Review Reports - DFT Insights into the Adsorption of Organophosphate Pollutants on Mercaptobenzothiazole Disulfide-Modified Graphene Surfaces

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This manuscript presents a comprehensive and well-executed computational DFT study on the adsorption of various organophosphate (OP) pollutants on MBTS-functionalized graphene. The topic is highly relevant to environmental monitoring and remediation. The manuscript is generally well-written and suitable for publication after minor revisions to clarify certain points and address the comments below.

The central conclusion is that G-MBTS offers advantages for sensing due to enhanced charge transfer and reduced HOMO-LUMO gaps, despite similar or slightly weaker adsorption energies compared to MBTS alone. This point is crucial but could be emphasized more clearly in the abstract and conclusions. Please explicitly state that the primary benefit of the graphene substrate is not necessarily strengthening adsorption but tuning the electronic properties for improved electrochemical response.
The introduction briefly mentions the selection criteria (agricultural relevance, frequent detection, functional group diversity). It would be beneficial to add a sentence or two in the introduction or the Computational Details/Results section explicitly linking this diversity to the expected range of interaction strengths (e.g., from weak dispersion for aliphatic glyphosate to strong π-π/chalcogen bonding for aromatic OPs), which is a key aspect of the study's findings.
The uniformly narrowed HOMO-LUMO gap (~0.79 eV) for all G-MBTS-OP complexes is a key result. The authors should briefly discuss the implications of such a small and consistent gap. Does this indicate the material may behave as a narrow-gap semiconductor upon adsorption, and how does this directly benefit sensing (e.g., increased conductivity, facilitating redox processes)?
Abstract: The phrase "MBTS alone displayed selective affinity" is supported by the range of energies, but the term "selective" could be slightly qualified (e.g., "a range of affinities, indicating potential selectivity").
Page 2, Introduction: The sentence "Density functional theory (DFT) calculations from that study highlighted the importance..." has an incomplete citation [12? -15]. Please correct this to the appropriate references.
Page 3, Figure 2: The caption "Chemical structures of the organophosphates under study" is present, but the figure itself is not included in the provided PDF. Please ensure all figures are included in the final submission.
The discussion of the η index in Table 3 is excellent. For clarity, consider adding a brief definition of the η index in the text when it is first mentioned (e.g., "the η index, a measure of the bond covalency").

Author Response

We sincerely thank Reviewer 1 for the careful reading of our manuscript and for the constructive comments and suggestions, which have helped us improve both the clarity and scientific rigor of the paper. All comments were addressed in detail, and the corresponding modifications have been incorporated into the revised version. The specific responses to each point are provided below. Reviewer comments are shown in black, and our responses appear in blue.

The central conclusion is that G-MBTS offers advantages for sensing due to enhanced charge transfer and reduced HOMO-LUMO gaps, despite similar or slightly weaker adsorption energies compared to MBTS alone. This point is crucial but could be emphasized more clearly in the abstract and conclusions. Please explicitly state that the primary benefit of the graphene substrate is not necessarily strengthening adsorption but tuning the electronic properties for improved electrochemical response.

We thank the reviewer for this valuable observation. The abstract, results and conclusions have been revised to explicitly highlight that MBTS anchoring primarily tunes the interfacial electronic properties rather than maximizing adsorption strength. In particular, we now state that all complexes exhibit a uniform and narrow HOMO-LUMO gap and enhanced charge redistribution, which are expected to improve electrochemical sensitivity.

The introduction briefly mentions the selection criteria (agricultural relevance, frequent detection, functional group diversity). It would be beneficial to add a sentence or two in the introduction or the Computational Details/Results section explicitly linking this diversity to the expected range of interaction strengths (e.g., from weak dispersion for aliphatic glyphosate to strong π-π/chalcogen bonding for aromatic OPs), which is a key aspect of the study's findings.

We expanded the end of the introduction to describe how the selected OP set spans from polar/non-aromatic to aromatic thiophosphates, covering dispersion, hydrogen bonding and chalcogen bonding regimes. This contextualizes the interaction trends discussed later.

The uniformly narrowed HOMO-LUMO gap (~0.79 eV) for all G-MBTS-OP complexes is a key result. The authors should briefly discuss the implications of such a small and consistent gap. Does this indicate the material may behave as a narrow-gap semiconductor upon adsorption, and how does this directly benefit sensing (e.g., increased conductivity, facilitating redox processes)?

Thank you for this suggestion. A dedicated paragraph was added to the results section, explaining that the uniform gap reflects a narrow band, charge responsive surface. The text now connects gap narrowing to enhanced sensitivity.

Abstract: The phrase "MBTS alone displayed selective affinity" is supported by the range of energies, but the term "selective" could be slightly qualified (e.g., "a range of affinities, indicating potential selectivity").

We appreciate the suggestion. The abstract has been revised accordingly. We replaced “selective affinity” with “a range of affinities, indicating potential selectivity” and preserved the numerical range to ground the statement.

Page 2, Introduction: The sentence "Density functional theory (DFT) calculations from that study highlighted the importance..." has an incomplete citation [12? -15]. Please correct this to the appropriate references.

This has now been corrected.

Page 3, Figure 2: The caption "Chemical structures of the organophosphates under study" is present, but the figure itself is not included in the provided PDF. Please ensure all figures are included in the final submission.

Figure 2 is now properly embedded and visible in the compiled manuscript. We believe this issue might have risen from a system upload error.

The discussion of the η index in Table 3 is excellent. For clarity, consider adding a brief definition of the η index in the text when it is first mentioned (e.g., "the η index, a measure of the bond covalency").

The eta index is now introduced at the beginning of the QTAIM discussion as a “covalency descriptor”.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a computational investigation of organophosphate pesticide adsorption on mercaptobenzothiazole disulfide (MBTS) and MBTS-functionalized graphene (G–MBTS) using density functional theory (DFT) with dispersion and solvation corrections. The study is timely and relevant to environmental monitoring and sensing of persistent organophosphates (OPs). The topic fits well within the scope of materials chemistry and environmental nanotechnology journals. The work is generally well organized, and the computational methodology is described clearly. However, some aspects require clarification, deeper analysis, and discussion to strengthen the scientific contribution. The manuscript presents valuable computational insights but would benefit from expanded discussion of environmental relevance, comparison with experiments, and analysis of key anomalies (e.g., glyphosate behavior). I recommend minor to moderate revision before publication.

Adsorption energies are reported in eV but lack conversion or discussion in terms of adsorption free energies or equilibrium constants under aqueous conditions. Without entropic contributions or temperature corrections, the practical significance of the binding strengths is uncertain.
The study focuses exclusively on isolated graphene nanoflakes and implicit solvent. Possible effects of surface defects, finite-size models, or explicit water molecules are not addressed.
Although related experimental studies are cited, no quantitative comparison of predicted adsorption energies or sensing behavior with experimental adsorption or electrochemical data is made.
The claim that narrower HOMO–LUMO gaps imply better electrochemical sensing could be elaborated. Conductivity of the hybrid material should ideally be supported by band-structure or charge-mobility calculations.
Glyphosate exhibits negligible adsorption on G–MBTS but large charge redistribution. The mechanistic explanation for this unusual combination is brief and warrants deeper discussion.
Some structural figures (e.g., BCPs in Fig. 5) may be difficult to interpret without clearer labeling or higher resolution.
Units for adsorption energies and charge transfer should be consistently presented and include the sign convention.
Table captions could better explain abbreviations (e.g., “FNT,” “TOK”) for readers outside the pesticide field.
Check references for formatting consistency and duplication (e.g., repeated PBE citations [17–19]).
Language is generally clear but can be tightened for conciseness in some introductory passages.
How would the inclusion of explicit water molecules or finite-temperature molecular dynamics affect the adsorption energies and charge transfer results?
Did you consider alternative graphene functionalizations (e.g., defects, oxygen groups) that may coexist with MBTS in practical materials?
How sensitive are the adsorption energies to the choice of functional and basis set? Have you benchmarked with a higher-level method?
Could you provide free-energy estimates (e.g., through vibrational entropy corrections) to evaluate adsorption at ambient conditions?
What experimental techniques would you propose to validate the predicted charge-transfer enhancements (e.g., electrochemical impedance, XPS)?
Do you expect competition between multiple pesticides in a mixture to alter the binding hierarchy observed for single molecules?
Can the predicted charge redistribution be directly correlated with measurable changes in graphene conductivity or sensor response?
Is the low adsorption energy but high charge transfer for glyphosate reproducible with different starting geometries or functionals?
Have you explored the possibility of covalent bonding or chemical degradation of MBTS on graphene under operational conditions?
Could defect engineering or co-modification (mentioned in the conclusions) be guided by specific computational descriptors from this study?

Author Response

We thank Reviewer 2 for the careful and insightful evaluation of our work. Their detailed feedback helped improve the methodological clarity, thermodynamic interpretation, and connection between the computational results and experimental sensing relevance. All suggestions were implemented in the revised version, as explained below. Reviewer comments are shown in black, and our responses appear in blue.

Adsorption energies are reported in eV but lack conversion or discussion in terms of adsorption free energies or equilibrium constants under aqueous conditions. Without entropic contributions or temperature corrections, the practical significance of the binding strengths is uncertain.

We appreciate this important comment. The Results and Discussion section now includes a short subsection titled “Thermodynamic considerations”, explicitly stating that all reported adsorption energies are 0 K electronic values including dispersion and BSSE corrections within the implicit solvent model. We also added that entropy and solvation effects will reduce the magnitude of these values and explained why moderate, reversible adsorption is advantageous for electrochemical sensing. This revision directly addresses the reviewer’s concern.

The study focuses exclusively on isolated graphene nanoflakes and implicit solvent. Possible effects of surface defects, finite-size models, or explicit water molecules are not addressed.

Statements acknowledging model limitations and future improvements have been added to the Results section. We note that explicit water and ion-pair modeling, as well as the inclusion of oxygenated defects, will further refine adsorption thermodynamics and may enhance stabilization of polar analytes such as glyphosate.

Although related experimental studies are cited, no quantitative comparison of predicted adsorption energies or sensing behavior with experimental adsorption or electrochemical data is made.

We fully agree. The Conclusions now explicitly propose electrochemical impedance spectroscopy, cyclic voltammetry, and X-ray photoelectron spectroscopy as direct validation methods. These techniques can monitor interfacial charge redistribution and are consistent with the sensing mechanism identified computationally.

The claim that narrower HOMO–LUMO gaps imply better electrochemical sensing could be elaborated. Conductivity of the hybrid material should ideally be supported by band-structure or charge-mobility calculations.

We appreciate this comment. The Results section following Table 2 has been expanded to explain that the uniform ~0.79 eV HOMO–LUMO gap represents a “preconditioned narrow-gap” regime typical of charge-responsive materials. The revised text clarifies that smaller gaps increase the interfacial density of states, facilitating electron exchange and enhancing impedance and voltammetric sensitivity. While full band-structure or charge-mobility simulations were beyond the present scope, the discussion now explicitly positions these DFT results as predictors of relative electronic sensitivity rather than absolute conductivity.

Glyphosate exhibits negligible adsorption on G–MBTS but large charge redistribution. The mechanistic explanation for this unusual combination is brief and warrants deeper discussion.

We expanded the discussion after Table 2 to explain that the weak stabilization arises from glyphosate’s highly polar, non-aromatic structure, which reduces π-surface contact. The large charge redistribution is attributed to long-range electrostatic polarization. The paragraph also suggests that co-functionalization or explicit solvation could improve glyphosate affinity while retaining electronic sensitivity.

Some structural figures (e.g., BCPs in Fig. 5) may be difficult to interpret without clearer labeling or higher resolution.

Figure 5 was regenerated at higher resolution with larger numerical labels corresponding to the entries in Table 3.

Units for adsorption energies and charge transfer should be consistently presented and include the sign convention.

All adsorption energies are now expressed in eV following the corrected formula (see attached PDF)

with larger positive values indicating stronger adsorption.

Charge-transfer signs are defined consistently in all tables and captions: negative values denote electron gain by the species listed.

Table captions could better explain abbreviations (e.g., “FNT,” “TOK”) for readers outside the pesticide field.

The abbreviations of each OP are presented in the manuscript just before the mention of Figure 2. The caption of said figure has also been updated to include them.

Check references for formatting consistency and duplication (e.g., repeated PBE citations [17–19]).

The references were reviewed and reformatted according to MDPI guidelines. The two PBE entries were retained intentionally, with the 1997 citation marked as “Erratum”.

Language is generally clear but can be tightened for conciseness in some introductory passages.

We thank the reviewer for this suggestion. The manuscript has gone through a general revision.

How would the inclusion of explicit water molecules or finite-temperature molecular dynamics affect the adsorption energies and charge transfer results?

A new paragraph titled “Thermodynamic considerations” discusses this explicitly. We note that explicit solvation and finite-T dynamics would reduce adsorption magnitudes due to entropic effects and continuous hydrogen-bond rearrangements. Nevertheless, the qualitative hierarchy of binding and charge-transfer trends is expected to remain, as these are governed primarily by molecular polarity and surface contact area.

Did you consider alternative graphene functionalizations (e.g., defects, oxygen groups) that may coexist with MBTS in practical materials?

Yes. This possibility is now mentioned in both the Results and Conclusions. We indicate that oxygenated defect sites or co-functionalization could further stabilize polar analytes (such as glyphosate) without compromising electronic sensitivity, and we suggest that future studies should explore these mixed-surface effects.

How sensitive are the adsorption energies to the choice of functional and basis set? Have you benchmarked with a higher-level method?

Alternative functionals and larger basis sets could shift absolute energies but the relative ordering among OPs should be robust. While we did not benchmark with a higher level method in this work, this has been our experience with previous studies.

Could you provide free-energy estimates (e.g., through vibrational entropy corrections) to evaluate adsorption at ambient conditions?

A discussion addressing this has been integrated into the Thermodynamic considerations paragraph. We explain that finite-temperature and entropic contributions will lower adsorption magnitudes but maintain the same qualitative ranking. Including full vibrational corrections would not alter the interpretation that G-MBTS acts as a reversible charge-transfer sensor.

What experimental techniques would you propose to validate the predicted charge-transfer enhancements (e.g., electrochemical impedance, XPS)?

The Conclusions section now explicitly mentions EIS, CV, and XPS as direct experimental probes of interfacial charge transfer.

Do you expect competition between multiple pesticides in a mixture to alter the binding hierarchy observed for single molecules?

Yes. Although our results point to a reversible, fast exchanging adsorption, competition is still expected at the G-MBTS interface as its site density is finite. We did not simulate co-adsorption in this work but its something to be taken into account for future experimental work.

Can the predicted charge redistribution be directly correlated with measurable changes in graphene conductivity or sensor response?

We would expect so. The text after Table 2 now explains how even small charge transfers can lead to measurable impedance or conductance changes. This connection is further supported by experimental analogs cited (Khan et al., RSC Adv. 2024).

Is the low adsorption energy but high charge transfer for glyphosate reproducible with different starting geometries or functionals?

As with the rest of the OP, we explored several initial placements and then proceeded with the lowest energy adsorbate pose found after geometry optimization. We did not compute adsorption energy or charge metrics for multiple converged poses nor re-optimize with alternative functionals. Thus, we refrain from claiming pose or functional reproducibility. First intuition might lead to thinking that every other conformation would have a smaller adsorption energy, as they had higher total energies, but this would not account for BSSE correction.

Have you explored the possibility of covalent bonding or chemical degradation of MBTS on graphene under operational conditions?

In the present DFT study we treat neutral, intact MBTS physisorbed on graphene and do not apply electrode bias or reactive pathways. However, in our prior experimental work with MBTS-modified carbon paste electrodes, we observed an oxidative process attributable to MBTS transformation under anodic conditions, which manifested as a distinct background peak and was discussed as electro-oxidation of MBTS (formation of 2,2′-thiobis(benzothiazole) and elemental sulfur).

Could defect engineering or co-modification (mentioned in the conclusions) be guided by specific computational descriptors from this study?

Yes, the present results identify two practical design descriptors that could guide defect or co-modification. Adsorption-induced charge redistribution and the complex’s HOMO-LUMO gap. Increasing the charge transfer while maintaining a narrow gap should maximize electrochemical sensitivity. The conclusions have been updated to state this.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript presents an entirely theoretical study of adsorption of several organophosphate pollutants on Mercaptobenzothiazole disulfide alone as well on Mercaptobenzothiazole on graphene.

The main (major) comment is about the magnitude of the adsorption energy.

As it is possible to show using a simple Arrhenius analysis (with a standard pre-factor of 10^13) for an adsorption energy of 0.73 eV the lifetime of the adsorbed molecule at 300 K is of the order of 0.2 s. So only MBTS-MAL (for which the calculated adsorption energy is 1.05 eV) would be stable at 300 K.

All the other molecules will desorb in fraction of seconds at room temperature!

The lifetime becomes even lower on graphene-MBTS. In this case also the most stable molecules considered here (MAL) has a computed adsoprtion energy of 0.8796 eV corresponding to a lifetime before desorption of 58 s.

Such short lifetimes imply that only at very high concentrations such molecule would have a significant probability to be detected.

Since no suggestion is given about which detection method might be used, the claims that this result ".highlight MBTS-functionalized graphene as a promising platform for the selective detection of
organophosphates in water." is not supported by the theoretical results shown in the manuscript.

In other words, at room temperature the lifetime of almost all the pesticides simulated here is so short that an extremely fast detection system would be required to detect them at low concentration.

So this calculations actually demonstrates the un-suitability of graphene-MBTS for the detection of all the molecules studied here, except MAL, unless extremely fast detecton methods (that are not even suggested) are used.

So I think the manuscript is possibly only of computational interest and for this reason I cannot recommend its acceptance here.

I have moreover the following minor comments:

1) According to equation (1), the adsorption energy should be negative for adsorption to be energetically favoured. Why are all the values positive?

2) Please define BCP acronym.

Author Response

We thank Reviewer 3 for the careful reading of our manuscript and for raising an important point regarding the physical meaning of the computed adsorption energies and their connection with the sensing mechanism. We have carefully addressed all concerns and revised the text accordingly. Reviewer comments are shown in black, and our responses appear in blue.

The manuscript presents an entirely theoretical study of adsorption of several organophosphate pollutants on Mercaptobenzothiazole disulfide alone as well on Mercaptobenzothiazole on graphene.

The main (major) comment is about the magnitude of the adsorption energy.

All the other molecules will desorb in fraction of seconds at room temperature!

Such short lifetimes imply that only at very high concentrations such molecule would have a significant probability to be detected.

In other words, at room temperature the lifetime of almost all the pesticides simulated here is so short that an extremely fast detection system would be required to detect them at low concentration.

So I think the manuscript is possibly only of computational interest and for this reason I cannot recommend its acceptance here.

We thank the reviewer for this insightful analysis. We fully agree that adsorption energies of 0.5–0.9 eV correspond to short residence times at room temperature. However, this behavior is not detrimental for electrochemical sensing, which relies on reversible charge-transfer interactions rather than long-lived physisorption. To clarify this point, we have:

Added a new paragraph titled “Thermodynamic considerations” explaining that the reported energies are 0 K electronic adsorption energies including dispersion and BSSE corrections. We discuss how entropy and solvation effects will further reduce adsorption strength, leading to rapid exchange analytes, which would be advantageous for fast, reversible electrochemical readout.
Clarified the detection context in the results and conclusions, explicitly stating that G-MBTS functions as a charge transfer sensor and not as a permanent adsorbent.
Specified feasible detection techniques (EIS, CV and XPS) that operate within the timescales expected from our calculations.

I have moreover the following minor comments:

1) According to equation (1), the adsorption energy should be negative for adsorption to be energetically favoured. Why are all the values positive?

We appreciate the opportunity to clarify this point. The equation has been corrected in the manuscript and it has been clearly stated that larger positive values correspond to stronger adsorption.

2) Please define BCP acronym.

The acronym has been defined at its first appearance in the Computational Details section and again in the Results section describing the QTAIM analysis.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

This revised manuscript can be accepted now.

Author Response

This revised manuscript can be accepted now.

We thank the reviewer for their positive assessment and recommendation for acceptance.

Reviewer 3 Report

Comments and Suggestions for Authors

The reply of the authors is reasonable and in the revised version they have specified that "sensing performance originates from charge-transfer
modulation rather than strong chemisorption. Therefore, the system should be considered a charge-transfer sensor rather than a permanent adsorbent. Further sensing engineering could be guided by following the computed FMOs and charge redistribution. Experimental validation through electrochemical impedance spectroscopy, cyclic voltammetry, and X-ray
photoelectron spectroscopy would provide direct evidence of the predicted interfacial charge transfer".

So in general I think the paper can be published in the present revised form. However, addition of some references demonstrating that, despite the low adsorption energy, charge transfer can be used for detection by electrochemical methods would make their claims less generic and more supported.

Author Response

We thank the reviewer for this valuable suggestion.

Three studies have been added to the Conclusions which support that physisorption-mediated charge transfer can produce strong electrochemical signals despite low adsorption energies.