The Impact of Arginine Side Chains on the Mechanism of Polycondensation of Silicic Acid in Bioinspired Mineralization

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this work, Lanuza and coworkers presented a theoretical investigation on the impact of arginine side chains on the condensation of silicic acids. The authors showed that the presence of guanidinium as a mimic of the arginine side chain doesn’t impact the dimerization of silicic acid molecules too much. By carefully reading, I found the following should be addressed before the manuscript can be considered further.

1. On line 120, in “deltaEelec”, “elec” should be subscript to “deltaE”. The unit of energy should be mentioned in the caption of Table 1.
2. The energy barrier and free energy barriers seem to decrease with the presence of guanidinium. The authors should spare their efforts on this. Would this be a kind of promotion exerted by the guanidinium?
3. Furthermore, the calculated barriers for MAM are smaller than those for AAM. The authors should double-check this, as the trend is different from those reported before.
4. It may not be fair that the authors only include the Guanidinium cation in the calculations. It would be more interesting to see that Guanidinium is accompanied by an anion. Further, the state of Guanidinium may depend on pH. It would be even more interesting if the authors could combine the condensation with the pH of the reaction system.
5. I noticed another paper (10.1021/acsomega.1c03235.) mentioned an alternative mechanism where the amine would abstract the proton of silicic acid and convert the acid into anion for the condensation to take place. One may wonder if Guanidinium may impact such a reaction in other ways.

 

Author Response

Comment 1: On line 120, in “deltaEelec”, “elec” should be subscript to “deltaE”. The unit of energy should be mentioned in the caption of Table 1.

Response 1: Thanks for pointing this out. The units have been added to the table and in the caption, the subscript has been changed.

Comment 2: The energy barrier and free energy barriers seem to decrease with the presence of guanidinium. The authors should spare their efforts on this. Would this be a kind of promotion exerted by the guanidinium?

Response 2: Indeed, the energy barriers decrease in the presence of guanidinium. We have now better clarified that “Although the present results indicate that guanidinium can facilitate the process, the reaction mechanism does not appear to change dramatically”

Comment 3: Furthermore, the calculated barriers for MAM are smaller than those for AAM. The authors should double-check this, as the trend is different from those reported before.

Response 3: Comparing our calculations with those appearing in 10.1021/acsomega.1c03235 and https://www.mdpi.com/1422-0067/20/12/3037, our models include an additional water molecule that helps stabilizing the transition states. We performed single point calculations on all the structures obtained, using PBE with dispersion correction and BLYP without dispersion correction, which is closer to the methodology used in these papers (the basis set is not the same, but it is reasonably similar)  and we obtained the same trend between MAM and AAM. With PBE-D3 we obtained energy barriers that are systematically lower by 4 to 5 kcal mol-1, whereas with BLYP the energy barriers are within 1 kcal mol-1. Therefore this effect is not related to the calculations setup but to the fact that we have an additional water molecule in the model.  

Comment 4: It may not be fair that the authors only include the Guanidinium cation in the calculations. It would be more interesting to see that Guanidinium is accompanied by an anion. Further, the state of Guanidinium may depend on pH. It would be even more interesting if the authors could combine the condensation with the pH of the reaction system.

Response 4: The effect of guanidinium appears to be reduced already when water molecules are intervening between the cation and the silicic acid. Thus one can intuitively expect that the effect will be even lower when another anion is added. However, given that the arginine sidechains are water exposed, it is not necessary that an anion is in close proximity. As far as the pH effect is concerned, the pKa of the guanidinium moiety in arginine is 12.48, which means that the molar fraction of deprotonated guanidine under the experimental conditions is of the order of 10-5. This is now stated in the text.

 

Comment 5: I noticed another paper (10.1021/acsomega.1c03235.) mentioned an alternative mechanism where the amine would abstract the proton of silicic acid and convert the acid into anion for the condensation to take place. One may wonder if Guanidinium may impact such a reaction in other ways.

Response 5: We apologize for not mentioning this paper, which actually was one of the starting points of our investigation. This is now cited in response to the previous point.

Reviewer 2 Report

Comments and Suggestions for Authors

Authors have studied using DFT B3LYP in vasp the two pathways for the polycondensation of silicic acid.  The manuscript would benefit if the authors can incorporate a few suggestions

1. The plots with Si and O of similar colors it hard to understand the ball and stick representations. Please label them with the color and the atom name. Further, provide the energy level diagram with all the thermodynamics in a plot for both cases. This will help understand the reaction, the two pathways, and the concerned TS.
2.  More importantly, while authors can use DFT B3LYP to make this inference, for a theoretical study, it is important to justify the usage of functional along with the comparision with if possible any wavefunction based method.  Even the usage of B3LYP over PBE or conventional functionals need to accessed or justified.
3. The differences in the observed bond lengths are nominal as shown in Table 1 and Table2. . Thus it becomes important to rationalise the observed differences in Free energy for the two mechanisms based on the parameter that might be more related to the reaction coordinate .
4.  The role of real solvent molecules versus the polarisable continuum model is also worth exploring.  how many water molecules do participate in the stabilization of reactant, product and TS and how they change the free energy for the two recation is worth exploring.

Author Response

Comment 1: The plots with Si and O of similar colors it hard to understand the ball and stick representations. Please label them with the color and the atom name. Further, provide the energy level diagram with all the thermodynamics in a plot for both cases. This will help understand the reaction, the two pathways, and the concerned TS.

Response 1: Labels have been added. We did not include the energy level diagram because we felt that it would be redundant with respect to the corresponding tables.

Comment 2: More importantly, while authors can use DFT B3LYP to make this inference, for a theoretical study, it is important to justify the usage of functional along with the comparison with if possible any wavefunction based method.  Even the usage of B3LYP over PBE or conventional functionals need to accessed or justified.

Response 2: We agree with the reviewer that the combination of methodology and functional might have appeared arbitrary in our previous submission. We have now stated more clearly the motivation for our computational approach, which we chose to make reference to previous results on this same reaction: The choice of the computational methodology was based on previous computational results on the same reaction we are investigating [34,35], and using the hybrid B3LYP functional over the GGA functionals BLYP or PBE because it has a better performance on structure optimization and single point energy calculations for main group elements.

Comment 3: The differences in the observed bond lengths are nominal as shown in Table 1 and Table2. . Thus it becomes important to rationalise the observed differences in Free energy for the two mechanisms based on the parameter that might be more related to the reaction coordinate.

Response 3: We thank the reviewer for pointing this out, as we realized there was an error in the tables, and we removed the first bond length column. It is now more apparent which are the geometric trends associated with the reaction.

Comment 4: The role of real solvent molecules versus the polarisable continuum model is also worth exploring.  how many water molecules do participate in the stabilization of reactant, product and TS and how they change the free energy for the two reaction is worth exploring.

Response 4: See also response to point 3 of reviewer 1. We have performed calculations including more water molecules. Increasing the number of water molecules does not have a major impact on the results of the calculations, which makes sense because we had already considered the most important interactions in the calculations reported in the first version. We have included the results of these calculations in the zotero repository.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I read the revised manuscript carefully and was happy to find the authors did a good job on revising the manuscript and addressing the comments of the reviewers. They have fully resolved my concerns, so I think the manuscript will be publishable now.