This is the review of the manuscript entitled “Suppressed atmospheric chemical aging of cooking organic aerosol particles in wintertime conditions” by Liu et al. This study investigates the chemical lifetime of cooking organic aerosol (COA) of three typical sources including canola oil, hot pot soup, and lard upon exposure to ozone. The second order reaction rate constants of ozone reacting with COA particles were determined as a function of temperature between -20 to 35 C. The reaction kinetics were derived by monitoring the degradation of the condensed phase organic by means of mass spectrometry and gas chromatography. It is observed that the reaction rate decreases by orders of magnitude as the temperature decreases. The temperature dependence of the reaction rate was fitted using a Vogel–Fulcher–Tammann (VFT) equation. This in turn was applied to predict the chemical lifetimes of COA across the globe for the months June and December. During northern hemispheric winter the chemical lifetimes of COA increased significantly.

This work is in the tradition of previous laboratory heterogeneous chemical reaction studies and thus fits the scope and audience of Atmospheric Chemistry and Physics. The experimental approach appears to be sound and the results present novel data. I have a few suggested minor revisions the authors should address before publication of this manuscript. Those points revolve around providing a few more details on the experimental approach and on data interpretation.

Minor comments:

Line 16: In heterogeneous reaction kinetics experiments typically the pseudo-first order decay is monitored. From this the second order rate constant could be derived. Do you imply a pseudo-second-order model that considers chemisorption as the rate-limiting mechanism of the process? Please elaborate.

Line 17: I am not sure if the statement of diffusion limitation is correct in this instance. If a reaction is diffusion-limited then the observed degradation does not reflect the actual reaction kinetics. However, changes in the reaction kinetics with temperature are observed. The authors likely meant to express that the second order rate constant is controlled by diffusion? As outlined further below, I would argue that a fit using a similar equation as the VFT description of viscosity is not a sufficient proof that only diffusion governed the observed temperature dependency of the reactivity.

Line 66: “However,…”. I do not understand this statement. If experiments are done correctly, reactive uptake measurements using aerosol particles or films result in the same reaction kinetics (Ammann et al., 2013). There are advantages and disadvantages for both approaches, e.g., gas-phase diffusion limitations. If the aim is to indicate that OA can remain in a metastable liquid phase, i.e., being supercooled (see, e.g., (Hearn and Smith, 2005)), which is less likely to occur using a film, then this has to be more clearly stated.

Furthermore, I think it would be fair to acknowledge the study by (De Gouw and Lovejoy, 1998).

Line 109: I would not call those concentrations “normal” and “high”. Both are unrealistically high. Please rephrase. You might want to express those concentrations also as a typically background and urban polluted ozone exposure time.

Line 113 and following: When determining the heterogeneous oxidation kinetics using aerosol particles, one has to pay attention to how reactivity scales with particle size (surface/volume). See e.g., studies by (Lim et al., 2017; Slade and Knopf, 2014). Have those experiments been conducted? How does the size distribution change prior to and after ozone exposure?

Somewhere in the introduction, to elevate the discussion, recent modeling studies that account for viscosity changes in multiphase chemical kinetics, could be briefly mentioned. E.g., (Berkemeier et al., 2021).

Section 3.3 and Table 1: I struggle to understand Table 1 and suggest elaborating this discussion more. When just quickly looking at the table, its meaning is not very clear. In the second column k2 is derived for only the oleic acid component in the types of particles given in column 1? Whereas column three reflects the reaction kinetics using a wider range of the mass spectrum. Maybe change the table or its description to make this clearer. Except for the value of the previous study, the data is derived from the same experiments? Does the difference in particle size distribution among the different aerosol source types matter when comparing their kinetics (Fig. S1)? See also comment above.

It may be worthwhile to mention that you are likely not gas-phase diffusion limited in the case of ozone uptake? I assume the uptake is sufficiently slow. Citing previous studies using canola oil or oleic acid might be helpful in this regard.

Line 197-198: Looking at Fig. S5 it seems the ratio was greater one for lowest temperature measurements. Could it be that surface-dominated oxidation resulted in more products that did not volatilize due to lower temperatures?

Line 244: Which transition (phase?) do you mean here?

Line 256-260: Condensed-phase diffusion is related to viscosity. However, I doubt, just because you can fit observations reasonably well with a VFT equation, though fit parameters are arbitrary and have no physical meaning, you can infer that only diffusion controls the entire oxidation process. This comes back to my comment in the abstract. There could be several processes going on in series or parallel which you are not resolving. See, e.g., (Pöschl et al., 2007; Berkemeier et al., 2021; Li and Knopf, 2021; Willis and Wilson, 2022). Clearly, your results demonstrate the importance of bulk diffusion but as long we cannot resolve all the intermediate steps, I suggest stating this observation more conservatively.

Technical corrections:

Line 121: Omit “also” since you already used “In addition,…”.

Figure 3: Typo in legend “Palmitic”.

Line 154: “species” should be “spices”?

Line 205: Missing “respectively”?

References

Ammann, M., Cox, R. A., Crowley, J. N., Jenkin, M. E., Mellouki, A., Rossi, M. J., Troe, J., and Wallington, T. J.: Evaluated kinetic and photochemical data for atmospheric chemistry: Volume VI - heterogeneous reactions with liquid substrates, Atmos. Chem. Phys., 13, 8045-8228, 10.5194/acp-13-8045-2013, 2013.

Berkemeier, T., Mishra, A., Mattei, C., Huisman, A. J., Krieger, U. K., and Poschl, U.: Ozonolysis of Oleic Acid Aerosol Revisited: Multiphase Chemical Kinetics and Reaction Mechanisms, ACS Earth Space Chem., 5, 3313-3323, 10.1021/acsearthspacechem.1c00232, 2021.

de Gouw, J. A. and Lovejoy, E. R.: Reactive uptake of ozone by liquid organic compounds, Geophys. Res. Lett., 25, 931-934, 1998.

Hearn, J. D. and Smith, G. D.: Measuring rates of reaction in supercooled organic particles with implications for atmospheric aerosol, Phys. Chem. Chem. Phys., 7, 2549-2551, 10.1039/b506424d, 2005.

Li, J. and Knopf, D. A.: Representation of Multiphase OH Oxidation of Amorphous Organic Aerosol for Tropospheric Conditions, Environ. Sci. Technol., 55, 7266-7275, 10.1021/acs.est.0c07668, 2021.

Lim, C. Y., Browne, E. C., Sugrue, R. A., and Kroll, J. H.: Rapid heterogeneous oxidation of organic coatings on submicron aerosols, Geophys. Res. Lett., 44, 949–2957, 10.1002/2017GL072585, 2017.

Pöschl, U., Rudich, Y., and Ammann, M.: Kinetic model framework for aerosol and cloud surface chemistry and gas-particle interactions - Part 1: General equations, parameters, and terminology, Atmos. Chem. Phys., 7, 5989-6023, 2007.

Slade, J. H. and Knopf, D. A.: Multiphase OH oxidation kinetics of organic aerosol: The role of particle phase state and relative humidity, Geophys. Res. Lett., 41, 5297-5306, 10.1002/2014gl060582, 2014.

Willis, M. D. and Wilson, K. R.: Coupled Interfacial and Bulk Kinetics Govern the Timescales of Multiphase Ozonolysis Reactions, J. Phys. Chem. A, 126, 4991–5010, 10.1021/acs.jpca.2c03059, 2022.

In their manuscript, Liu et al. investigated the atmospheric lifetime of cooking organic aerosol across the temperature range of -20 to 35C. They performed laboratory experiments with three types of cooking aerosol precursors: canola oil, hot pot base, and lard. Precursors were heated and emissions were passed through an aerosol flow tube containing ozone. Through SV-TAG and ACSM measurements, they measured changes in aerosol composition upon ozone exposure at different temperatures and then estimated pseudo-second order rate constants as a function of reactor temperature. They suggested that changes to aerosol viscosity at lower wintertime temperatures may increase the chemical lifetime of COA to upwards of 1 hour, while in the summer these aerosols may persist for only a few minutes. I enjoyed reading this paper and I think it fits well within the scope of Atmospheric Chemistry and Physics. I have a few minor comments and requests for clarification.

Line 18: Do you mean “increased” viscosity?

Line 30: How much did COA contribute in these European cities that are cited? Would be interesting to add some quantitative information here, as you did for the Los Angeles example.

Line 33: Do you mean that the mass fractions of COA are highly variable in the Chinese cities you mentioned? If so, please clarify. Also, does the range 8-33% come from the papers cited in the line above?

Line 52: “Much longer than the time scale” needs to be quantified; which time scale? I assume you are referring to “in the order of a few minutes” from the prior sentence? Does this account for both daytime and nighttime chemistry?

Line 97: Can you describe any possible particle losses in the stainless steel storage container? Three hours seems like plenty of time for losses to occur. Is there mixing in the container during this time? What was the size distribution of particles prior to entering the storage container for coagulation, and why was growing the particles necessary prior to ozonolysis experiments? Just to ensure you were comparable to COA particles in Beijing (mentioned on the next line)? And finally, do the size distributions in figure S1 account for all experiments? If so, the setup looks to be quite reproducible, which is great to see, but I suggest clarifying this. Can you comment on whether the difference in size distribution between COA types could be contributing to differences in reactivity between the COA types?

Line 108: 450 ppb is still a very high ozone exposure. I suggest re-phrasing how you label 450 ppb as “normal” ozone levels.

Line 134: Can you comment on the source of the halogens?

Line 135: Where is tygon tubing used in your setup? Can you comment on any particle losses to the tygon tubing? Do you have any blank/background samples to characterize the effect of chemical components (like the plasticizers you mention) coming off the tubing, or chemical compounds of interest sticking to the tubing?

Is section 3.1 all room temperature in the flow tube?

Figure 4: It is nice to see the comparison here between the COA types and their signals pre- and post-ozone introduction. I am not surprised to see the spectra looking so similar, since we are getting significant compound fragmentation in the ACSM. I think having a version of this figure with SV-TAG data, similar to Figure 2, would potentially yield more useful chemical information for readers. For instance, you could bring Figure S3 over from the SI and maybe include an inset that zooms in to some peaks of interest that react away significantly with ozone.

Line 163-164: Can you comment more specifically on what the differences are with the hot pot COA relative to the other types?

Line 234: What are your criteria for these transitions? What does “more pronounced” mean, quantitatively?

Line 245: Can you take the fatty acid composition you measured in Figure 3 and estimate the glass transition temperature of each of the mixtures? For instance, as in this paper: https://doi.org/10.5194/acp-18-6331-2018

Figure S5: The positive slope for the -11C experiment seems like an anomaly. You mentioned that these data were not used to calculate k2. Can you comment on if these experiments were repeated at all? Did the positive slope for this temperature occur every time? I suggest including information about replicate experiments in the Methods section or maybe even with Table S1.