This smog chamber study investigated the effects of NH₃, SO₂, and NO_x on SOA formation under UV irradiation from 1,2,4-trimethylbenzene (TMB) over a period of 5-6 hours. The smog chamber was monitored in real time using a scanning mobility particle sizer; gas analyzers for SO₂, NO_x, NH₃, and O₃; and gas chromatography-flame ionization for TMB. Offline analysis used infrared spectrophotometry, ion chromatography, and liquid chromatography-mass spectrometry. The study found that the presence of NH₃ and SO₂ both individually and synergetically increase SOA yield, and that SO₂ speeds nucleation, possibly through uptake onto H₂SO₄ surfaces. New organosulfates were identified and reaction schemes and structures were proposed for some; some organosulfates had molecular weights consistent with TMB-derived aerosol components found in the atmosphere. NH₃ was found to react to form organic nitrogen compounds in the aerosol phase, but only in the presence of SO₂, attributed to formation of ammonium sulfate. Aerosol components had a wide range of volatility almost nine orders of magnitude, as predicted by elemental composition from mass spectrometry.

This study is a good match for the scope and aims of ACP, and it does an excellent job conveying its own novelty even if the reader is not well versed in aromatic aerosol chemistry. It represents a necessary contribution to the aerosol research community's understanding of SO₂ and NH₃ dynamics, and has implications for reactions of a large number of aromatic VOCs as well as human health in the findings regarding enhanced UFP fraction. 

Major Comments

This article is solid, if not concise, and all sections contribute to the main point as a cohesive whole. The introduction, methods section, and discussion of limitations are excellent, and while there is far more description of results than there is analysis of results I believe that both the analysis and discussion of atmospheric implications are sufficient. All of my suggestions for improvement are minor, relating to clarity and readability.

Minor Comments

line 13: Consider a brief explanation of the atmospheric relevance of TMB in the abstract.

line 69-70: "Equivocally not originated from biogenic VOCs" is a little unclear. Possibly change to "unidentified OSs with C2-C25 skeletons that may not have originated from biogenic VOCs" or similar.

line 254-55: Make it clear whether ammonium sulfate particles were introduced during experiments or as an independent wall loss experiment. Is this dependent on humidity or hygroscopicity of the particles? Not VERY important, but may be a limitation worth discussing.

line 284: Were these from two single experiments? If so, include experiment numbers. Otherwise I want to know the standard deviations for these values.

line 291-95: Before noting that UFP are more harmful to human health, you may want to make it explicit in the text analysis of figure 1 that increased SO₂ concentration increases the fraction of UFP in the aerosol size distribution.

line 323: I would be interested in more detail about the mechanics of H₂SO₄ as a "condensed surface for key compounds." I take this to mean reactive uptake or heterogeneous reactions of VOCs with H₂SO₄. However, you don't justify this assertion; I might recommend citing Wang et al. 2010 (doi.org/10.1021/es9036868) and/or Zhang et al. 2019 (doi.org/10.1021/acsearthspacechem.9b00209) for VOC heterogeneous reaction with sulfuric acid surfaces.

line 329-30: As neither this work nor that of Julin et al. attempts to explain the "nonlinear dynamics of aerosol populations," rather than attributing your results to these dynamics it might be more effective to say something like "The nonlinear response of the mean particle diameter to SO₂ initial concentration is similar to results found by Julin et al. (2018) in a modelling study."

line 336: Is this the mean particle diameter after 30 minutes? Be explicit. You defined initial growth rate as average for 0-30 minutes, but did not define a timeframe for mean diameter.

line 383-84: If you believe you have not sufficiently justified the necessity of chemical composition studies, I recommend you do so in the introduction instead.

line 589: Figure 9 may belong in the supplement, because it has only a brief mention in the text that is limited to the observation that positive mode MS spectra skew toward lower m/z's.

line 594, 600: The contents of Figure 10 are also not discussed in the text, so it could be moved to the supplement as well.

line 509-10: use comparable measures to compare the SO₂-involved and SO₂ free conditions. I would recommend including the multiplicative factor of the change for both.

line 524-26: The sentence "Initial growth rate... as shown in Fig. S6" is redundant, and to the next sentence "In SO₂-free experiments... initial growth rate of aerosol particles" I would add something like "compared to the SO₂-involved condition."

General comments

The authors have studied SOA formation from photo-oxidation of 1,2,4-trimethylbenzene under different concentrations of NOx, SO2 and NH3 in an indoor smog chamber. The reactants and the particle generation and growth were monitored using a series of standard instrumentations. The chemical functional groups were characterized by ATR-FTIR in each experiment and inorganic constituents were analyzed using ion chromatography. The molecular level information is provided using UPLC-HRMS and dd-MS2 scans. Ten new organosulfates are identified in the presence of SO2, including 3 in which the origin was previously unknown and some previously reported to be originated from biogenic precursors. Formation mechanisms for 8 of the newly identified organosulfates are proposed based on previous literature. Their results indicate that SO2 is a key parameter for ultrafine particle formation. A synergistic effect of NH3 and SO2 in particle formation is also shown, indicating the importance of reducing both SO2 and NH3 emissions to improve lowering PM. Their results also suggest that ammonium sulfate form by the reaction of NH3 with H2SO4 facilitate aerosol formation and growth through condensation of organic vapors.

This article advances the current knowledge of aerosol formation from photo-oxidation of a typical aromatic hydrocarbon in the presence of NOx, SO2 and NH3, therefore is of interest to the scientific community of ACP. The manuscript is well written and organized. The experiments are well executed, methods are explained adequately, results are discussed thoroughly, and conclusions are well supported. I have some minor comments to improve the quality of this manuscript that are listed below. I recommend accepting this manuscript for publication in ACP with minor revisions.

Specific comments

Line 13 ­­ – Indicate the major emission sources of TMB.

Line 45 – Define ‘certain regions.

Line 54-56 – Add reference/s.

Line 98-99 – Briefly indicate what MCM is and add a link to the version.

Line 210 – Indicate how the mass calibration was performed.

Line 200 – Indicate the mass resolution in full MS, top N in dd-MS2, isolation width (mass window) etc

Line 277 – Add initial growth rates to Table 1 or SI.

Line 277-280 – Can the authors elaborate the reasons for the observed non-linear response of the particle diameter with initial SO2 concentrations?

Figure 4 – Indicate what error bars represent.

Figure 5 – Indicate the experiment numbers relevant to a) to f)

Lines 346-348 – Add detailed information of the common products observed in SO2 free and SO2 involved experiments to the SI.

Table S3 – Indicate clearly whether this table shows the compounds that are detected in both SO2 free and SO2 involved experiments with NH3 or the products only formed from the experiments involved both NH3 and SO2.

Tables S2 and S3 – Add UPLC retention times.

Technical corrections

Line 105 – Add ‘in the atmosphere’ to the end of “Given the ubiquity of SO2, NH3, and TMB…

Line 200 – It is better to write it as data-dependent MS/MS (dd-MS2) scans

Line 206 – Add B after 3%.

Figure 6 – Label the red structures as OS-226, OS-228…etc. (Authors may replace the chemical formula with their abbreviated names as the structures are shown.)

Figure 10 – Match the color of the TMB on the figure with carbon number (should be light blue as it has 9 carbons)

Overview:

The manuscript by Yang et al. examined the effect of SO₂ and NH₃ on the formation of SOA from 1,2,4-trimethylbenzene photooxidation. After the injection of SO₂ and/or NH₃, the apparent yield of the SOA increased after wall loss correction, which demonstrated the synergistic effect of SO₂ and NH₃ in facilitating SOA formation. The authors also used ATR-FTIR, IC, and UPLC-MS to systematically analyze the particle phase composition and identified various inorganic and organo-sulfates compounds. My main comments are about the control experiment setup in this manuscript, which are discussed below. Other than the main comments, the manuscript is written clearly and comprehensive.

Major Comments:

The authors demonstrate that the addition of SO₂ and/or NH₃ during the photo-oxidation experiments can enhance the SOA formation and yield. However, part of the aerosol growth can arise from the formation of H₂SO₄ or (NH₄)₂SO₄particles even without the presence of any SOA, which should be deducted from the enhancement effect of SO₂ and/or NH₃.

I suggest the authors add in three control experiments (pure SO₂, pure NH₃, and SO₂+NH₃ mixtures but with similar levels of OH radicals and UV intensity using H₂O₂ or other OH generators) without any VOCs to rule out the formation of inorganic species contributing to the SOA. Figure S7 shows the mass spectra with and without SO₂ and/or NH₃ are similar, suggesting that maybe inorganic species formed with OH are the likely source of enhancement. It would be important to understand that after ruling out this part of the inorganic aerosol formation, how much enhancement the SO₂and/or NH₃ would add to the yield of the SOA.

Another comment I have is that the enhancement of the SOA from the addition of SO₂ and/or NH₃ can also be attributed to the increase of the surface area from the formation of inorganic species, which shifts the gas-particle equilibrium more to the particle side. Can the authors discuss more about the effect of this shift of equilibrium? It seems Figure 4 shows that the first three experiments after adding SO₂ seem to follow this rule.

Minor Comment:

L217: Can the author make an additional plot in the SI or show how the size dependent wall loss factor is generated?

Figure 1: It would be better to change the plots all in the same scale for easy comparison of different conditions.

L305: an extra space before “to”

L367: Please add the recent paper by Chen et al. 2020 also demonstrate that OS-228 could be from isoprene-derived SOA.

L389: The author should also consider adding Zhang et al. 2019 here, which discussed the effects of inorganic sulfates to organosulfates conversion in affecting aerosol growth, multiphase chemistry, and pH.

L525: As mentioned, OS-228 was observed by Chen et al. 2020 from the aging of isoprene-SOA. Even though the source might be different from the OS-228 here, the author should mention it in the conclusion instead of defining OS-228 as from unknown source.

References:

Chen, Y., et al. (2020). "Heterogeneous Hydroxyl Radical Oxidation of Isoprene-Epoxydiol-Derived Methyltetrol Sulfates: Plausible Formation Mechanisms of Previously Unexplained Organosulfates in Ambient Fine Aerosols." Environmental Science & Technology Letters 7(7): 460-468.           

Zhang, Y., et al. (2019). "Joint Impacts of Acidity and Viscosity on the Formation of Secondary Organic Aerosol from Isoprene Epoxydiols (IEPOX) in Phase Separated Particles." ACS Earth and Space Chemistry 3(12): 2646-2658.