Theoretical Study on the Mechanism of CO* Electrochemical Reduction on Cu(111) under Constant Potential
Round 1
Reviewer 1 Report (Previous Reviewer 3)
manuscript can be published in the present form
Author Response
Thanks for the reviewer's approval.
Reviewer 2 Report (Previous Reviewer 2)
Although the authors have revised and resubmitted the manuscript, the quality of the manuscript does not improve obviously. Here are my comments.
1. Please add the H-C distance of TS in Figure 1 and H-O distance of TS in Figure 3.
2. The authors calculated the transition states at different potential conditions. The authors still need to do vibrational frequency calculations to verify the transition states and the imaginary frequencies should be provided in SI.
3. In terms of CO2 reduction, the main problem in this field is the product selectivity, such as CH4, CO, HCOOH etc. However, the reduction pathways of CO have limited connection to these possible products, hence the novelty of this manuscript is strongly limited. It is insufficient to publish a paper by calculating only two reaction steps, and few insights are seen from current manuscript.
I recommend the authors do an English editing. Also check the grammar of the sentences such as "this two electrocatalytic reactions", "calculation on this two electrocatalytic reactions with including the solvent effect"
Author Response
Response to reviewer 2 comments:
Although the authors have revised and resubmitted the manuscript, the quality of the manuscript does not improve obviously. Here are my comments.
point1: Please add the H-C distance of TS in Figure 1 and H-O distance of TS in Figure 3.
response1: revised as suggested.
point2: The authors calculated the transition states at different potential conditions. The authors still need to do vibrational frequency calculations to verify the transition states and the imaginary frequencies should be provided in SI.
response2: Revised as suggested. frequency calculations are provided in supporting information.
point3: In terms of CO2 reduction, the main problem in this field is the product selectivity, such as CH4, CO, HCOOH etc. However, the reduction pathways of CO have limited connection to these possible products, hence the novelty of this manuscript is strongly limited. It is insufficient to publish a paper by calculating only two reaction steps, and few insights are seen from current manuscript.
response3: Thanks for the comments. This is the drawback of our research, we will consider more reaction steps in our futher studies.
I recommend the authors do an English editing. Also check the grammar of the sentences such as "this two electrocatalytic reactions", "calculation on this two electrocatalytic reactions with including the solvent effect"
response: Revised as suggested.
Reviewer 3 Report (New Reviewer)
The article entitled "Theoretical study on the mechanism of CO* electrochemical reduction on Cu(111) under constant potential" by Mei et al. reported an interesting theoretical study on the mechanistic chemistry of CO reduction on Cu(111). Especially, the constant potential method was used in this study, which is novel and interesting. I, therefore, recommend it for publication after minor revision.
1) The colour codes of the atoms are missing in Figs. 1 and 3.
2) It’s better to mention the reference potential here. For example, unit V may be changed to V vs. SHE.
3) The cutoff energy of 400 eV may be not high enough. The authors need to justify this choice.
4) Why is the unit of the reaction coordinate angstrom in Figs. 2 and 4? It should be unitless. At the same time, the molecular state of the reactant/intermediate/product along the reaction coordinate should be clearly provided in the figures.
5) It’s better to provide a brief description of how the number of electrons (n) was calculated using the constant potential method because this is one of the most important parameters in this paper.
There are some grammatical errors. It's better to further improve English writing.
Author Response
Response to reviewer 3 comments:
The article entitled "Theoretical study on the mechanism of CO* electrochemical reduction on Cu(111) under constant potential" by Mei et al. reported an interesting theoretical study on the mechanistic chemistry of CO reduction on Cu(111). Especially, the constant potential method was used in this study, which is novel and interesting. I, therefore, recommend it for publication after minor revision.
1) The colour codes of the atoms are missing in Figs. 1 and 3.
response1: Revised as suggested.
2) It’s better to mention the reference potential here. For example, unit V may be changed to V vs. SHE.
response2: Revised as suggested.
3) The cutoff energy of 400 eV may be not high enough. The authors need to justify this choice.
response3: Revised as suggested. The results of convergence test are given in SI.
4) Why is the unit of the reaction coordinate angstrom in Figs. 2 and 4? It should be unitless. At the same time, the molecular state of the reactant/intermediate/product along the reaction coordinate should be clearly provided in the figures.
response4: (1).Each value of the reaction coordinate corresponds to the relative position of each atom in the reaction system, thus it is unit wise. (2)Because there are so many lines in the figure and the positions of the transition state on the reaction coordinates under different potentials are not necessarily similar, more text annotations may affect the cleanness of the picture.
5) It’s better to provide a brief description of how the number of electrons (n) was calculated using the constant potential method because this is one of the most important parameters in this paper.
response5: Revised as suggested. We added a brief description in the computational methods section.
Reviewer 4 Report (New Reviewer)
In their manuscript entitled “Theoretical study on the mechanism of CO electrochemical reduction on Cu(111) under constant potential”, Shange Mei and Wanzhen Liang present a fully computational study on reduction of CO on a Cu surface. Periodic DFT simulations are performed to investigate potential intermediates, products, and the underlying reaction mechanism, i.e., the COH* as well as the CHO* pathways. Of particular interest is the effect of applying an external (electric) potential.
The manuscript is written in an elegant fashion, the results are presented adequately and all conclusions drawn by the authors are justified by the data. Publication of the manuscript is recommend in its present form.
Language wise, the manuscript is at a high standard with mere minor stylistic issues.
Author Response
Thanks for the reviewer's approval.
Reviewer 5 Report (New Reviewer)
In the manuscript, the authors employed fixed potential DFT computations to provide mechanistic insights into the conversion of CO on the surface of Cu(111) to CHO and COH intermediates. They studied the impact of the applied potential on the above reactions. The authors' findings merit publication, but some comments must be addressed.
1) To complete the study, the authors should discuss CO poisoning at
different potentials. Examining the binding strength and modes of CO at
different potentials would be beneficial since the ultimate objective is to
convert CO into value-added chemicals. Furthermore, the binding strength and
applied potential may influence zero-point energy (ZPE) and/or entropic
corrections, which are critical for determining free energy that ultimately governs
kinetics.
2) Since applied potential can prompt structural changes, the authors
should graph potential vs surface energy, as well as a shift in the d-band
center.
3) In Figures 2 and 4, the meaning of reaction coordinates in the present
reaction needs to be explicitly defined in the text. It would also be
preferable to use the symbol for angstrom.
Overall, the authors' study sheds light on the complex kinetics of CO
reduction on Cu, which is crucial for improving industrial processes and
enhancing the production of valuable chemicals.
Author Response
Response to reviewer 5 comments:
In the manuscript, the authors employed fixed potential DFT computations to provide mechanistic insights into the conversion of CO on the surface of Cu(111) to CHO and COH intermediates. They studied the impact of the applied potential on the above reactions. The authors' findings merit publication, but some comments must be addressed.
1) To complete the study, the authors should discuss CO poisoning at
different potentials. Examining the binding strength and modes of CO at
different potentials would be beneficial since the ultimate objective is to
convert CO into value-added chemicals. Furthermore, the binding strength and
applied potential may influence zero-point energy (ZPE) and/or entropic
corrections, which are critical for determining free energy that ultimately governs kinetics.
Response 1: Thanks a lot for the suggestion. Considering the difficulty of performing calculation on fragments in a constant potential giant canonical system, we used integrated COHP to represent the bonding strength between C atom and Copper atom.
2) Since applied potential can prompt structural changes, the authors
should graph potential vs surface energy, as well as a shift in the d-band
center.
response 2: Calculation on bulk Cu is needed to get the surface energy, however, it is unsupported in the eNEB code, because it is based on the electrolyte model.
a d-band center shift of Cu slab vs potential figure are added.
3) In Figures 2 and 4, the meaning of reaction coordinates in the present
reaction needs to be explicitly defined in the text. It would also be
preferable to use the symbol for angstrom.
response 3: revised as suggested.
Overall, the authors' study sheds light on the complex kinetics of CO
reduction on Cu, which is crucial for improving industrial processes and
enhancing the production of valuable chemicals.
Round 2
Reviewer 2 Report (Previous Reviewer 2)
Thanks the authors to address my concerns and their efforts to improve the quality of the manuscript. However, the manuscript still needs major revision due to:
1. The authors stated the importance of CO poisioning at different potentials, however which potential do they use to plot Figure 5?
2. I assume the authors calculated the d-band center averaged on all Cu atoms. It would be beneficial to plot the trend of the surface Cu atom, which is more relevant to CO adsorption/poisioning.
3. For Figure 6, the authors should include the results of another spin state.
4. grammar mistake "The reason why this minimum point appears may be because the shift of the d-band center of copper". an extra dot at line 138
5. If the authors wanted to study CO poisoning, the calculations at various CO coverages should be done to support the poisoning effect.
6. From Table S2, several imaginary frequencies are approaching 0, e.g. 15.11, 17.16 20.28, 27.68 etc. Have the authors checked the vibration modes of these frequencies?
7. From Figure S1 (2), the vibrations cannot reflect the favorability of reaction pathways, it can only indicate the transition state is not a stable transition state and the authors need to find one which clearly shows the H stretching in between H2O and CO.
Author Response
Response to reviewer 2 comments:
Thanks the authors to address my concerns and their efforts to improve the quality of the manuscript. However, the manuscript still needs major revision due to:
1. The authors stated the importance of CO poisioning at different potentials, however which potential do they use to plot Figure 5?
response 1 : Figure 5 uses the data of 0.0V. The DOS and COHP figure are plotted to show the electronic structure of *CO adsorbed on a Cu(111) slab, since the variation of potential do not change the electronic structure obviously, we did not plot the results under other potentials.
2. I assume the authors calculated the d-band center averaged on all Cu atoms. It would be beneficial to plot the trend of the surface Cu atom, which is more relevant to CO adsorption/poisioning.
response 2 : Thanks very much for the suggestion, we have revised it as suggested.
3. For Figure 6, the authors should include the results of another spin state.
response 3 : Thanks very much for the suggestion, but a 3*3*4 atoms Cu slab with CO molecules is not a spin polarized system and our results show that the data is exactly the same for both spins.
4. grammar mistake "The reason why this minimum point appears may be because the shift of the d-band center of copper". an extra dot at line 138
response 4 : Thanks very much for the suggestion, we have revised the grammar mistake.
5. If the authors wanted to study CO poisoning, the calculations at various CO coverages should be done to support the poisoning effect.
response 5 : Thanks very much for the suggestion, we have add the results of different CO coverages, please see the manuscript for details.
6. From Table S2, several imaginary frequencies are approaching 0, e.g. 15.11, 17.16 20.28, 27.68 etc. Have the authors checked the vibration modes of these frequencies?
response 6 : We apologize for our carelessness. Generally, the vibration mode of the imaginary frequency of the TS will not be changed by the variation of potential, only the frequency will be affected. We have checked those small frequencies again and find that the CHO results of -0.8V, -1.2V, -1.6V and COH result of -0.9V are not true. We re-ran the calculation at a higher precision level for those structures with small imaginary frequencies and ensure the vibration mode of the imaginary frequency of the TSs are the same for same reaction.
7. From Figure S1 (2), the vibrations cannot reflect the favorability of reaction pathways, it can only indicate the transition state is not a stable transition state and the authors need to find one which clearly shows the H stretching in between H2O and CO.
response 7 : Thanks very much for the comment. The COH path is quite simple since it only contains a proton transfer process, thus the eNEB calculation is not likely to give wrong results. However, after several recalculations at different settings, we still did not observe the "H stretching in between H2O and CO" mode of the imaginary frequency. We suspect that there is no real transition state for this reaction and the barriers of COH path we obtained are due to the transformation of the orientation of water molecules. Here we provide a previous research paper to support this opinion, J. Am. Chem. Soc. 2021, 143, 6152-6164. [doi: 10.1021/jacs.1c00880]. In this paper, the authors calculated COH path and find that no transition state observed in the COH path (see Figure 3 of this article), which means the bond break between the water molecule and the proton (or between the CO molecule and the proton, from the other side of reaction) needs no activation energy in this reaction. This phenomenon may be due to the presence of hydrogen bonds. The main difference between their results and ours is that the charged slab will affect the orientation of the water molecule and that's possibly where the barrier comes from.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
This work focuses on a puzzling step in the CO2 reduction reaction – the formation of CHO* and COH* intermediates – at a constant potential. Accordingly, the author utilized a grand canonical density functional theory method. The author concluded that the formation of CHO* and COH* via studied paths is potential dependent.
The utilized method is relatively new to the electrocatalysis field and, thus, deserves more attention. Let me recommend references from a recent review paper by Abidi et al. [10.1002/wcms.1499]. In that review, the expression for Eq. 1 looks different. I agree with Abidi et al. [10.1002/wcms.1499] that (in Eq. 1) free energy should be written instead of enthalpy. The author must explain why there should be enthalpy (like in the original publication by Duan and Xiao [10.1021/acs.jpcc.1c02998]). The choice of the reference value of 4.6 eV in Eq. 2 must be explained in more detail. I suspect that some units in Eqs. 2 and 3 require conversion via division by e.
It needs to be clear which NEB variation was used in the study. Curreently, NEB details are not given in the manuscript. Thus, I doubt whether the true maxima in Figures 2 and 3 were reached. In addition, a low number of states on the MEP in Figure 2a makes me question the results. I suggest improving consistency between Figures 2 and 3 by presenting subplots of the same type for the same set of potentials. For example, Energy (eV) vs reaction coordinate and Charge (e) vs point of the MEP. Such a change will ease the comparison of the figures. The same applies to Tables 1 and 2, which should be presented similarly.
Bader charges are still used for the charge analysis because there were no working codes to run other charge analyses. Nowadays, there are such codes, and I suggest presenting (perhaps in the Supporting Information) DDEC and GRID charges: https://sourceforge.net/projects/ddec/files/ and https://github.com/theochem/grid. The
Please correct
Line 130: "form H3O species" to "form H3O+ species" as from the chemical point of view, H3O is an extremely unstable molecule.
Line 162 "table .such" to "table 2"
The author will benefit from uploading the computed data to one of the open repositories, like https://www.catalysis-hub.org/. For example, I could suggest an NEB description if only the data were openly available.
Overall, this study and its finding are of some interest for the electrocatalysis community and can be published after a revision. The revision should include improving readability by rewriting the text with a focus on clarity.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In this manuscript, the author conducted some studies on the CO reduction to COH or CHO on Cu(111) surface. Although the author used a constant potential model, I don't think this can be regarded as novelty since there are many works working on the same field. In addition, some of the results in the manuscript are doubtful, and hence I would like to reject the manuscript.
1. The the novelty of this manuscript is limited. The author only compared the CO reduction to COH or CHO. However this step has been thoroughly studied before and the corresponding results are even covered in the CO2 reduction reaction process. See J. Am. Chem. Soc. 2017, 139, 1, 130–136; https://doi.org/10.1073/pnas.2202931119.
2. The way that the author conduct the H-shuttling is different with Nie et al.'s work. It is counterintuitive that a H3O+ radical exists in near surface area. In addition, the transition state structure is doubtful, the H is still too close to the H2O, which is similar as the initial state. In addition, will a water molecule adsorb onto a surface by using the methods? If so, why would not a H3O+ decompose to surface H and H2O?
3. I don't agree "the reported theoretical works mostly used models that contains constant charge". There are works featuring constant potential calculations. e.g. ACS Omega 2019, 4, 17, 17269–17278
4. What does the x-axis mean in Figure 2a and 4a? It seems the reaction coordinates have no real meaning.
5. CO2 reduction is a more complicated process, but the author only calculated two steps. I don't think the current results can support the "relatively negative production potential of methane in Cu catalyzed CO2 electroreduction experiments". In addition, Cu(111) may not be the most active surface for CO2RR on Cu. So the conclusion is unconvincing.
6. The k-points are not large for a 3x3 supercell. Can you verify the settings do not affect the results? Especially for a Bader charge analysis.
Comments on the English writing of the manuscript:
1. Some sentences are not suitable for academic writing, such as the long subject "Recycling carbon dioxide (CO2) in the environment and converting it into useful fuels and feedstocks is a promising green chemistry strategy". The sentence is difficult to understand: "Such a system that varies its number of electrons to keep its potential constant during structural changes can be treated as a grand canonical system", "The activation and reaction energies decrease from 0.85 eV and 0.79 eV for U = 0.0V to 0.62 eV and 0.62 eV for U =−1.6V as the potential becomes more negative"
2. Grammatical errors, e.g. "which may due to", "We performed a more detailed mechanistic study ..." however no comparisons are made.
3. Lack the description of NEB in Computational Methods.
4. Some sentences lack the logic linking: "This is because the system charge of the IS, TS and FS are changed to keep the potential constant (Figure 2(b))." I don't know what the meaning of "this" is, considering this sentence is the first sentence of a paragraph.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The manuscript describes the CO2 reduction on Cu surface by unraveling the selectivity between to competing intermediates COH and CHO. The manuscript is presented well and can be accepted for publication after minor revisions.
1. The main drawback of the manuscript is not talking about hydrogen evolution reaction. At studied potentials, HER is also very competing on Cu surfaces. Authors should account fir HER possibilities.
2. Manuscript should be checked thoroughly for English corrections.
Author Response
Response to reviewer 3 comments
Point 1: The main drawback of the manuscript is not talking about hydrogen evolution reaction. At studied potentials, HER is also very competing on Cu surfaces. Authors should account fir HER possibilities.
Response 1: We agree that HER is a strong competitive side reaction, and it should be considered for comprehensive analysis. However, it seems we do not have enough time to calculate them during the revision, we will consider them in future works.
Point 2: Manuscript should be checked thoroughly for English corrections.
Response 2: English writing will be checked, thanks for advice.
Round 2
Reviewer 1 Report
In this review, I will go through the previous point of Reviewer #1.
Point 1.
(1) Arbitrary switching of E or A to H or G is unacceptable according to their thermodynamic definitions. If the potential energy is used, then equations should include E rather than H; or it must clearly explain how E transformed to H.
(2) 4.4 and 4.8 are not two experimental values, so one must not average them. Besides, Neurock is referring to very old works. For a more modern overview, see 10.1021/ja073946i. Please find more recent studies to ground your choice.
Point 2.
Please write "Climbing image-nudged elastic band (CI-NEB)". Please provide a link to eNEB code (if available) or write "eNEB method" (if there is only a method and no code).
I'm afraid I have to disagree with the statement that 1 image is enough to locate the transition state with CI-NEB. If the eNEB somehow allows identifying the true energy maxima (barrier) in one step, please explain how eNEB does it in the text.
---
Please rewrite the text with a focus on clarity. Use Grammarly or Writefull to fix the grammar. Try Quill or ask ChatGPT to refine the text.
Author Response
Response to reviewer 1 comments:
Point 1.
(1) Arbitrary switching of E or A to H or G is unacceptable according to their thermodynamic definitions. If the potential energy is used, then equations should include E rather than H; or it must clearly explain how E transformed to H.
(2) 4.4 and 4.8 are not two experimental values, so one must not average them. Besides, Neurock is referring to very old works. For a more modern overview, see 10.1021/ja073946i. Please find more recent studies to ground your choice.
Response 1:
(1) Since the paper presented the eNEB method did not explain why it is 'enthalpy', here we try to explain and make it sounds reasonable.
The H used here is a concept under electrochemical conditions, rather than thermodynamical, it can be regarded as corrected energy. As we can see, the enthalpy expression under electrochemical conditions,
H(r, n) = E(r, n) - (n - n0)φ
is similar to the enthalpy expression, H = E + pV, they both contain an internal energy term and a generalized force multiple generalized displacement term. The difference is that the former is defined in constant electrochemical potential and the latter is defined in constant pressure.
In case more people get confused, we changed 'enthalpy' to 'corrected energy'.
(2) Thanks very much for providing more recent work on estimating the absolute SHE, this paper and some works mentioned in the paper have been referenced in our manuscript.
The absolute potential of SHE has 3 definitions and at least 5 research groups obtained 5 results of it, since the estimating of the absolute SHE is quite difficult to work and no certain value was determined up to now, it appears that 4.6 V is an acceptable value according to the paper(4.2 ± 0.4) and the previous works mentioned in the paper.
Point 2.
Please write "Climbing image-nudged elastic band (CI-NEB)". Please provide a link to eNEB code (if available) or write "eNEB method" (if there is only a method and no code).
I'm afraid I have to disagree with the statement that 1 image is enough to locate the transition state with CI-NEB. If the eNEB somehow allows identifying the true energy maxima (barrier) in one step, please explain how eNEB does it in the text.
Response 2:
The eNEB code can be found at https://github.com/penghao-xiao/Electrochemical-barrier/releases .
According to Henkelman's paper that presented the CI-NEB method in 2000, ( J. Chem. Phys. 113, 9901 (2000); doi: 10.1063/1.1329672) the image with the highest energy will climb to the saddle point as long as the CI-NEB calculation converged, thus a CI-NEB calculation with 1 image will reach the same transition state as that with 10 images if the calculation finished normally. The main reason why we use more images in some CI-NEB calculations is that we usually do not know if there exists more than 1 saddle point on the MEP path. However, we already know from Nie's work that there's only one maxima point on the MEP of the CHO path, so we can use this saddle point structure as the climbing image to approach the results.
---
Please rewrite the text with a focus on clarity. Use Grammarly or Writefull to fix the grammar. Try Quill or ask ChatGPT to refine the text.
Response :
The grammar of the text is fixed using Grammarly, we tried to make the text clearer.
Reviewer 2 Report
Thank you for addressing my concerns.
Author Response
Thanks for comments and suggestions
