5). The situation is, of course, further complicated by the existence of enantiomers. In my opinion, Scheme 1 must be redrawn to be correctly labelled.

Response: Done

Referee: 2 In contrast, GM’s Scheme 3 and Figure 3 indicate structure 2 yielding product 4 via step k4 with an endo-TS and product 5 via step k1 with an exo-TS, and structure 3 yielding product 4 via step k2 with an exo-TS and product 5 via step k3 with an endo-TS. All very confusing! Similarly, both the endo and exo transition structures have the same (R) configuration around the S atom: in its present form, it is not obvious that the paper has anything to say on the topic of S/R inversion at S and on ‘forbidden inversion’ (kinv >> k1 or k2) or Curtin-Hammett conditions (kinv << k1 or k2). GM assumed that 2 and 3 would be formed in equal amounts from 1, which would imply that the equilibrium constant for epimerisation between 2 and 3 were equal to unity. If that were true, then it would also imply that k1 = k3 and k2 & k4, rendering the same simple expression for the intramolecular KIE (eq 2) for both forbidden inversion and Curtin-Hammett cases. However, since species 2 and 3 in the present work are not epimers, and the energies of the epimeric species do not appear to be provided, it is not obvious to this reviewer what the present work can say on this question.

Response:
The core analysis of Scheme1 is that in the presence of isotopic substitution at carbon, there are two asymmetric centres in the system, which leads to 22 = 4 stereoisomers, which occur as pairs of enantiomers. Since the latter have the same energies, this means that there are only two distinct transition states to be located, which are labelled as endo and exo by both the original authors GM and ourselves. We have also identified that the reactants have two low energy conformations, again labelled endo or 2 and exo or 3. We now include a succinct summary of this in the article.

For the purpose of replicating the original study, we have used 2 as the reactant (as shown in Scheme 1 for endo-TS) in Table 1. For selected examples only (B3LYP+GD3+BJ/Def2-QZVPP and B2PLYP-D3/Def2-QZVPP) we also report in Table 1 the calculated KIE for the individual endo and exo-TS, using both reactant conformers, thus covering all the stereochemical combinations. The reactant conformer dependency is small (KIE = 2.106 using rotamer 3 and 2.066 using rotamer 2 for B3LYP+GD3+BJ/Def2-QZVPP and respectively KIE = 2.213/2.215 for B2PLYP-D3/Def2-QZVPP). The individual contributions of endo-TS and exo-TS to the KIE are then combined using equation 1 (which includes a weighting derived from the relative free energies of the two transition state isomers) and the results are then discussed in section 3.4.1 of the article and elsewhere. We consider this fully addresses the concerns made by the referee above.

Referee: 2 The energies of the exo and endo TSs are clearly not the same, but BLR present results in the deposited data only for non-deuterated species and they do not present any details of their KIE calculations. In the spirit of accessibility that they are very rightly championing, it would be very helpful to include results for each isotopologously distinct species.

Response:
We are grateful to the referee for recognising the spirit of identifying both the accessibility and findability (FAIR) of the data. To help readers find such data, we have devised the following search query to identify the data the referee is requesting, which does indeed exist! https://commons.datacite.org/?query=media.media_type:chemical/x-gaussian-log+AND+media.media_type:text/plain+AND+(titles.title:*Exo*+OR+titles.title:*Endo*) This query is now also included in a renamed Data availability and discovery statement section of the article, which reveals as many as 73 datasets with the type of data the referee asks for! The information can also be found or “discovered” by following the hierarchy of collections from the master collection at DOI: 10.14469/hpc/13058 down to e.g. DOI 10.14469/hpc/13057 (Mechanism of the Swern oxidation: Transition state kinetic isotope effect calculations) where text files containing the details asked for are included, as per e.g. specific eentry 10.14469/hpc/13031. Other collections can also be found by this procedure, such as DOI: 10.14469/hpc/13327 (Mechanism of the Swern oxidation: Kinetic isotope effects for novel substituents). Finally we have also included what is now starting to be described as a “Finding AID” in the form of a FAIR Table, DOI: 10.14469/hpc/13370 where links to the data (labelled in the table as either TS or R) are provided for all the calculated KIE, along with 3D animated models of the transition structuref. The data is available as the outputs of the Kinisot program and the log files used as inputs to this program. We have added a summary of the above to the article, in the spirit of “digital discovery” theme of the journal itself. We leave it to the editors whether including discovery information as part of the data availability statement in this manner is appropriate, or whether it is better suited for a separate new section or indeed in a separate article (see below).

Referee: 2 Since each benzaldehyde product is obtained from a racemic mixture of deuterated benzyl alcohol enantiomers, the intramolecular KIEs should be obtained either from the respective TS energies alone (Curtin-Hammett) or from the respective products of an equilibrium constant for formation of each epimer multiplied by the appropriate rate constant k1, k2, k3 or k4.

Response:
All observed KIE for the mono deuterated species were obtained from the ratio of the calculated primary/secondary values, as originally performed by GM, via solution of Equation 1. This equation absorbs the equilibrium constant for the formation of each epimer.

Referee: 2 There are some minor issues: e.g. symbols for physical quantities that should be italicised; use of ‘transition state’ where transition structure is better (see: IUPAC Glossary of Terms in Physical Organic Chemistry 2021); ‘imaginary mode wavenumber’ where ‘imaginary wavenumber for the transition vector’ is meant, and several spelling errors/typos. GM’s work used the Bigeleisen equation (not the Bigeleisen-Mayer equation) and it is not clear what is meant by the statement that the Kinisot program is ‘based on a much earlier implementation of these equations’.

Response: All corrected

I applaud the authors on their demonstration of FAIR data, but I cannot recommend publication of this paper in its present form.

Referee: 3
Rzepa and coworkers reported a very systematic computational/DFT (B3LYP, B3LYP+D, wB97X-D, B2PLYP-D3 method) study to re-visit one previous experimental and computational KIE study on SWERN oxidation (ref. 9). The current systematic study attempts to understand and resolve the previously claimed discrepancy between the measured and computed KIEs in ref. 9 by examining effects on computational settings, solvent, basis sets, DFT functionals and dispersion correction on the computed KIE values. Notably and surprisingly, larger basis sets were found to be essential to match/resolve the measured and computed KIEs based on transition state theory. They also suggest that using a substituent of NMe2 can give a larger KIE value for the future experimental verification. Extensive data generated in this study are collected/stored in FAIR data mode, rather than conventional ESI file. Overall, this systematic study give interesting and less-common method comparison about KIE calculations, which should be useful in computational chemistry community. I would recommend this work after the authors address below minor issues. However, I am not sure whether this work fits themes of Digital Discovery, in which I found most papers relates to AI/ML. I let the editor to make a decision in this point.

Response:
As noted elsewhere, and in direct response to an incorrect assertion by Referee 2, who apparently had not discovered all the data we had provided in relation to KIE calculations using the Bigeleisen equation, we have now added an expanded Data availability and discovery statement, which we feel could usefully be included in most journal articles and seems particularly appropriate to be first illustrated in the present journal! The point could be made that data availability on its own is not optimally useful if it cannot be easily discovered - or indeed re-used or easily interoperated, not only by a human but especially by an AI-driven machine. I make the point here that conventional data availability statements can often relate to a single PDF document which can exceed 1000 pages in length, which is only really processable by a human and as a result the data within such a document is often not at all easily discovered, and almost certainly is not AI-Ready. So we ask the editor their opinion on this particular aspect.

1. Referee: 3 If possible, add a specific base name in Scheme 1.

Response: Done

2. Referee: 3 As to “Replication of Energies”, it should be helpful for the readers to follow if they put their obtained values using different settings in a Table(s).

Response: We have styled the energies quoted from the ESI of ref 9 in italics and our replication values without italics to help follow. We are happy to accept any other form of styling that the editors may suggest.

3. Referee: 3 Fig. 3 (x/y-axis), it may be better to use element labels better than atom index? e.g., C1-O-S-C2? Moreover, “Species 2 (endo) has a C-O torsion angle of ~10° and species 3 (exo) ~165°” I suggest the authors to use labels or highlight the two values for 2 & 3 in Fig. 3.

Response:Done. The torsion of +10 has been modified to +190 to better correspond to the figure axis and the labels added.

4. Referee: 3 wB97XD --> wB97X-D

Response: Done

5. Referee: 3 What is the main problem for underestimating KIE values when a small basis set was used? A problem of weak non-covalent interactions? Could the authors add some discussion?

Response: We have added a comment in the conclusions section that the proton transfers studied here are non-linear, which may be associated with the need to use higher quality basis sets. Clearly more examples are needed. We will try to identify these in a follow up study and thank the referee for this suggestion.

6. Referee: 3 Accurate prediction or reproduction of a (intramolecular) primary (H/D) KIE value by computational methods is quite changeling. Tunnelling could have some influence. Instead, I would propose to perform parallel secondary (H/D) KIE using Ph-CH2-OS-R and Ph-CD2-OS-R substrate. Generally, secondary (H/D) KIE value is smaller and can be reliably evaluated by computational methods.

Response:
The original article does not report any results using the Ph-CD2-OS-R substrate, so there are no values to compare with computed ones. We are not in a position to synthesize such species as part of the present project. We do note however that the inclusion of files containing eg the full force constant matrix (either the .log file and the .fchk files) makes it trivial for anyone to compute new isotopic substitutions should they wish, unlike the original article ref 9, where the required force constant matrix was not reported in the ESI.

04-Jun-2024

Dear Dr Rzepa:

Manuscript ID: DD-ART-12-2023-000246.R1
TITLE: Modelling kinetic isotope effects for SWERN Oxidation. DFT-based Transition State Theory is OK.

Thank you for submitting your revised manuscript to Digital Discovery. I am pleased to accept your manuscript for publication in its current form. I have copied any final comments from the reviewer(s) below.

You will shortly receive a separate email from us requesting you to submit a licence to publish for your article, so that we can proceed with the preparation and publication of your manuscript.

You can highlight your article and the work of your group on the back cover of Digital Discovery. If you are interested in this opportunity please contact the editorial office for more information.

Promote your research, accelerate its impact – find out more about our article promotion services here: https://rsc.li/promoteyourresearch.

If you would like us to promote your article on our LinkedIn account [https://rsc.li/Digital_showcase] please fill out this form: https://form.jotform.com/213544038469056.

We are offering all corresponding authors on publications in gold open access RSC journals who are not already members of the Royal Society of Chemistry one year’s Affiliate membership. If you would like to find out more please email membership@rsc.org, including the promo code OA100 in your message. Learn all about our member benefits at https://www.rsc.org/membership-and-community/join/#benefit

By publishing your article in Digital Discovery, you are supporting the Royal Society of Chemistry to help the chemical science community make the world a better place.

With best wishes,

César de la Fuente, Ph.D.
Associate Editor, Digital Discovery

Reviewer 1

According to the author’s response, my concerns were already considered by them and left for future work. Therefore, I recommend the publication of the paper in its present form and await with expectation their follow-up work, in which my remarks will be properly addressed.

Reviewer 2

The authors have addressed the points raised in the original review adequately.
I endorse their request for the Editor to a decision regarding the suitability of their data availability and discovery statement.