The manuscript “Dual roles of inorganic aqueous phase on SOA growth from benzene and phenol” provides new insight into the SOA formation processes from the oxidation of gaseous benzene and phenol under various HC:NOx ratios. To date, experimental studies show a negative related NOx dependence of SOA formation yield from the oxidation of aromatic hydrocarbons. The work presented herein combines experimental chamber investigations with a complex modeling system to deeply explore the heterogenous chemistry within SOA particles with respect to various relevant environmental parameters (i.e., acidity of SOA particles, SOA thermodynamical equilibrium, partitioning coefficients, temperature and RH). Authors employed a variety of modeling tools and used available atmospheric databases (MCM, EPI Suite) to design a tool for predicting the SOA mass under different atmospheric conditions by mean of heterogenous reactions in a two media particle system (inorganic/organic liquid phases) and by gas-particle partitioning processes. Acid-catalyzed formation of a persistent phenoxy radical (PPR) in wet inorganic aerosols and its desorption into the gas phase is hypothesized to be responsible for ozone consumption, thus lowering the atmospheric oxidation capacity near human settlements. Significant improvements were made to the in-use UNIPAR model by integrating HOM and H-PPR sequences to accommodate a new gas mechanism driven by the oxidation of benzene and phenol.

Both the experimental and the modeling part are well presented through the manuscript. I recommend this manuscript for publication in ACP after the following concerns are addressed.

Initial manuscript evaluation: Major revisions

You may be more explicit in the abstract about the “Dual roles of inorganic aqueous phase”. For instance, “Data presented herein highlights the impact of aqueous phase on SOA generated through benzene and phenol oxidation. The roles of the aqueous phase consist in: (1)…. and (2)… .

A discussion regarding minimal incremental reactivity index (MIR) (Carter, 1994/ https://doi.org/10.1080/1073161X.1994.10467290) and photochemical ozone creation potentials (Jenkin et al., 2017/ https://doi.org/10.1016/j.atmosenv.2017.05.024) of monocycle aromatics would add considerable impact to your current findings and highlight the atmospheric implications.

To what extent would the competing reaction of PPR with the dissolved NO₂ in the inorganic phase affect the UNIPAR/H-PPR model (Kleffmann et al., 1998/ https://doi.org/10.1016/S1352-2310(98)00065-X)? Same question for the catechol gas-phase reactions with ozone (Obeid et al., 2024/ https://doi.org/10.1016/j.envpol.2023.122743; Coeur-Tourneur et al., 2009/ https://doi.org/10.1016/j.atmosenv.2008.12.054; Thomas et al., 2003/ https://doi.org/10.1002/kin.10121)

How is k_{off_phenoxy} calculated? Is it assumed to be equal to k_{off_phenol}? If so, explain why and how an order of magnitude in between the considered value impact the model? Does the model incorporate Leighton equilibrium in predicting the gas-phase O₃, NO₂ and NO concentrations?

Kwok and Atkinson SAR on monocyclic aromatics follows the regression log (k/cm³ molecule^-1 s^‐1) = ‐11.6 – 1.39 Σσ⁺, where σ⁺ are the Hammett constants for electrophilic substitution by Brown and Okamoto (1958/ https://doi.org/10.1021/ja01551a055). If you are using EPI Suite software to estimate the gas kinetic rate coefficients for multi-hydroxy benzenes with vicinal OH groups the software may underestimate the values (Roman et al., 2022/ https://doi.org/10.5194/acp-22-2203-2022).

Also, you could calculate and provide in the discussions sections a relative drop in NO₂, O₃ and SOA mass concentration when applying the UNIPAR with and without H-PPR.  

Using the current dataset for the UNIPAR/H-PPR, could you estimate the SOA mass distribution from the oxidation of 2-methylphenol and catechol under similar conditions?

Technical corrections, minor questions and suggestions:

Affiliation is not indicated for the authors.

Abstract

L10: gas oxidation or phenol or benzene… > gas oxidation of phenol or benzene…

L25: oxidation, about 53% of the… > oxidation, up to 53% of the…

Across the manuscript you have no consistency expressing the units (i.e., L227: g mol^-1, L241: g/L). Choose one way to express the units.

Introduction

L41: oxidation rate (i.e., 1.21571E-12 at 298K) > oxidation rate (i.e., 1.22 × 10^-12 cm³ molecule^-1 s^‐1 at 298K) [REFERENCE NEEDED]. Be consistent with the units and the order of magnitude across the manuscript and the supplement material.

L41: but its SOA yield is high > [provide a range for observed SOA formation yield and the corresponding cited paper/ papers].

L59: The lifetime is long also due to a p-π conjugated system also help for stabilizing the phenoxy radicals.

L85: delete “^{4-9, 52}”.

L86: of phenol or benzene > of phenol and benzene

Experiment section

L 109: Specify the instrument and the operating conditions used to monitor the HCs concentration presented in Fig 3. What were the sensitivities and the corresponding relative uncertainties for NO/NOx (Villena et al., 2012/ https://doi.org/10.5194/amt-5-149-2012) and O₃ (Spicer et al., 2012/https://doi.org/10.3155/1047-3289.60.11.1353) photometers? In what extent these uncertainties would affect the experimental findings?

L 116: Regarding the SOA seeds, were particle diameters the same for all experiments? Do you account for differences in SOA surface concentration in the UNIPAR model?

L 117: sulfate, ammonium, nitrate ion peaks in aerosol. > sulfate, ammonium and nitrate ion signals in aerosol phase.

L 120: species (Sulfate, nitrate… > species (sulfate, nitrate…

UNIPAR SOA model

L153: You stated that “Both organic-phase oligomerization and aqueous reactions of reactive species in inorganic phase yield non-volatile OM in the model”. Except for PPR, right?

HOM Formation

L178: The reaction rate constants > The gas phase reaction rate coefficients

PPR Formation

L183: A citation is needed for the branching ratios.

L222: R is a gas constant (8.314 J mol^-1 K^-1). > R is a gas constant (8.314 J mol^-1 K-¹) and T the absolute temperature.

L245: k1 number and units. Also, a reference should be cited here for adduct formation.

L276: with H-PPR and without H-PPR. > with H-PPR and without (w/o) H-PPR.

Evaluation of the impact of H-PPR on SOA Formation: aerosol acidity

L306: connected tothe > connected to the

Sensitivity of SOA formation to NOx 320 level, Temperature, and RH

L321: temperatures (278K, 288K and 298K) > temperatures (278 K, 288 K and 298 K)

L324: 2022 (between 6:30 AM to5:30… > 2022 (between 6:30 AM to 5:30…

L336: 2.82E-11 cm³/molecule s^-1 at 298K >  …… and  [REFERENCE IS NEEDED}

L337: 1.2E-12 cm³/molecule s^-1 >

Conclusion and atmospheric implications

L370: the heterogeneously produced PPR production occurs via…. > the H-PPR occurs via….

L371: OH radical > OH radicals that are

References

L444: New York2002

L460: doi [REMOVE UNDERLINE]

L464: New York2002

L489: doi [REMOVE UNDERLINE]

L493: with NO 2> with NO₂

L510: doi [REMOVE UNDERLINE]

L518: doi [REMOVE UNDERLINE]

L525: the absence of NO_x, the absence of NO₂

L530: doi [REMOVE UNDERLINE]

L541: doi [REMOVE UNDERLINE]

L548: doi [REMOVE UNDERLINE]

L554: p-amino… > p-amino…

L563: doi [REMOVE UNDERLINE]/check doi

L572: doi [REMOVE UNDERLINE]

L580: m-xylene/doi [REMOVE UNDERLINE]

What were the wall deposition and the dilution rates for SOA, phenol, ozone and NO₂.

L624 and L625: 4:00 > 16:00 (as in Figs. 3 and 4)

L633: HO2 ….. RO2 > HO₂ ……RO₂

L639: Figure 2: the k₂ decomposition coefficient should be changed to match k_phenoxy used in the manuscript body.

L642: k1 subscript

L645: Figure 3 and Figure 4 could be split into two parts, one for phenol and one for benzene. Try to use the same style (denotions, legend) for all the figures presented in Fig 3. and Fig. 4. Some left/right and down ticks would fit decently for all the figures. In Fig 3(B) and 3(C), Fig 3(H) and 3(I) use the same scale for better comparison. There is a different trend of HC in Fig 3 (A), (B), (C) in the first two hours compared with others for phenol. In Fig 3 (G-L) you have some variations. Are those in the uncertainty domain for your measurements? Fit the expiatory text of the experiment in the corresponding figure.

L646: Use subscripts for the inorganic species in figures and also in figure capitations!

L663: 2ppb > 2 ppb; L664: 298K > 298 K

L666 and L675: Same observations as in other figures. Figure 7 (C) scale, axis titles and legend color are different. Subscripts for OMAR and OMP to be consistent with the text. 298K > 298 K.

Supplement material

Stoichiometric coefficients

Please verify Eq. 13 > A1, B1, C1, D1 parameters!

Section 4: Check the subscripts for chemical compounds (i.e., H2O2) and superscript for units and large numbers. Italic font for notations (i.e., k_ph)

The study by Jiwon Choi et al. “Dual roles of inorganic aqueous phase on SOA growth from benzene and phenol” is a combination of experimental and modeling data that give an insight on oxidation of gaseous benzene and phenol and the formation of secondary organic aerosol. In this work the authors showed negative relation of SOA growth (of phenol and benzene) to NO_x levels in high NO_x regions by using several databases and models. Furthermore, the simulations in the current work showed the aspect of increasing significance of persistent phenoxy radical with decreasing NO_x levels. Significances of persistent phenoxy radicals that are produced during wildfire plumes and their impact on retarding the atmospheric oxidation in urban areas are highlighted.

The manuscript is well written and fits well in the scientific scope of ACP. I recommend the manuscript to be published after some minor revision:

- Line 63: I suggest to add an author to the UNIPAR model: https://doi.org/10.5194/acp-14-4013-2014 .

- Line 77:  You might specifiy: “In this study, we hypothesize [based on chamber experiments and complex model data] that the production …”.

- Line 85: citation style changed or the numbers need to be deleted.

- Line 152: Do you also mean to include the persistent phenoxy radicals or not? “Both organic-phase oligomerisation and aqueous reactions of reactive species in inorganic phase yield non-volatile OM in the model”.

- Line 183: There is a citation/reference missing for the values in the brackets.

- Line 336/337: There is a citation/reference missing for the values in the brackets.

- Figure S1: Please check the caption of the second y-axis.

General comments:

- missing affiliation indication for authors.

- check the subscripts for chemical compounds in the text, figures and figure capitations.

- check the consistency of the units in the manuscript and remain with one style.

- Reference section: please remove the lines under the DOI link.

The manuscript “Dual roles of inorganic aqueous phase on SOA growth from benzene and phenol” provides coupled experimental and modelling evidence of the suppression of atmospheric oxidation capacity and SOA growth due to the formation of persistant peroxy radicals formed during the production of benzene and phenol derived SOA. The Heterogeneous Persistant Phenoxy Radical Model was derived with a new explicit mechanism for the formation of HOM and H-PPR and utilised in the UNIPAR model to predict the formation of SOA from multiphase reactions of phenol and benzene. The addition of H-PPR into model was found to increase suppression of SOA growth with this suppression found to further increase with increasing aerosol acidity.

After the following concerns are addressed, I recommend this manuscript for publishing by ACP.

Major corrections:

While the importance of this work on urban areas specifically is mentioned, I don’t feel this is quantitively explored enough in the implications section. I would suggest restructuring this section as the conclusion currently reads more like a discussion to me with new ideas still being introduced (i.e. ln 406” Phenol is the most abundant first-generation product from the oxidation of benzene…”) and a lot of use of generalisations in language such as “about” or “generally”. Slightly more quantitative implications would help to cement the importance of this work. What urban areas of the world is this most likely to effect? Areas with more biomass burning and wildfires or areas such as Chinese megacities and haze dominated regions?

Section 4.1 – suggest slight restructuring/rewording for increased clarity as it is a bit difficult to follow at present.

Adding in the factors of suppression for the different model scenarios may help to add context to level of suppression of SOA growth exhibited. At present, throughout the manuscript this is not directly given a number.   

Have you considered natural emissions of benzene and phenol such as over polar oceans or in the marine boundary layer? (Wohl et al., 2023 Sci. Adv.)

How competitive is NO₃ oxidation of phenol to give C₆H₅O compared to the reactions with OH and O₃? Is this significantly fast as a dark reaction to be considered important? When NO3 is the oxidiser is the oxidation capacity of the system still suppressed with increasing aerosol acidity?

Minor corrections:

Throughout the manuscript the consistency of inclusion of units and unit formatting needs to be checked as does the subscripts in names i.e. RO₂.

Throughout the manuscript there is repeated mentions of other phenolic compounds. In ln 281 it is mentioned that “The radical scavenging ability of phenols by forming phenoxy radicals is in the order of pyrogallol > 1,2,4- benzenetriol >catechol > hydroquinone > resorcinol ≈ phloroglucinol”. This being said, why was the focus of the study not expanded to include some of the more active phenols, especially as phenol can form  

Consistency in nomenclature is needed. For example pyrogallol and catechol are not given in chemical nomenclature, but 1,2,4-benzenetriol is. Suggest changing to hydroxyhydroquinone or changing the others, ie pyrogallol to 1,2,3-benzenetriol, or catechol to 1,2- benzenediol.

I suggest adopting a consistent colour for benzene and for phenol in all figures. Being different colours in every figure reduces readability.

Technical revisions:

Abstract:

1. Missing author affiliation numbers
2. Is the mention of NOx limited regimes worth highlighting more?
3. Add in what the implication of this research is atmospherically.

Introduction:

1. Add in some statistics and refs of the prevalence of benzene and phenol atmospherically.
2. Ln 35 – where globally is represented by 20 % and where 90 %?
3. Ln 45 – references needed
4. Where is the benzene oxidation path important? Why is it hard to study benzene oxidation specifically?
5. Expand the mention of wildfire SOA to mention briefly atmospheric implications
6. Ln 66- what relative humidity are you defining as “wet” and is it consistent over all experiments?
7. Ln 69 – add more recent references for aerosol acidity experiments (e.g. Deng, Yange, et al. "Temperature and acidity dependence of secondary organic aerosol formation from α-pinene ozonolysis with a compact chamber system." Atmospheric Chemistry and Physics8 (2021): 5983-6003. Surratt JD, Lewandowski M, Offenberg JH, Jaoui M, Kleindienst TE, Edney EO, Seinfeld JH. Effect of acidity on secondary organic aerosol formation from isoprene. Environ Sci Technol. 2007 Aug 1;41(15):5363-9. doi: 10.1021/es0704176. PMID: 17822103.)

Model description:

1. Ln 131 – typo, correct to UNIPAR
2. Figure 2 is too cramped with the ending ‘e’ of ‘intermediate’ cut off

Results and Discussion (Section 4.1):

1. Figure 3 – none of the axes are labelled or given units, these also need to be added to the figure caption.
2. Both Figure 3 and 4 need to be reworked to help readability. I suggest splitting the figure into benzene and phenol, or perhaps phenol have dashed lines and benzene filled. At present the figure is hard to digest. The figure titles are also overpowering and the legend is too small and needs reordering so the predictions and corresponding experiments are side by side.
3. Ln 264 –How is improved quantified? What is this in relation to?
4. Ln267 – should this be explained explicitly earlier as an indirect amplification of the scavenging of O₃?
5. Ln 272 – “accurate gas simulation” – why is it defined as accurate?
6. Ln 277 – how are you defining “highly acidic aerosol”?
7. Ln 278 to 283 - feel a bit out of place.
8. Ln 286 -289 – unclear why this is said here.
9. Figure 7 – axis C label and legend font colour should be black not grey. And plot A is a side-by-side boxplot while B and C are not.

Results and Discussion (Section 4.2):

1. Ln 297 – Is “FS value” defined previously? If not, define here.
2. Ln 299 – insert reference for deliquescence point of seed aerosol
3. Ln 308 – why do you think that a larger conc of inorganic seed suppresses SOA mass? – for a given RH there is less water per seed particle? i.e. each surface has a thinner liquid microlayer?
4. Ln 325 – why was 30 ppb chosen for the initial concentration?
5. Ln 336 – 2.82E-11 reformat as 2.82 10^-11 and provide reference
6. Ln 227 – 1.2E-12 reformat as 2 10^-12 and provide reference
7. Ln 336 to 340 – Is this repeated information, is it needed here?
8. Ln 350, Section 4.2.3 – is this section not also testing the senstivitiy, as opposed to the uncertainty?
9. Ln 355 – only the sensitivity of the benzene simulations are included here. What not also Phenol?

Supplementary:

1. Figure S2 is quite cramped – consider opening up the schematic more to improve readability
2. Throughout supplementary – subscripts on inorganic species names are missing as are superscripts for units