From the journal Environmental Science: Atmospheres Peer review history

11-May-2022

Dear Dr Ammann:

Manuscript ID: EA-ART-03-2022-000015
TITLE: Benzediols Induce Local Ordering of Water Molecules near the Liquid-Vapor Interface

Thank you for your submission to Environmental Science: Atmospheres, 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.

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/esatmos?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/esatmos) 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

Environmental Science: Atmospheres 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 from CASRAI, https://casrai.org/credit/) 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 http://www.rsc.org/journals-books-databases/journal-authors-reviewers/author-responsibilities/ for more information.

I look forward to receiving your revised manuscript.

Yours sincerely,
Dr Claudia Mohr

Associate Editor, Environmental Science: Atmospheres

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

Reviewer 1

Yang et al. present an interesting study of benzenediols and how these surfactants influence the molecular interfacial structure of the water solvent using a combination of XPS/NEXAFS and MD/DFT simulations.
The article is well-written and is in principle suitable for the journal’s scope. I would like the authors to address the following issues before I recommend the work to be published:

• A schematic of the experimental setup would be helpful to the reader
• For this experimental setup, doesn’t the liquid jet travel in vacuum? Can evaporative cooling of the jet play a role?
• Abbreviation “ML” in the last paragraph on page 2 should be defined properly, as should be “C1s” and “O1s” at first use
• Please be consistent in the notations, e.g. “O K edge” or “O K-edge”, “O1s” or “O 1s” etc.
• Results section: Figure 2 does not exist. Please re-number the figures consistently.
• Could the authors also discuss what role dissociation of the hydroxyl group can play in aqueous solution, how this might affect the spectra?
• The relevance of this work to atmospheric science is rather weak. Ice nucleation is indeed a most important field of research and the influence of the surfactants on the local molecular order is interesting. But even strong cooling enhances the local tetrahedral structure of water. Why are the investigated surfactants of particular interest in this context?
• The authors discuss pi-stacking which is likely driven by the mostly hydrophobic interactions of the surfactants: It is energetically favorable for the stronger molecular interaction between the water molecules to form and thus minimize the exposed hydrophobic parts of the surfactant. It is well known that the interfacial structure of water is modified in the presence of hydrophobic surfaces. Could the authors please discuss this perspective, too?
• It is still not clear to me what mechanism makes “…resorcinol and orcinol molecules locally induce a more tetrahedral coordination among water molecules.” The OH groups might do this locally, but it is difficult to see how this would give such a significant effect on the liquid’s interfacial structure. Could the authors please clarify?

Reviewer 2

Review report on “Benzenediols Induce Local Ordering of Water Molecules near the
Liquid-Vapor Interface” by Yang et al.

The manuscript addresses the possible ordering of water molecules at the water surface induced by two organic molecules, resorcinol and orcinol, in connection to ice nucleation. Aqueous solutions of resorcinol and orcinol have been studied using core and valence level photoelectron spectroscopy, electron-yield NEXAFS, and molecular dynamics simulations.

The photon-energy-dependent intensity of the C1s core level photoelectron spectra were used to study the surface propensity of the two organics.
The qualitative result that both molecules are surface enriched is solid, and the determination of the surface enrichment using a model represents an advanced attempt to quantify the surface enrichment. Some points regarding the core level photoelectron spectra:
The intensity of photoelectron peaks is angular dependent, and this angular dependence is in general photon energy dependent. IS this considered in the model, and if so, how?
On page 2, it is stated that “during the experiment, we made use of linearly polarized light at 0°”. Does this mean that the photoelectrons were detected parallel to the polarization direction? How would this affect relative intensities compared to using for example the magic angle?
On page 6, it is stated that the peak fitting was done with ”consistent full width at half maximum”. What does consistent mean in this context?

Molecular dynamics simulations qualitatively support the conclusions about surface enrichment of the organics. The MD simulations also indicate an increased ordering of the water molecules in the outermost layer(s). Rather small changes in the experimental O1s NEXAFS spectra were interpreted as supporting the organics-induced ordering of the water molecules in the outermost layer(s). Some points regarding the O1s NEXAFS spectra:
The O1s NEXAFS spectra were measured using electron yield in the kinetic energy range of 412 - 437 eV. Why was this lower kinetic energy range used instead of the O1s KVV Auger?
The O1s NEXAFS spectra will contain contributions from the oxygen atoms in both H2O and the organic molecule. Could this be the reason for the small observed differences in the NEXAFS spectra? Could the organics induce small changes in the electronic structure of the water which could cause the observed differences?
Theoretical spectral calculations starting from geometries generated by the MD simulations would be very helpful for the interpretation of the NEXAFS spectra.

Valence level photoelectron spectra were recorded for the solutions with the organics, and compared to neat water and ice. The spectra for water and the solutions are quite similar, but differ from ice. The differences between the spectra for water and the solutions is qualitatively discussed, but considered too small to make any definite conclusions.
Some points regarding the valence level photoelectron spectra:
The valence level photoelectron spectra were recorded with a photon energy of 600 eV, i.e. in a not very surface sensitive mode. Why was not a lower photon energy resulting in higher surface sensitivity chosen to emphasize the changes in the outermost molecular layers?
The valence level photoelectron spectra will contain contributions from both H2O and the organic molecule. Could this be the reason for the small observed spectral differences?
On page 9, spectral peaks are denoted “absorption features”. This is an unusual nomenclature.

Some minor points:
In the introduction, “hygrophilic OH groups” are mentioned. Should that be hydrophilic?
The authors should check the numbering of the figures throughout the manuscript, as this does not seem to be consistent between the figures and the text.
Figure 7: The discussion around figure 7 starts with panel b, which in that case should be a. The authors should also make the color scheme described in the text, the figure caption and the figure consistent.

Reviewer 3

See attached file.

Response to Review Comments

Referee: 1

Yang et al. present an interesting study of benzenediols and how these surfactants influence the molecular interfacial structure of the water solvent using a combination of XPS/NEXAFS and MD/DFT simulations.
The article is well-written and is in principle suitable for the journal’s scope. I would like the authors to address the following issues before I recommend the work to be published:

Response: We would like to thank the reviewer for his positive assessment.

A schematic of the experimental setup would be helpful to the reader

Response: Since we have used the liquid jet XPS setup in a completely standard configuration described in previous publications, adding a figure in the main text is not warranted. We have added schematics of the core part of the setup to the ESI, Figure S1, to clarify the configuration and angles.

For this experimental setup, doesn’t the liquid jet travel in vacuum? Can evaporative cooling of the jet play a role?
Response: Yes, the chamber pressure is 10-3 mbar. Additional details have been added to the methods section, including the extent of evaporative cooling.

Abbreviation “ML” in the last paragraph on page 2 should be defined properly, as should be “C1s” and “O1s” at first use

Response: Done. In the revised version, we have removed that abbreviation; for the labels of the photoemission peaks, we have defined the label as referring to the core level from which the excitation occurs when they are mentioned first in the methods section.

Please be consistent in the notations, e.g. “O K edge” or “O K-edge”, “O1s” or “O 1s” etc.

Response: Done; O K-edge has been used everywhere; and ‘O 1s’ with space in between.

Results section: Figure 2 does not exist. Please re-number the figures consistently.

Response: We apologize for the oversight. Figures have been renumbered and all references checked everywhere, both in main text and ESI.

Could the authors also discuss what role dissociation of the hydroxyl group can play in aqueous solution, how this might affect the spectra?

Response: The pKa of resorcinol and orcinol is 9.4 and 9.6, respectively, thus dissociation is not relevant under neutral conditions as in our solutions. A note has been added to the methods section.

The relevance of this work to atmospheric science is rather weak. Ice nucleation is indeed a most important field of research and the influence of the surfactants on the local molecular order is interesting. But even strong cooling enhances the local tetrahedral structure of water. Why are the investigated surfactants of particular interest in this context?

Response: Organic monolayers have been brought up as potentially relevant ice nucleation relevant materials also for the atmospheric context (refs 11-20 in introduction). Phenolic species are ubiquitous in the atmosphere in emissions of biomass burning, thus e.g., in wildfire plumes. Their stacking behavior is much more pronounced than with simpler aliphatic fatty acids, so that ordered islands may exist in aerosol particles or cloud droplets at mass fraction for which ordered layers would not exist with fatty alcohols. This point and the atmospheric relevance of the chosen organics is now better introduced. Related to the point of tetrahedral ordering at lower temperature: indeed the population of tetrahedral configurations is increasing with decreasing temperature; water homogeneously nucleates at -38°C. Thus, the role of water ordering by substrates at higher temperatures is exactly one of the aspects important for heterogeneous ice nucleation. We have revised the initial part of the introduction about ice nucleation and the role of organic ice nucleating substances to clarify this.

The authors discuss pi-stacking which is likely driven by the mostly hydrophobic interactions of the surfactants: It is energetically favorable for the stronger molecular interaction between the water molecules to form and thus minimize the exposed hydrophobic parts of the surfactant. It is well known that the interfacial structure of water is modified in the presence of hydrophobic surfaces. Could the authors please discuss this perspective, too?

Response: If we correctly understand, the reviewer is wondering whether water clustering (by hydrogen bonding) can prevent the direct (hydrophobic) interaction among the surfactants at the interface. As the reviewer pointed out, the interfacial structure of water is modified by the presence of even small nonpolar solutes (1,2). Studies in literature have suggested that in the solvation of small nonpolar molecules (radius less than ~ 1 nm) the association of two solutes is weakly favorable and entropically driven. On the contrary, association of larger nonpolar solutes is highly favorable and enthalpically driven (2,3). Moreover, water hydrogen bonding strength is closer to nonpolar solutes(4,5). Based on all of that and since RES and ORC can be regarded as small solutes, the conclusions would be a weakly favorable association of the RES and ORC at the interface. However, RES and ORC have an amphiphilic and aromatic character, which makes their behavior different from simple hydrophobic cavities or aliphatic chains. As we also reported here in our manuscript for RES and ORC, in general at the air/liquid water interface amphiphilic solutes tend to be adsorbed with the hydrophilic part toward the liquid phase and the hydrophobic part exposed to the gas phase: This anchors the molecule to the condensed phase, which may favor the assembly of the adsorbates by hydrophobic interaction(1,6). In addition, pi-pi interaction energetically competes with water hydrogen bonding (e.g., benzene-benzene interaction energy in the gas is ~4kcal/mol) (7). Finally, to the best of our knowledge, the interplay between hydrophobicity and water hydrogen bonding at the air/liquid water interface is still not fully explored, at least for RES and ORC as a function of their interfacial concentrations. Our classical MD simulations report clear evidence of RES and ORC stacking at the interfaces, and similar features were also observed for other aromatic compounds in our previous works on the ice pre-melted liquid water surfaces (8,9). However, as we also remarked in the main manuscript, a more quantitative analysis of the adsorbate self-assembly calls for a better description of the pi-pi interaction, which is beyond the capability of the classical MD employed here and more suited for some ab-initio MD approach. We really appreciate this comment: more discussion on this aspect will certainly add value to the manuscript. A note with respect to the uncertainty of the calculation of the stacking behavior has been added.

Response: It is still not clear to me what mechanism makes “…resorcinol and orcinol molecules locally induce a more tetrahedral coordination among water molecules.” The OH groups might do this locally, but it is difficult to see how this would give such a significant effect on the liquid’s interfacial structure. Could the authors please clarify?

Response: Simulations studies of ice nucleation on hydroxylated inorganic substrate (e.g., kaolite) (10), alcohol monolayers (11), and Phloroglucinol Dihydrate (PD) crystals (12) have reported the formation of a ordered interfacial water layer in contact with the substrate, which precede the heterogeneous ice nucleation. The ordering of this interfacial water layer is controlled by the distribution of hydrogen bonding (-OH) and hydrophobic groups on the organic crystal surface(12): this and the molecular details of the interfacial substrate-water interaction (e.g., strength of hydrogen bonding between water and the organics (12) the interfacial electric field (13,14) etc.) determine the ice nucleation efficiency. In this work, the interfacial density (i.e., molecules/nm^2) of RES and ORC at the liquid water interface is lower than those experienced at the organic crystal/water interfaces discussed above, but the same features are still present here (even if the signals are less evident due to the lower interfacial concentration). The hydrophobic interaction among solutes orders the adsorbates at the interface, as shown in Figure 6 with an indication of self-aggregation that locally increase the interfacial concentration. This drives the interaction between the hydroxyl groups of the solutes and water to the formation of a more ordered interfacial water layer: this is evident in Figure 7c where the tetrahedral order distribution is more shifted to higher q-value with RES and ORC in solution (green and red line), than on the pure liquid water case (blue line). Also note that from the measured surface excess, we had roughly an already quite densely packed layer of RES and ORC for which we observed the trends towards more tetrahedral ordering in the NEXAFS spectra. To make the presence of an air/liquid water interface that is more ordered than the pure liquid water one even more evident, we added in the ESI figure S7. The figure shows the distribution of the order parameter, q, as a function of the coordinate perpendicular to the interface, Z. Z=0 corresponds to the bulk region. At the interface, defined as Z larger than the Gibbs Dividing Surface (GDS), the order distribution of liquid water is higher in the 2M ORC solution (green line) than on the pure liquid water one (blue line), in agreement with the conclusions from Figure 7c. We thank the reviewer for this valuable comment: we have now better highlighted this in the manuscript.

Referee: 2

The manuscript addresses the possible ordering of water molecules at the water surface induced by two organic molecules, resorcinol and orcinol, in connection to ice nucleation. Aqueous solutions of resorcinol and orcinol have been studied using core and valence level photoelectron spectroscopy, electron-yield NEXAFS, and molecular dynamics simulations. The photon-energy-dependent intensity of the C1s core level photoelectron spectra were used to study the surface propensity of the two organics.
The qualitative result that both molecules are surface enriched is solid, and the determination of the surface enrichment using a model represents an advanced attempt to quantify the surface enrichment. Some points regarding the core level photoelectron spectra:
The intensity of photoelectron peaks is angular dependent, and this angular dependence is in general photon energy dependent. Is this considered in the model, and if so, how?
On page 2, it is stated that “during the experiment, we made use of linearly polarized light at 0°”. Does this mean that the photoelectrons were detected parallel to the polarization direction? How would this affect relative intensities compared to using for example the magic angle?

Response: Reply to both points: we agree that we have not provided enough details about the normalization procedure and calculation of cross sections. In the revised submission, we have added Figure S1 to the ESI to clarify the geometric configuration with the incident photon beam, its polarization and the electron detection axis. Then, we have expanded the description of the calculation of the photoemission intensity ratios to explicitly include the beta parameter in the total ionization cross section that describes the dependence on the polarization angle and takes into account the orbital geometry.

On page 6, it is stated that the peak fitting was done with ”consistent full width at half maximum”. What does consistent mean in this context?

Response: Indeed, this was not entirely clear. This was meant to be the same widths for all carbon peaks. We have reformulated to clarify.

Molecular dynamics simulations qualitatively support the conclusions about surface enrichment of the organics. The MD simulations also indicate an increased ordering of the water molecules in the outermost layer(s). Rather small changes in the experimental O1s NEXAFS spectra were interpreted as supporting the organics-induced ordering of the water molecules in the outermost layer(s).

Response: We agree that the changes to the NEXAFS spectra were rather small and therefore used cautious language for the interpretation and only mention a ‘trend’ towards more tetrahedral coordination.

Some points regarding the O1s NEXAFS spectra:
The O1s NEXAFS spectra were measured using electron yield in the kinetic energy range of 412 - 437 eV. Why was this lower kinetic energy range used instead of the O1s KVV Auger?

Response: This kinetic energy window remains free from valence photoemission peaks travelling through while scanning the photon energy. Otherwise, a complex deconvolution process would be necessary to obtain the true Auger yield. This aspect has been added to the methods section.

The O1s NEXAFS spectra will contain contributions from the oxygen atoms in both H2O and the organic molecule. Could this be the reason for the small observed differences in the NEXAFS spectra?

Response: For the O 1s signal intensity contribution we have estimated that the density of the organic-OH-oxygens is about one order of magnitude less than those of water molecules within the topmost bilayer of water. The probe depth of the electron yield NEXAFS is between 2 and 3 nm, reducing this number to the percent level or below. We have added a note on this in the discussion.

Could the organics induce small changes in the electronic structure of the water which could cause the observed differences?

Response: Yes, the presence of the organic at the interface is changing the hydrogen bonding network and, thus, the electronic structure of water. The changes in the NEXAFS spectra reflect small perturbations in the outermost non-occupied molecular orbitals, but more through changes in the local geometry of the hydrogen bonding network rather than by changing the energy levels that would appear as shifts in the position of the absorption peaks.

Theoretical spectral calculations starting from geometries generated by the MD simulations would be very helpful for the interpretation of the NEXAFS spectra.

Response: Yes, we totally agree, and this is part of current work and interest in our group. NEXAFS calculations have become possible (e.g., see Discussion in Ref.15) but are still computationally and memory intense calculations, especially for soft matter systems, such those considered here. In soft matter systems, different solvent arrangement (e.g., hydrogen bonding arrangements) are possible, which calls for averaging many spectra calculated over different MD snapshots for a fair comparison with the experimental ones. At the present date, these are very intense calculations, which require preliminary time-consuming testing and fine tuning of the computational set-up.

Valence level photoelectron spectra were recorded for the solutions with the organics, and compared to neat water and ice. The spectra for water and the solutions are quite similar, but differ from ice. The differences between the spectra for water and the solutions is qualitatively discussed, but considered too small to make any definite conclusions.
Some points regarding the valence level photoelectron spectra:
The valence level photoelectron spectra were recorded with a photon energy of 600 eV, i.e. in a not very surface sensitive mode. Why was not a lower photon energy resulting in higher surface sensitivity chosen to emphasize the changes in the outermost molecular layers?

Response: The reviewer is correct. We chose a photon energy of 600 eV to be able to acquire the valence level photoelectron spectra (VLPS) of water, aqueous solutions and ice with the same probing depth. Indeed, as specified in the manuscript, the VLPS of ice has been acquired at the ISS beamline. While the SIM beamline is a double undulator, providing high photon flux in a broad energy range, the ISS beamline is a bending magnet, producing lower photon fluxes. A decent VLPS could be acquired only with 600 eV photon energy. Thus, the limitation came through the need to have a comparison between water and ice. A note has been added to the methods section.

The valence level photoelectron spectra will contain contributions from both H2O and the organic molecule. Could this be the reason for the small observed spectral differences?

Response: Yes, this is one of the reasons we were cautious about concluding from these spectra. We have attempted at obtaining valence spectra from pure condensed phase resorcinol at the ISS beamline as those for ice; however, the high vapor pressure did not allow stable measurements.

On page 9, spectral peaks are denoted “absorption features”. This is an unusual nomenclature.

Response: True; corrected.

Some minor points:
In the introduction, “hygrophilic OH groups” are mentioned. Should that be hydrophilic?

Response: Yes, corrected.

The authors should check the numbering of the figures throughout the manuscript, as this does not seem to be consistent between the figures and the text.

Response: Yes, revised and corrected.

Figure 7: The discussion around figure 7 starts with panel b, which in that case should be a. The authors should also make the color scheme described in the text, the figure caption and the figure consistent.

Response: Agreed; we have revised the order and labelling.

Referee: 3

Yang et al. present results from electron spectroscopy experiments and molecular dynamics simulations to describe the behavior of resorcinol and orcinol at the liquid-vapor interface. Both the findings from the simulations and the experiments show resorcinol and orcinol to be enhanced at the liquid-vapor interface. Both molecules induce ordering effects in the water molecules at the surface which could have implications on the ice nucleating activity of droplets containing resorcinol or orcinol. The results should be interesting for the atmospheric science community, but the link to compounds found in the atmosphere and processes that occur in the atmosphere needs to be strengthened in order to fit within the scope of the journal. I recommend publication after the major concerns below are satisfactorily addressed, and the manuscript is substantially revised to improve the clarity and readability to non-experts in photoelectron spectroscopy or molecular dynamics simulations.

Response: Thanks for the very detailed and thorough reading, understanding and providing a lot of useful comments. A review that we don’t see everyday anymore. We assume and hope the paper has been of sufficient interest to warrant the time invested!

1 Major concerns
The major weakness of the manuscript is its relevance to the atmosphere. The reader needs to be convinced that the results from this work are of some relevance to compounds and processes that occur in the atmosphere. This has not been satisfactorily addressed by the manuscript. This is not a commentary on the scientific merits of the experiments, simulations, and analysis. This is a major concern due to the scope of the journal.

Response: Accepted; the introduction has been expanded to better motivate and introduce the phenolic compounds. Though, we have kept it reasonably short.

1.1 The title
The title needs to be revised. Resorcinol is a benzenediol. Orcinol is not. It is a methylbenzenediol. Phenols would be a more accurate term, but it is a very large class of compounds. You could also just say resorcinol and orcinol if you want to be specific.

Response: Agreed; we have now changed to just mention orcinol and resorcinol in the title. Mentioning ‘phenols’ or ‘phenolic compounds’ would give the flavor that we claim our results apply to all phenolic compounds.

1.2 Phenolic compounds as atmospherically relevant compounds
Very little context for phenolic compounds in the atmosphere is given. There is only one reference cited (Hawthorne et al., 1988), and it is specific to wood smoke pollution from stoves. The reader needs to be convinced about the importance of phenols as a class of compounds that can be found in the atmosphere.

Response: done
1. How much mass of phenolic compounds can be found in the atmosphere? How are they distributed in the atmosphere? Are they highly localized? Do they have distinct vertical profiles? What is their lifetime in the atmosphere?

Response: Done, some information has been added to the introduction.

2. Introduction, paragraph 2. “The role of such fatty alcohol monolayers as IN active material has also been suggested to be of minor relevance, because the complex mixtures of organic material in atmospheric aerosol particles would rarely allow the formation of ordered monolayers." Couldn't the same be said of phenolic compounds?

Response: As explained just after that sentence, the stacking behavior is much different. Self assembly occurs at much lower concentrations than for fatty alcohols that require either compression or special properties of their 2D pressure-area phase diagrams. Mixtures would hardly form such ordered layers. We have reformulated this paragraph to make this more clear.

3. Besides burning wood, are there any other sources, either anthropogenic or biogenic, of phenols to the atmosphere?

Response: The sources are now better explained.

4. Do anthropogenic and biogenic sources produce different phenols?

Response: As mentioned, we are providing more information in the revised introduction, but also do not provide an extensive review of sources of phenols.

5. Do resorcinol and orcinol represent anthropogenic or biogenic phenols?

Response: They are just the most simple model species, representative of phenolic compounds irrespective of the real source in the atmosphere. We have avoided an extensive description of which ones may derive from biogenic or anthropogenic sources, since none of the source compounds is purely anthropogenic or biogenic. We now mention the two major pathways, biomass burning (which can be anthropogenic as well), and gas phase oxidation of aromatic compounds (from a variety of sources).

6. Calling resorcinol and orcinol “simple proxies of atmospheric phenolic compounds" implies that those two compounds are representative in some way of all phenolic compounds in the atmosphere. I find this doubtful, since Hawthorne et al. (1988) identified many heavier or polycyclic species from their wood smoke samples that probably wouldn't be well represented by resorcinol or orcinol. In fact, some polyphenols have been found to have anti-ice nucleating activity (Koyama et al., 2014).

Response: We have now called them ‘model compounds’, and have restricted them to monomeric phenols; we also mention the methoxy substituted compounds as a separate family, as we don’t want to make RES and ORC to be implicitly representative also of methoxyphenols without touching by experiment.

7. I didn't get a sense for why of all phenols that can be found in the atmosphere, resorcinol and orcinol were chose as “proxies." There are three benzenediol isomers. Why resorcinol and not catechol or hydroquinone? Was the choice of orcinol deliberate? That addition of the methyl group opposite the diol is pretty tantalizing. Was there a specific hypothesis being tested here?

Response: We have decided to start with two simple species with 2 OH groups to start with. Catechol is very reactive and would be difficult to keep stable for the experiments. Resorcinol has very high solubility; and the distance between the two OH-groups is similar to that in phloroglucinol. The methyl group in orcinol may support upward orientation; orcinol was thus expected to be more surface active, thus featuring high coverage at lower bulk concentration, and thus expected to exhibit less solute effects in the solution. We have added these aspects to the introduction.

1.3 Resorcinol and orcinol as ice nuclei
Ordering of water at the liquid-vapor interface is interesting because of the implications it might have on ice nucleation activity of compounds that can induce such ordering. The last sentence of the manuscript states, “Considering the presence of phenolic compounds in the atmosphere from both anthropogenic and biogenic origins, our findings are of potential importance for heterogeneous ice nucleation in the atmosphere." However, there is no discussion of the ice nucleation of phenolic compounds even though it has been the subject of past research. A cursory search of the literature found: 1. Parungo and Lodge (1965) measured the freezing onset temperature of resorcinol using a cold stage. The freezing onset temperature was predicted pretty well using Hammett's sigma function (Hammett, 1937). 2. Knollenberg (1969) measured the ice nucleation activity of resorcinol using a Bigg-Warner ice nuclei counter. Also provided is a solubility curve for resorcinol as a function of temperature. 3. Komabayasi and Ikebe (1961) studied the number of ice crystals formed from hydroquinone, an isomer of resorcinol, seed particles in a cold chamber. Is it necessary to know that resorcinol and orcinol inducing ordering of water at the surface when we can parameterize ice nucleation activity of those two compounds using measurements in an ice nuclei counter? Will this molecular level understanding be useful in atmospheric applications like weather or climate models?

Response: We thank the reviewer for pointing to these additional references, which we have added to the introduction. Clearly, parameterized ice nucleation rates from measurements can more directly be used in models. Our work contributes to better understanding why some materials nucleate ice better than others and thus adds to the scientific basis without providing parameters for models. We have added a sentence on this at the end of the introduction section and modified the end of the concluding paragraph.

2 Minor comments
2.1 Introduction
1. In the introduction, it should be stated which ice nucleation pathways are affected by ordering effects at the surface.

Response: Ok, added mentioning of heterogeneous ice nucleation more explicitly, along with the role of water ordering.

2. MD simulations are only given a brief mention in the introduction. Some sort of explanation as to what these simulations can do and what information they provide should be given in the introduction. This should help with the flow and readability in the methods section.

Response: In the revised manuscript, a paragraph introducing the simulations has been added to the introduction section.

2.2 Section 2.1
1. Paragraph 1. Some information about the solubility of resorcinol and orcinol would be helpful context. I assume the reason for the difference in concentration is due to the lower solubility of orcinol. Are the solutes completely dissolved at the measured concentrations?

Response: Yes, it is important to stay well below the solubility limit to avoid clogging in the liquid jet nozzle. The values for solubility have been added. And as also pointed out further below, ORC has in turn higher surface propensity, and essentially a monolayer is present at 0.2 M.

2. Section 2.1, paragraph 2: “. . . the capillary was surrounded by a cooling jacket to tune the liquid temperature." Tuned to what temperature and why?

Response: The operation of the jet is more stable at lower temperature; a lower background pressure can be achieved and less water needs to be pumped. A note has been added.

3. In section 2.1, the relationship between photon energy, photoelectron kinetic energy, and core level binding energy should be explicitly stated, otherwise readers unfamiliar with XPS will not understand how setting the photon energy from the beamline results in a given photoelectron kinetic energy.

Response: Agreed; we have added this information.

4. How is the inelastic mean free path calculated? It should be stated in section 2.1.

Response: We used the SESSA software (mentioned in the ESI); we have now added this information and a reference to this section.

2.3 Section 2.2
In general, this was a difficult section to read. If I were reading it in a published paper, I probably would have just skipped over it. I actually found the material in the supplement a bit easier to read in part because of the structure provided by the subheadings (“Simulation Box preparation" and “Molecular Dynamics. . . "). There is also considerable overlap between the text in the main manuscript and the supplement. I would suggest rewriting this section with a non-user of MD simulations in mind and leave the gory details to the supplement. For example, something like “Classical MD relies on a force field. . . " that appears in the supplement might be better served in the main manuscript.

Response: Accepted, in the revised manuscript, we have considerably simplified the description of the simulations and moved the details to the SI. We have tried to avoid redundancy as much as possible.

1. I don't understand the choice of concentrations chosen for the MD simulations. I now see this is explained in the supplemental material. Please move it to the main text. Why weren't 0.01 M solutions simulated? Is orcinol even soluble to 2 M? I see it's acknowledged in section 3.1 that 2 M is above the solubility of orcinol in water. So why do this simulation? Is the simulation illustrative of any physically realizable system?

Response: There are limitations in the system size that can be achieved by (even classical) MD simulations. To study the self-interaction of (for example) 8 solutes in a 0.01 M aqueous solution, at least 44000 water molecules in the simulation box are needed. Larger systems also mean a longer simulation time for equilibration and sampling. Regarding the 2 M ORC solutions, even if 2 M ORC is above the solubility of ORC in water, the simulation has still a physical meaning. For example, it allows us to compare ordering of water at similar RES and ORC concentration. Moreover, the 2 M ORC simulation permitted us to highlight the physicochemical mechanism of ordering of interfacial water, and make a comparison with those observed at the water-organic crystal interface. But we agree with the reviewer, this point has not been well enough explained in the manuscript. We have moved the discussion on the concentration from the SI to the main manuscript and expanded the discussion.

2. Paragraph three, starting from “In Table S3 and Figure S4. . . ". These are results. Move them to Section 3.

Response: done

3. Paragraph four. Explain the Ih abbreviation.

Response: done

4. Paragraph four: “All classical MD simulations were performed using. . . " This should be said earlier in the section.

Response: Done, moved to the first paragraph of this section.

5. Paragraph five: “DOS [sic] were average [sic] over 150 snapshots. . . " I think it's better to give the exact number.

Response: Done and language corrected.

6. Paragraph five: \DOSs for bulk ice. . .were calculated at the temperature T=Tm-14. . . " This is repeating what has been said in the previous paragraph. Just say they were done at the same temperature as the ice slab MD simulation.

Response: done

7. Paragraph five: “As [a] benchmark test, Figure S6 shows the DOS. . . " This is starting to sound like results, and, as written, this paragraph feels like it should belong in the supplement. Please consider streamlining this paragraph.

Response: done

8. Paragraph five: \CP2K molecular dynamics package was used to perform the DOS calculations." Again, should probably come closer to the beginning of the paragraph instead of the end.

Response: done

2.4 Section 3.1
1. I assume multiple C 1s spectra were added together to create the spectra plotted in Figure 1. What was done to ensure that the photoemission counts from 2 M resorcinol are comparable to the photoemission counts from 0.2 M orcinol?

Response: Orcinol is more surface active. Therefore, at low kinetic energy, the intensities are quite similar, because the surface coverage is similar, in spite of the lower bulk concentration. Intensities are directly comparable between the two graphs, as the y-axis is count rate, thus normalized to the number of sweeps taken during the acquisition. We have revised the figure and y-axis label to clarify this. We have also added discussion of this at the end of the first paragraph as it directly demonstrates the higher surface activity of ORC.

2. Paragraph 2: “the aromatic rings have an upward orientation with respect to. . . " Please refer the reader to Figure S2.

Response: Ok, done

3. Paragraph 2: “. . . which are located on the vacuum side of the interface" Maybe a mention of the pressure inside the measurement chamber in section 2.1 would be helpful.

Response: Ok, done, also in response to the other reviewers.

4. Paragraph 2: "The reason for the apparently less pronounced impact of orientation on the C1s component fractions for RES than for ORC is that the bulk concentration of RES was much higher than that of ORC. . . " What about when the bulk fractions were the same? You have measurements of C 1s component fractions for orcinol and resorcinol both at 0.01 M.

Response: We have added more information about the meaning of the kinetic energy dependent measurements to make this more clear. We have measured 0.01 M solutions, but the signal intensity ratios were scattering too much for a meaningful analysis of the orientation due to too high signal-to-noise levels. We could use these though to derive the surface coverage later from the total C 1s intensity.

5. Paragraph 3. More information about the C/O ratio with depth needs to appear in the main text.

Response: Ok, we considerably expanded this paragraph to explain how this ratio can be intuitively understood for a surfactant system.

6. Paragraph 3. "e.g. inelastic mean free path of the photoelectron, bulk element atomic densities, photon flux, total photoionization cross section and geometry of the experimental setup." Please make it clear if this is a complete list of factors effecting photoemission intensity. "Geometry of the experimental setup" is vague. Please explain. Does it include the alignment with the electron analyzer and the transmission function through the analyzer?

Response: In response to this comment, we have extended the description of the normalization procedure and calculation of photoemission intensities in the methods section and in more detail in the ESI to make it clear.

7. Paragraph 4. "our analysis does not consider whether they randomly float on the surface or aggregate into islands." You have non-zero amounts of C 1s signal past the most surface sensitive mean escape depth, so the molecules don't all go to the surface. What does that say about the validity of your model?

Response: Note that at the larger kinetic energy, the signal is still dominated (80%) by atoms within the first layer of thickness corresponding to the mean free path. Thus, the profile of the ratio is representing a cumulative depth profile. Yet, molecules in the bulk have an increasing contribution at larger kinetic energy, leading to changes of the slopes in Figure 4. We note for the time scales of the liquid experiment, bulk – surface equilibrium is established, and the bulk concentration is practically unaffected by the enrichment at the surface. As mentioned above, the interpretation of the C/O ratio as a function of kinetic energy are now better introduced in the previous paragraph. We have added a note at the end of this paragraph to mention that the bulk concentration remains essentially unaffected when the surfactants populate the surface.

8. Paragraph 3-4. the D parameter is not explained in the main text. It makes reading Figure 4 extremely confusing. Consider choosing a different symbol.

Response: Ok, we agree; we completely changed the presentation of the fits, by directly fitting the calculated signal intensity ratios to the measured ones. Results are identical in terms of the layer thickness, but easier to understand and describe.

9. Paragraph 5. Give a reference for the orcinol solubility.

Response: We have added values for the solubility in the methods section.

2.5 Section 3.2
1. What is the relevance of the 1 M NaCl NEXAFS spectrum?

Response: It was meant to be a reference for a case, where the opposite effect is observed for an established system in the bulk, to kind of putting the observed effects into perspective. We have now removed the NaCl spectrum to keep the figure simpler. Though, we have put a sentence in the text in the next paragraph about the solute effect of simple salt ions.

2. ". . . so that electrons contributing to the electron yield may. . . originate from deeper in the bulk than the MED at this kinetic energy." Can you provide an estimate of the contribution of these electrons that come from deeper in the bulk?

Response: Unfortunately, we are not yet in the position to quantify this effect, which is the topic of an ongoing project.

2.6 Section 3.3
1. “The noise level of the spectra does not allow making a statistically significant statement." Was this due to insufficient measurement statistics, or is it a limitation of valence photoemission measurements?

Response: We believe that this is mostly due to insufficient measurement statistics. While the acquisition time can in principle be extended, small movements of the jet also lead to fluctuations and thus more variability that then affects the ratio between gas phase and condensed phase contributions. Though, for the present dataset, the conclusiveness is mostly limited by the poorer signal-to-noise ratio of the ice spectrum, which was measured at another beamline with a lower photon flux. A note has been added after this sentence.

2.7 Figures
After the first, the remaining figures are misnumbered. I refer to them as they appear in the captions.

Response: We are sorry for this mistake. We have carefully revised and checked all figure numberings in main text and ESI.

1. The image quality of Figure 1 is poor.

Response: We have redone Figure 1, in response to another comment above. We are providing a high resolution version with the submission.

2. The quality of Figure 3 is poor. The top panel is slightly narrower than the bottom panel. The y-axis caption for panel (a) overlaps the axis tick labels. The numbers on the top axis aren't level. There appear to be a lot of artifacts in the images, for example (a) and MED.

Response: We have redone this figure, now Figure 2, with better aligned labels and panels.

3. Please make all four subplots in Figure 4 the same size.

Response: We have redone this figure, now Figure 3, in response to another comment above.

4. Figure 5. Consider making the lines in subplots (b), (c), and (d) thicker. They are a bit difficult to read as is.

Response: We have provided high resolution images with the submission to resolve this. Since the lines are fairly close to each other, thicker lines seem not ok.

5. Figure 7. I understand why you might want to explain panel (b) first, but it looks really strange. You could try swapping panels (a) and (b). Please also explain the meaning of the shaded region in the caption.

Response: Agreed, as mentioned in response to another comment above, we have rearranged this figure, now Figure 6, and we have added the information about the shaded regions to the caption.

6. Figure 8 has been modified from the analogous figure in Dr. Yang's thesis. Please revise the caption to reflect this. There is no x-axis label for panel (b).

Response: Agreed, for this figure, now Figure 7, we have removed the difference spectra from the earlier version provided in Dr. Yang’s thesis. We have revised the caption to correct this. The X-axis label has been changed to ‘binding energy (eV)’

2.8 Supplement
1. Please number your sections. It is unnecessarily difficult to find things in the supplement, and I was forced to search in the supplement quite often.

Response: Agreed, we have numbered all ESI sections and provided references to these in the main text.

2. Page S3-4. Please provide references for the density of resorcinol and orcinol in their pure condensed phase.

Response: Ok, done.

3. Page S4. How do you determine the factor A? Same for the factor B on page S5.

Response: This has now been clarified; the unknown transmission function contained in both A and B cancels in the ratio. This has been made more explicit.

4. Page S4. “Since the surface excess of both orcinol and resorcinol solutions are about the same. . . " Rewrite for clarity. Breaking this sentence into two or three will probably help.

Response: Ok, done.

5. Page S6. Do you use surface tension measurements to derive any Γ values? Remove if you don't.

Response: Agreed, we have removed this, as we have no surface tension measurements or data available for these solutions. Though we have added a sentence to the main text to mention that the obtained surface excess is similar to that for phenol and cresol, with a corresponding reference.

6. Page S7. Change “vapour" to “vapor" for consistency.

Response: Ok, done; we assume the editors will correct US vs UK English as appropriate.

7. It took me a while to figure out what NVT and NpT meant. I think the abbreviation is unnecessary.

Response: We have now explained the abbreviations NVT and NpT in the manuscript. We prefer to keep them because they allow an MD expert to immediately grasp the type of simulation.

8. Page S8. If you insist on also using the Celsius scale, the symbol requires the degree sign. Also decide whether you want to write “z-dimension" or “Z-dimension".

Response: Done, and everything is now given in K.

9. Page S9, “Gibbs Diving [sic] Surface and Solute Surface Excess". Move this section to come before the previous section. Rephrase the first sentence so it makes sense with the citation style.

Response: Done; this seems to have resulted from some autocorrect function.

10. Page S9. The singular of “species" is also “species."

Response: done

11. Figure S5. There is no label for the x-axis.

Response: Done

3 Technical and language comments
The manuscript and supplement could use a careful round of language editing. The use of the present continuous tense (e.g. “This effect is being discussed in terms of the formation. . . ") is a bit awkward. The copy editor will hopefully give you comments about stylistic consistency (e.g. whether units should be written as 2M or 2 M). Please also edit the supplement following the same guidelines.

Response: Ok, we have polished the language as far as possible within our capabilities; we have now used 2.0 M, O 1s, etc. throughout the text.

1. In the environmental significance statement: “Ice nuclation [sic] is still poorly understand [sic] at the molecular level. Some specific organic compounds are known as particulary [sic] good ice nuclei. . . "

Response: Ok, corrected

2. Edit for clarity:
(a) The last sentence of the abstract.

Response: Ok, adapted based on revised discussion of that part.
(b) Section 2.2, paragraph four: “. . . the tetrahedral arrangement of each water molecule with respect to the other." Should this be “with respect to its [four nearest] neighbors" or “with respect to one another"?

Response: Agreed, corrected

(c) Section 2.2, paragraph four: "The Ih ice slab was equilibrated. . . belonging to the water molecule i."

Response: Done
(d) Section 2.2, paragraph five: ". . . similar strategies have been already successfully explored. . ."

Response: Done

(e) Section 3.2. "Looking at the pure liquid water slab. . . this is why the p(q) peaks are at considerable smaller q-values. . . q-values are closer to the unity." Too long and too confusing.

Response: Done, split up into several sentences.

3. Introduction, first paragraph: ". . . terrestrial and oceanic water bodies" This terminology confuses me. Does "oceanic water bodies" refer to anything else besides oceans? Aren't all oceans terrestrial?

Response: Terrestrial water refers to all water on the land surface and in the subsurface, differentiated from that of oceans. But we have amended to make it more clear, so that all terms in that sentence are at the same level of hierarchy. ‘Aqueous or sea spray’ was also not ideal, since sea spray is most of the time already part of the aqueous aerosol category.

4. Introduction, first paragraph: “Despite its importance and century long investigations. . . " Can the authors give a reference for a century old investigation of atmospheric ice nucleation? DeMott et al. (2011) state in reference to ice nucleation research: “. . .when such research began in earnest more than 60 yr ago."

Response: To circumvent a discussion around when ice nucleation research really started, we have revised to refer to ‘a large body of literature…’.

5. Introduction, second paragraph. Is the choice of “hygrophillic" intended?

Response: Corrected to hydrophilic.

6. Section 2.1, paragraph 2: PEEK

Response: Ok, done

7. Section 2.1, paragraph 3. The “ML" abbreviation has not been introduced yet.

Response: Ok, abbreviation removed.

8. The numbering of the figures past the first one is incorrect.

Response: Ok, corrected.

9. In discussing the temperatures of the simulations, pick one temperature scale and stick to it.

Response: Changed to K everywhere.

10. The 0.2 M orcinol simulation is described as having both 5400 TIP4P water and ≈5400 TIP4P water. Be consistent, and be precise.

Response: Ok, clarified.

11. “Gibbs Diving Surface" appears seven times. It's Gibbs Dividing Surface, no?

Response: Of course, corrected

12. Please include article titles and DOI numbers in both reference lists! Please also ensure the references are stylistically consistent.

Response: We have used the most recent RSC style file for endnote for this journal.
 
References
1. Chandler, D., Interfaces and the driving force of hydrophobic assembly. Nature 2005, 437 (7059), 640-647 DOI: 10.1038/nature04162.
2. Dallin, B. C.; Yeon, H.; Ostwalt, A. R.; Abbott, N. L.; Van Lehn, R. C., Molecular Order Affects Interfacial Water Structure and Temperature-Dependent Hydrophobic Interactions between Nonpolar Self-Assembled Monolayers. Langmuir 2019, 35 (6), 2078-2088 DOI: 10.1021/acs.langmuir.8b03287.
3. Huang David, M.; Chandler, D., Temperature and length scale dependence of hydrophobic effects and their possible implications for protein folding. Proceedings of the National Academy of Sciences 2000, 97 (15), 8324-8327 DOI: 10.1073/pnas.120176397.
4. Grdadolnik, J.; Merzel, F.; Avbelj, F., Origin of hydrophobicity and enhanced water hydrogen bond strength near purely hydrophobic solutes. Proceedings of the National Academy of Sciences 2017, 114 (2), 322-327 DOI: 10.1073/pnas.1612480114.
5. Silverstein, K. A. T.; Haymet, A. D. J.; Dill, K. A., The Strength of Hydrogen Bonds in Liquid Water and Around Nonpolar Solutes. Journal of the American Chemical Society 2000, 122 (33), 8037-8041 DOI: 10.1021/ja000459t.
6. Gladich, I.; Berrens, M. L.; Rowe, P. M.; Pereyra, R. G.; Neshyba, S., Solvation and Stabilization of Single-Strand RNA at the Air/Ice Interface Support a Primordial RNA World on Ice. The Journal of Physical Chemistry C 2020, 124 (34), 18587-18594 DOI: 10.1021/acs.jpcc.0c04273.
7. Grant Hill, J.; Platts, J. A.; Werner, H.-J., Calculation of intermolecular interactions in the benzene dimer using coupled-cluster and local electron correlation methods. Physical Chemistry Chemical Physics 2006, 8 (35), 4072-4078 DOI: 10.1039/B608623C.
8. Kania, R.; Malongwe, J. K. E.; Nachtigallová, D.; Krausko, J.; Gladich, I.; Roeselová, M.; Heger, D.; Klán, P., Spectroscopic Properties of Benzene at the Air–Ice Interface: A Combined Experimental–Computational Approach. The Journal of Physical Chemistry A 2014, 118 (35), 7535-7547 DOI: 10.1021/jp501094n.
9. Heger, D.; Nachtigallová, D.; Surman, F.; Krausko, J.; Magyarová, B.; Brumovský, M.; Rubeš, M.; Gladich, I.; Klán, P., Self-Organization of 1-Methylnaphthalene on the Surface of Artificial Snow Grains: A Combined Experimental–Computational Approach. The Journal of Physical Chemistry A 2011, 115 (41), 11412-11422 DOI: 10.1021/jp205627a.
10. Cox, S. J.; Raza, Z.; Kathmann, S. M.; Slater, B.; Michaelides, A., The microscopic features of heterogeneous ice nucleation may affect the macroscopic morphology of atmospheric ice crystals. Faraday Discussions 2013, 167 (0), 389-403 DOI: 10.1039/C3FD00059A.
11. Qiu, Y.; Odendahl, N.; Hudait, A.; Mason, R.; Bertram, A. K.; Paesani, F.; DeMott, P. J.; Molinero, V., Ice Nucleation Efficiency of Hydroxylated Organic Surfaces Is Controlled by Their Structural Fluctuations and Mismatch to Ice. Journal of the American Chemical Society 2017, 139 (8), 3052-3064 DOI: 10.1021/jacs.6b12210.
12. Metya, A. K.; Molinero, V., Is Ice Nucleation by Organic Crystals Nonclassical? An Assessment of the Monolayer Hypothesis of Ice Nucleation. Journal of the American Chemical Society 2021, 143 (12), 4607-4624 DOI: 10.1021/jacs.0c12012.
13. Gladich, I.; Neshyba, S., Molecular Dynamics of Ice, Ice Surfaces and Impurities on Ice. In Chemistry in the Cryosphere, WORLD SCIENTIFIC: 2020; Vol. Volume 3, pp 173-257.
14. Glatz, B.; Sarupria, S., Heterogeneous Ice Nucleation: Interplay of Surface Properties and Their Impact on Water Orientations. Langmuir 2018, 34 (3), 1190-1198 DOI: 10.1021/acs.langmuir.7b02859.
15. Kühne, T. D.; Iannuzzi, M.; Del Ben, M.; Rybkin, V. V.; Seewald, P.; Stein, F.; Laino, T.; Khaliullin, R. Z.; Schütt, O.; Schiffmann, F.; Golze, D.; Wilhelm, J.; Chulkov, S.; Bani-Hashemian, M. H.; Weber, V.; Borštnik, U.; Taillefumier, M.; Jakobovits, A. S.; Lazzaro, A.; Pabst, H.; Müller, T.; Schade, R.; Guidon, M.; Andermatt, S.; Holmberg, N.; Schenter, G. K.; Hehn, A.; Bussy, A.; Belleflamme, F.; Tabacchi, G.; Glöß, A.; Lass, M.; Bethune, I.; Mundy, C. J.; Plessl, C.; Watkins, M.; VandeVondele, J.; Krack, M.; Hutter, J., CP2K: An electronic structure and molecular dynamics software package - Quickstep: Efficient and accurate electronic structure calculations. The Journal of Chemical Physics 2020, 152 (19), 194103 DOI: 10.1063/5.0007045.

26-Jul-2022

Dear Dr Ammann:

Manuscript ID: EA-ART-03-2022-000015.R1
TITLE: Orcinol and Resorcinol Induce Local Ordering of Water Molecules near the Liquid-Vapor Interface

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

After careful evaluation of your manuscript and the reviewers’ reports, I will be pleased to accept your manuscript for publication after revisions.

Please revise your manuscript to fully address the reviewers’ comments. 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.

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/esatmos?link_removed

(This link goes straight to your account, without the need to log in 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/esatmos) 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

Environmental Science: Atmospheres 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 from CASRAI, https://casrai.org/credit/) 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 http://www.rsc.org/journals-books-databases/journal-authors-reviewers/author-responsibilities/ for more information.

I look forward to receiving your revised manuscript.

Yours sincerely,
Dr Claudia Mohr

Associate Editor, Environmental Science: Atmospheres

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

Reviewer 1

I'm satisfied with the revisions and answers provided by the authors and recommend publication.

Reviewer 3

I would like to see a revision to the atmospheric relevance paragraph in the introduction and the correction of some errors before accepting the manuscript for publication. <br><br>

The addition of the paragraph dealing with the atmospheric relevance of phenolic compounds is welcome, but it is a bit off the mark. The cited literature will refer to phenols generally (Volkamer et al., 2002, Palm et al., 2020), methoxyphenols specifically (Hawthorne et al. 1988, Yee at al., 2013, Liu et al., 2022), or cresols specifically (Olariu et al., 2002), but never resorcinol or orcinol explicitly.<br>

Please read Finewax et al. (2019) (10.1021/acsearthspacechem.9b00112) and references therein for the atmospheric relevance of resorcinol. As for orcinol, Hawthorne et al. (1988) detected catechol in their wood stove burning samples and Alves et al. (2011) (10.1016/j.scitotenv.2010.12.025) detected both catechol and hydroquinone from wildfire samples, so you can argue that you are measuring a more stable and less toxic isomer of what's been detected in the atmosphere. You could additionally try to make a stronger argument for why resorcinol and orcinol are good model compounds for phenols found in the atmosphere. (Besides the diol is similar to phloroglucinol, which has high IN activity.)<br><br>

"For instance, phloroglucinol, probably one of the most IN active materials overall…"<br>
Please be careful with your citations. Reference [6] does not mention phloroglucinol.<br><br>

"Phenolic compounds occur as constituents in atmospheric particles in the ng to µg m^-3 range."<br>
The citation is specifically for methoxyphenols, of which orcinol and resorcinol do not belong.<br><br>

"While RES is very soluble (1000 g/L (20 °C))…"<br>
PubChem does not give this combination of solubility and temperature. The first entry is ≥100 g/L at 20 °C. 717 g/L is given a couple of times for 25 °C.<br><br>

"…the solubility of ORC is much lower (80 g/L (20 °C))."<br>
PubChem <i>does not</i> give the solubility of orcinol. This number is also different from the value quoted in section 2.2 (24 g/L).<br><br>

"We assume the density of RES and ORC in this layer to be the same as that of their pure condensed phase…"<br>
PubChem <i>does not</i> give the density of orcinol.<br><br>

Since the journal will not copy-edit the supplement:<br>
"Gibbs diving [sic] surface" still occurs twice in the supplement.<br><br>

Page S6, change "parametrized" to "parameterized."

Response to review comments – round 2

Referee: 1
Comments to the Author
I'm satisfied with the revisions and answers provided by the authors and recommend publication.

Response: We would like to thank this reviewer for the positive assessment.

Referee: 3

Comments to the Author
I would like to see a revision to the atmospheric relevance paragraph in the introduction and the correction of some errors before accepting the manuscript for publication.

Response: We would like to thank this reviewer for checking through everything again and pointing out the additional issues.

The addition of the paragraph dealing with the atmospheric relevance of phenolic compounds is welcome, but it is a bit off the mark. The cited literature will refer to phenols generally (Volkamer et al., 2002, Palm et al., 2020), methoxyphenols specifically (Hawthorne et al. 1988, Yee at al., 2013, Liu et al., 2022), or cresols specifically (Olariu et al., 2002), but never resorcinol or orcinol explicitly.

Response: The purpose was indeed to include studies that report a range of different phenol derivatives with different substitutions, including methyl and methoxyphenols. Olariu et al report methylbenzenediols, but not the same isomer as orcinol. We have added Schauer and Cass (2000) to include both phenols and these other species in terms of atmospheric concentration in the first general paragraph.


Please read Finewax et al. (2019) (10.1021/acsearthspacechem.9b00112) and references therein for the atmospheric relevance of resorcinol. As for orcinol, Hawthorne et al. (1988) detected catechol in their wood stove burning samples and Alves et al. (2011) (10.1016/j.scitotenv.2010.12.025) detected both catechol and hydroquinone from wildfire samples, so you can argue that you are measuring a more stable and less toxic isomer of what's been detected in the atmosphere. You could additionally try to make a stronger argument for why resorcinol and orcinol are good model compounds for phenols found in the atmosphere. (Besides the diol is similar to phloroglucinol, which has high IN activity.)

Response: Thank you for pointing to the Finewax et al work. In addition to references from therein, we have added a specific reference for orcinol (Vicente et al., 2019) and one including also other methylbenzenediol isomers (Ruzickova et al., 2022). This should provide enough credibility that methylbenzenediols are relevant species.


"For instance, phloroglucinol, probably one of the most IN active materials overall…"
Please be careful with your citations. Reference [6] does not mention phloroglucinol.

Response: Indeed, correct, Komabayasi et al. only mention pyrogallol, a different trihydroxybenzene isomer. So we removed the Komabayasi reference.

"Phenolic compounds occur as constituents in atmospheric particles in the ng to µg m^-3 range."
The citation is specifically for methoxyphenols, of which orcinol and resorcinol do not belong.

Response: We have added Schauer and Cass (2000) to include both families.

"While RES is very soluble (1000 g/L (20 °C))…"
PubChem does not give this combination of solubility and temperature. The first entry is ≥100 g/L at 20 °C. 717 g/L is given a couple of times for 25 °C.

Response: We have now corrected the solubility of resorcinol to 717 g/L at 25°C and cite Yalkowski et al. Instead of PubChem.


"…the solubility of ORC is much lower (80 g/L (20 °C))."
PubChem <i>does not</i> give the solubility of orcinol. This number is also different from the value quoted in section 2.2 (24 g/L).

Response: We have replaced PubChem by CHemSpider http://www.chemspider.com/Chemical-Structure.13839080.html, which lists estimates between 16 and 104 g/L. We were not able to find a compilation or original literature with an experimental value of the solubility of orcinol in water.

"We assume the density of RES and ORC in this layer to be the same as that of their pure condensed phase…"
PubChem <i>does not</i> give the density of orcinol.

Response: We have replaced these by references to ChemSpider for both species.

Since the journal will not copy-edit the supplement:
"Gibbs diving [sic] surface" still occurs twice in the supplement.

Response: Corrected!

Page S6, change "parametrized" to "parameterized."

Response: Done

17-Aug-2022

Dear Dr Ammann:

Manuscript ID: EA-ART-03-2022-000015.R2
TITLE: Orcinol and Resorcinol Induce Local Ordering of Water Molecules near the Liquid-Vapor Interface

Thank you for submitting your revised manuscript to Environmental Science: Atmospheres. 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 Environmental Science: Atmospheres. 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.

We will publicise your paper on our Twitter account @EnvSciRSC – to aid our publicity of your work please fill out this form: https://form.jotform.com/211263048265047

How was your experience with us? Let us know your feedback by completing our short 5 minute survey: https://www.smartsurvey.co.uk/s/RSC-author-satisfaction-energyenvironment/

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

With best wishes,

Dr Claudia Mohr

Associate Editor, Environmental Science: Atmospheres