The manuscript describes results from laboratory based experiments to elucidate the gas phase atmospheric degradation chemistry of two VOC, namely 3,3-dimethylbutanal (henceforth DMBal in this review) and 3,3-dimethylbutanone (DMBone). There is much novel and valuable data presented here, but some improvements in presentation and discussion are required.

The introduction sets out the likely atmospheric sources and sinks of the two VOC, and outlines previous work. The experimental section is a little lacking in detail. For example, it may be that the photochemistry used to generate OH from CH3ONO is familiar to the authors, but it was not to this reviewer (nor many readers). If you are not going to elaborate here, at least include a reference to a literature paper describing these details. Further, I am not familiar with lamps of output 350 nm. Please describe.

There is a wealth of data presented in the results section, useful both for rate coefficient determinations and for product yields. Data presented in plots (e.g. Fig. 1) appear to be of high quality. My main concern regarding this manuscript concerns the data presented in Table 1, and the associated discussion. The data for DBBal + Cl looks good, both as presented in Fig. 1 and in Table 1. However, for determinations of k for the other two reactions, there appears to be an unusually wide spread of results. For DMBone + Cl, these values (in 10^-11 cm³ molecule^-1 s^-1) range from 3.03 to 5.22. The results obtained using 2-methyl-2-butanol as a reference VOC all cluster at the lower end of this range of results, whereas results using other reference compounds tend to agree on k ~ 5x10^-11 cm³ molecule^-1 s^-1. These observations are surely worthy of comment in the following text, together with some analysis of the provenance of the reference k-data. Is the data on 2-methyl-2-butanol + Cl (or any other of the reference reactions) well established? All worth a few more sentences, though to conclude on this reaction, if no prior studies were available then results from this work represent a considerable contribution to our understanding of atmospheric chemistry.

More concerning is kinetic data in Table 1 on DMBone + OH (Table 1). Results from this work range from k / 10^-12 cm³ molecule^-1 s^-1 = (0.96 +/- 0.11) to (1.92 +/- 0.59), exactly a factor of two, considerably larger than I would expect from a relative rate study. There is no clear sense here that the reference reactions are responsible for this inconsistency. Nor (if the data quality in Fig. 1 is in any way exemplary), was this likely a result of random noise from experiment to experiment. So, having ruled out two potential problems, we should consider other sources of error. Was there another experimental factor exerting a malign influence on the data? I note that we cannot tell from Table 1 whether experiments were conducted at 254 nm (where photolytic effects may be serious) or 350 nm, where the more complex precursor chemistry may introduce other secondary effect. Details such as precursors used should be included in Table 1. The uncertainties quoted for each experiment indicate that data quality could be quite varied; were all datasets nonetheless proportional in appearance – all having intercepts of zero and no evident curvature? Perhaps all such plots should be included in the S.I. Were there any difficulties in differentiating FTIR peaks of DMBone from precursors or products? An overall weighted average yielded (1.25 +/- 0.05)x10^-12 cm³ molecule^-1 s^-1. The surprisingly small (4%) uncertainty in k does not appear to reflect the inconsistencies encountered from one experimental determination to the next. There is more analysis needed and more discussion around the above points required to justify the closing statement from line 179 “These data are in good agreement with the values obtained in this study, thereby contributing to the accurate determination of the rate coefficients.”

There follows good discussion of k results, both in terms of the reactivity of different oxidants, of different functional groups (aldehyde and ketone) and of the impacts of structural changes within the VOC on reactivity. However, I found the absence of any comparison with one of the most recent and sophisticated SAR formulations surprising. The authors should compare results obtained here with those calculated using Jenkin et al. (2018) doi.org/10.5194/acp-18-9297-2018

The product studies for both reactions appear commendably detailed. Methods appear sound. My only concern here was the unidentified problem with the DMBone kinetic data (see above). This may have derived from FTIR retrievals. Might this propagate into errors in product yields?

There follows a discussion of atmospheric implications. The points raised here all seem reasonable, as do the various estimates of lifetimes and POCP. One point to note would be that photolysis lifetimes were estimated elsewhere based upon measured spectra, but that no quantum yield data was available. There are consequently large uncertainties in the rate of photolysis for either of these VOC.

Typos / minor concerns:

• should be an italic “k” throughout the manuscript;
• Table 1 header is confusing. It states that k is in units of cm³ molecule^-1 s^-1, but in fact the k values throughout the table are in different units of 10^-10, 10^-11 or 10^-12 cm³ molecule^-1 s^-1, depending on which reaction is being reported. I suggest that the most sensible way to report this is to list all k values in one consistent set of units, e.g. 10^-11 cm³ molecule^-1 s^-1. At the very least, remove the misleading statement “k in units of cm3 molecule-1 s -1” from the table header.
• Similar comments re. k values and powers of ten for Table 2.
• Line 166 “an” to “and”

General Comments:

The manuscript investigates the oxidation processes of two VOCs, specifically 3,3-dimethylbutanal and 3,3-dimethylbutanone, by Cl (with and without NO), NO3, and OH radicals. The authors provide extensive and detailed information on the chemical kinetics, reaction pathways, and resulting products. This study makes a significant contribution to our understanding of the oxidation mechanisms of 3,3-dimethylbutanal and 3,3-dimethylbutanone, particularly those involving Clorine atoms. However, I have some concerns regarding the treatment of wall loss. Once this issue is addressed and improve the quality of certain figures, the manuscript will be ready for publication in ACP.

Regarding the wall loss description (Line 149), while the results indicate minimal losses for 3,3-dimethylbutanal (3%) and 3,3-dimethylbutanone (0%), it is essential to account for potential wall loss impacts for oxidants like chlorine, N2O5, and NO3 on the rate constant results. Heterogeneous reactions, such as wall loss, may introduce uncertainties in chamber studies. Previous chamber studies have reported wall loss rates for these Cl, N₂O₅ , and NO₃ (e.g., https://doi.org/10.5194/acp-2020-360 and https://doi.org/10.1016/j.cplett.2009.03.047). The authors should provide more detailed data on wall loss rates to reinforce the reliability of their findings.

Some figures need quality improvement. For instance, Figure 2 requires a legend, and Figure 6 shows low resolution with small, hard-to-read text; it may be better suited for the supplementary document. Additionally, Figure 8 is difficult to read and should be rearranged to enhance readability.

Specific Comments:

1. Line 259: Does "x3" in Channel II refer to three attack sites? There is no clear explanation of what "x3" means in this context. The same issue arises in Line 264 for Channel III.
2. Footnote Description table1: The footnote description is unclear. It is not specified whether "a" uses 10^-10, "b" uses 10^-11, and "c" uses 10^-12, or what KR represents. Please revise this for clarity.
3. Figure 8 : In Channel III, (49% X=Cl ); 2% X= OH....NO3), there should be  no ")" inside.

The review is attached.

General comments and main concerns:

The article “Reactivity study of 3,3-dimethylbutanal and 3,3-dimethylbutanone: Kinetic, reaction products, mechanisms and atmospheric implications” by Aranda et al., is written in line with several previous studies performed by this research group.

The manuscript consists of two parts related to kinetic and mechanistic investigations on the reaction of two carbonyls (one ketone and one aldehyde) with atmospheric oxidants (Cl, OH, and NO3). The manuscript comprises a first part including three kinetic studies of 33DMbutanal+Cl, and 33DMbutanone with CL and OH radicals and a second part dedicated to product studies performed with chlorine and hydroxyl radical for both carbonyl compounds and gas-phase product studies initiated by NO3 radicals for the reaction with the aldehyde.

The manuscript needs consistent English corrections and also needs to be restructured on the sections related to gas-phase product presentation. The structure of the manuscript is tangled and difficult to follow, especially for the gas-phase product section. I suggest a structure with subsections for results and discussion representative for each studied reaction.

One main concern is related to product studies where, despite the extensive degradation mechanisms shown in the article, the products tentatively quantified are formaldehyde, acetone, and 22DMpropanal. No other products could be really supported by the presented results.

The mechanisms should be redesigned in a much simpler form to emphasize the findings and to highlight the m/z for the compounds identified in the present study. The expected products in the dotted line in the mechanism should be supported by the identified m/z in the mass spectra. Please add in the article, supportive information from mass spectra corresponding to chromatogram peaks.

Other concern is related to PAN formation and organic nitrate formation which are mainly suggested to be formed by generally accepted mechanisms due to the concentrations of the reactants involved in the present study. Please discuss the formation of this compound in terms of interferences due to the secondary sources of PAN which could be formed, for example, PAN formation from acetone degradation and huge NOx concentration. This source would explain also the delay in the formation of PAN.

The techniques employed in the present study can identify multiple other products present in the designed degradation mechanism. Could the authors reconsider the product formation interpreting the mass spectrum of each peak in their study? IR spectra did not show an OH absorption band around 3600 cm-1 according to the proposed alcohol products? There are also reference spectra that could help to identify possible formation products (eg. Biacethyl, etc.). A statement at line 592 indicates the formation of formic acid “which was not quantified”. Why do not provide all the possible information in the article? I suggest discussing the residual spectrum in terms of remaining IR features that are not allocated to the known products.

Please include in the paper all the plots for all the reference compounds used in kinetic experiments for 33DMbutanal with Cl. The scientific community wants to see how well the linearity of the kinetic plots and which of the reference compounds worked better. Why the other kinetic plots are not properly represented?

An important concern is related to the photolysis of 3,3-Dimethylbutanal. Tadic et al., found significant photolysis for 3,3-Dimethylbutanal in the spectral range used in the present study. How do the authors comment on the missing correction for the photolysis in their kinetic and product studies? How significant is the photolysis? There is a competition between photolytic lifetime and reactive lifetime, how do you comment on that?

Specific comments:

Could authors add information about the concentration of the radical precursors?

Table 1 – 33DMbutanal+Cl – the error for the second exp with cyclohexane is wrong.

• Please give an explanation to “d” in the table
• Please comment in the text for the difference of 25% between the kcarbonyl values for the reaction of 33DMbutanal+Cl
• Please comment in the text for the difference of 50% between the kcarbonyl values for the reaction of 33DMbutanone+OH. 0.96 and 1.92 are completely unrealistic for this study. Could you explain the lowest kcarbonyl values for the reaction of 33DMbutanone+OH using 1-butanol and cyclohexane as reference compounds?
• The following rate constants were recommended by McGillen database:
  • 1-butanol + OH: 9.14e-12 cm³molec^-1s^-1
  • 2-methyl-2-butanol + OH: 3.42e-12 cm³molec^-1s^-1
  • 2-propanol + OH: 5.24e-12 cm³molec^-1s^-1

Why the study does not use these values as the authors mentioned in the article but uses other values as the kOH = 9.8±2.0 for 1-butanol for example?

The reaction of acetylperoxy with HO2 could be better represented by a more recent paper: Winiberg, et al., 2016, Direct measurements of OH and other product yields from the HO₂  + CH₃C(O)O₂ reaction, Atmos. Chem. Phys., 16, 4023–4042, https://doi.org/10.5194/acp-16-4023-2016, 2016. This could explain the formation of ozone, peroxides, and OH radicals. Please discuss the effect of OH radicals formed in the reaction with chlorine atoms on the formation products. The reactivity towards OH and chlorine could help with these discussions.

The yields from nitrated gas-phase products are not very well represented. Could authors explain the curved shape?

Please give more information about the error calculation for kinetic results and product yields.

Minor comments:

Please consider “;” between the cited reference citation in the article text body.

Please revise the way used to include the cited literature in the article text body.

In the introduction section, the information about the formation of 33DMbutanone as a product from d 3,3-dimethyl-2-butanol degradation is presented a couple of times.

Line 104. Please use only one literature-cited reference for the experimental details found in the previous publication. Multiple citations, in this case five, lead to an increase in self-citations and this is unwanted.

Please avoid commas for Figure 1 and all the kinetic figures that are included in the article body and supplement.

Line 185. Please use the fundamental atmospheric chemistry reference literature (Finlayson-Pitts book for example) to support the attempt related to faster reactions with chlorine radicals.

Line 260 Please mention which SAR approach the authors considered.

The authors Aranda et al. have conducted a series of experiments examining the oxidation of a pair of saturated, functionalized carbonyl molecules. The experiments cover the three major relevant atmospheric oxidants (OH, Cl, and NO3 radicals) and include both relative-rate measurements and studies on major product formation to inform chemical mechanisms. The overall significance of different reactions and products is discussed in the context of general atmospheric chemistry. The authors also highlight some current unknowns regarding the chemistry of aldehydes, a significant and reactive class of volatile organic compounds. The authors have put a substantial amount of effort into these tasks and the interpretation of their data. I have several questions related to the oxidation mechanisms discussed in the paper. If these and minor comments are addressed, I believe the paper should be accepted for publication.

I suggest the authors take care when discussing the Cl-initiated oxidation of 3,3-dimethyl butanal. Previous works studying the oxidation of aldehydes (Iwasaki et al., 2008; Rayez et al., 2011; Singh et al., 2009) observed a strong deactivating effect of the CHO group that extended over multiple carbons and did not fit well with SAR predictions; Singh et al. (2009) wrote "the substituent factor for -CHO is significantly less than one, and that the group has a deactivating effect over several carbon atoms along the alkyl chain.” It is not clear to me whether the work of Carter (2021) used by the authors for estimating functional group factors accurately reflects these prior works. I therefore suggest that the authors evaluate the prior references mentioned here (or others, as relevant) and determine whether the 0.4 substituent factor, and the discussion in sections 3.1-3.2 and the proposed chemistry in Figure 8, are reasonable. It may not be feasible with the current experimental data, but experimental verification of SAR predictions for Cl reaction with aldehydes would be beneficial in understanding Cl/aldehyde chemistry.

The authors discuss major RO2 radical reaction channels in section 3.2. Can the authors make any estimate of the relative prevalence or importance of the different RO2 radical reaction partners (RO2, HO2, OH/Cl) during these experiments? Without some sort of estimate of RO2 reaction branching ratios, it is difficult to determine the significance of the different identified products and their associated proposed formation pathways. To put this another way, how realistic are the reaction conditions and calculated product yields?

The authors mention the clear loss of some primary products during later experiment times (Figure 4). Can the authors provide an estimate of when secondary chemistry and product formation may become relevant during each experiment? This could include oxidation or photolysis reactions of primary products. It is not currently clear whether primary products may be more or less reactive or photolabile than the parent molecules, which complicates the interpretation of the yield results and the variety of products identified through FTIR and GC-MS analysis.

Minor Comments

General: I found some discussion of parent and product structures difficult to follow. I would suggest showing structures of parent molecules in Table 2 somewhere in the text as well as adding a label or number to some products or intermediates in Figures 7 and 8 to make referencing these structures within the text more clear.

Line 201: might the decrease in reaction rate when moving to 3,3-dimethyl butanal be due to the fact that many of the abstractable H are now primary, with a lower inherent reaction rate towards OH and Cl, rather than steric factors?

Line 311, Figure 7, and elsewhere: I’m not familiar with the “alkoxy nitrate” compounds the authors note as forming from reaction of NO or NO2 with alkoxy radicals (e.g., (Atkinson, 2007)). Can the authors provide more background on the formation of these molecules?

Lines 489-491: Atkinson (2007) also wrote that “this ‘'prompt’’ decomposition of alkoxy radicals formed from the exothermic RO2 + NO reaction appears to be important for alkoxy radicals with a barrier to decomposition of approximately 9 kcal mol^-1 or less, with prompt decomposition being unimportant for alkoxy radicals with higher barriers to decomposition.” If the authors assert that alkoxy radicals formed in this work undergo “prompt” decomposition, they must present work supporting this conclusion (for example, utilizing the methods proposed in the already cited work of Vereecken and Peeters, 2009).

Line 493: RO2 + Cl à RO is also a potential reaction (Maricq et al., 1994) that may be relevant to consider.

Figure 8: For channel III, might intramolecular hydrogen shifts occur from wither the initial RO2 or RO radicals? Based on prior estimates (Vereecken & Nozière, 2020; Vereecken & Peeters, 2010), these aldehydic H-shifts are expected to be relatively fast. Might such reactions contribute to the larger variety of product structures observed for Cl reaction compared to OH, given the greater importance of channel III for Cl reaction?

Technical comments

Line 265:  “mayor” à major

Line 561: “…spectrum of 22DMpropanoic [acid].”

References

Atkinson, R. (2007). Rate constants for the atmospheric reactions of alkoxy radicals: An updated estimation method. Atmospheric Environment, 41(38), 8468–8485. https://doi.org/10.1016/j.atmosenv.2007.07.002

Iwasaki, E., Nakayama, T., Matsumi, Y., Takahashi, K., Wallington, T. J., Hurley, M. D., & Kaiser, E. W. (2008). Kinetics and Mechanism of the Reaction of Chlorine Atoms with n -Pentanal. The Journal of Physical Chemistry A, 112(8), 1741–1746. https://doi.org/10.1021/jp077525z

Maricq, M. M., Szente, J. J., Kaiser, E. W., & Shi, J. (1994). Reaction of Chlorine Atoms with Methylperoxy and Ethylperoxy Radicals. The Journal of Physical Chemistry, 98(8), 2083–2089. https://doi.org/10.1021/j100059a017

Rayez, M. T., Rayez, J. C., & Villenave, E. (2011). Theoretical approach of the mechanism of the reactions of chlorine atoms with aliphatic aldehydes. Computational and Theoretical Chemistry, 965(2–3), 321–327. https://doi.org/10.1016/j.comptc.2010.11.025

Singh, S., Hernandez, S., Ibarra, Y., & Hasson, A. S. (2009). Kinetics and mechanism of the reactions of n -butanal and n -pentanal with chlorine atoms. International Journal of Chemical Kinetics, 41(2), 133–141. https://doi.org/10.1002/kin.20383

Vereecken, L., & Nozière, B. (2020). H migration in peroxy radicals under atmospheric conditions. Atmospheric Chemistry and Physics, 20(12), 7429–7458. https://doi.org/10.5194/acp-20-7429-2020

Vereecken, L., & Peeters, J. (2010). A structure-activity relationship for the rate coefficient of H-migration in substituted alkoxy radicals. Physical Chemistry Chemical Physics, 12(39), 12608–12620. https://doi.org/10.1039/c0cp00387e