the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 02 Jul 2025
Formation of highly oxygenated organic molecules from α-pinene photooxidation: evidence for the importance of highly oxygenated alkoxy radicals
Abstract. Highly oxygenated organic compounds (HOMs) from α-pinene oxidation are of great interest because of their importance in secondary organic aerosol (SOA) formation. Despite intensive investigations, the mechanisms of HOM formation from first-generation peroxy radicals to HOM-peroxy radicals (HOM-RO2·) and to HOM-closed shell products are not well understood. One reason is that HOM-alkoxy radicals (HOM-RO·) are likely to contribute to the propagation of oxidative radical chains (alkoxy-peroxy pathway) because isomerization of functionalized alkoxy radicals can compete with their fragmentation (and reaction with O2), as shown by theoretical kinetics. However, HOM-RO· reaction steps are difficult to verify in mechanisms. In this work, we have investigated HOM formation by varying the significance of the alkoxy-peroxy pathway as a function of NOX, OH·, and CO. HOM-RO· are likely formed with high branching ratios in reactions of HOM-RO2· with peroxy radicals (0.6) and NO (0.64) in analogy to simpler alkoxy radicals. We provide experimental evidence that for HOM-RO· the branching into isomerization is about 50 % (±14 %). Thus, HOM-RO· can play a central role in HOM formation, since alkoxy-peroxy pathways can compete with direct autoxidation. We observed significant concentrations of HOM-RO2·, despite fast termination by NO, and shifts to higher O/C for HOM-RO2· and termination products with increasing NO. At NO concentrations >1.5 ppb, the alkoxy-peroxy pathway may even prevail in propagating the oxidative radical chain leading to HOM formation. The increasing sink of HOM-RO2· with increasing concentration of peroxy radicals and NO is compensated by an increasing source via the alkoxy-peroxy pathway.
- Preprint
(1333 KB) - Metadata XML
-
Supplement
(1588 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-2772', Anonymous Referee #1, 23 Jul 2025
Kang et al. uses high resolution NO3- ToF mass spectrometry to investigate the peroxy alkoxy pathway to highly oxygenated organic molecules (HOMs). Since the alkoxy radical cannot be detected using CIMS, they infer alkoxy reactions occurring by the parity of oxygen and hydrogen number, where odd oxygen C10H15Ox peroxy radicals are assumed to have formed via an alkoxy radical intermediate. There are some limitations to this method (e.g. oxygen parity only works through one generation of peroxy-alkoxy isomerization and NO3- CIMS is only sensitive to HOM RO2) and many assumptions about complex RO2 and RO chemistry. Nevertheless, the story of this paper, summarized in Figure 8, is sound and may fundamentally shift how we think about HOM formation. I find this study interesting and believe the audience of ACP will too. Before being published I request the following comments to be addressed.
Specific comments:
1. My main concern is that the discussion often neglects the importance of R structure for the peroxy-alkoxy pathway. This includes RO formation and RO isomerization, both of which are broadly parameterized here. While there are instances where structure is addressed (e.g. Figure 1), below are a few places where I would like to see a deeper discussion,
- Figure 1 is a great illustration of how a step of alkoxy chemistry changes the RO2 hydrogen parity, in this case from an even to odd number. However, mechanism ABE produces a less oxygenated RO2. This contradicts the paper conclusion that the peroxy alkoxy pathway leads to more oxygenated organic compounds. A discussion on R here would help clarify branching between mechanisms ABCD and ABE.
- It is difficult to interpret changes in the RO2 and RO branching chemistry when the ratio of OH:O3 oxidation changes. Peroxy and alkoxy branching is sensitive to R, and so it matters whether the peroxy radical you make is coming from OH addition, OH abstraction, or ozonolysis. Please address how the ozonolysis rate is accounted for in the AP turnover when you adjust O3, NOx, CO, or light aperture.
- Line 84: ROO-OOR -> 2RO2 + O2 branching ratio seems to be largely parameterized and probably varies much more. Please clarify if using structure activity relationships.
- Line 427: I don’t agree that RO isomerization will be a statistical probability as you make it out here. It will be extremely dependent on R structure. RO isomerization will create new functionalized R backbones that may favor or inhibit future isomerization steps. Please clarify if using structure activity relationships and also their applicability.
- Line 798: Please clarify what you mean by the alkoxy-peroxy step does not rely on specific chemistry of alpha-pinene. I agree these results are applicable (and important!) for large, functionalized VOCs, but would this statement hold true for alkanes, especially those with 5 or fewer carbon atoms?
2. Although it is pointed out that NO3- CIMS is selective towards measuring HOMs, there is little discussion as to what it cannot efficiently detect. Including more information on this is important for readers to interpret what's shown in the plots and also what's not shown (e.g. less oxygenated RO2 that precede HOM RO2).
3. For the NOx experiments, your aim is to increase [NOx] to produce RO from RO2+NO reactions. There is a good discussion in the main text and SI about how NOx affects OH and thus the alpha pinene turnover rate. Have you calculated/modeled how additional NOx affects OH:HO2? The previous CO experiments show that increased HO2 shuts down peroxy alkoxy pathways, which could also explain your data in the NOx experiments.
4. In Figure 5, the x-axis is plotted as the NOx concentration. Table S1 indicates almost all of NOx is NO2. Can this be changed to NO concentration since it is RO2+NO reactions that produce alkoxy radicals?
5. Figure 6 is interesting but may be better suited for the SI. The text points out that signal is changing due to [OH] which is not corrected for. The comparison to MCM does not fit in with this specific study since autoxidation is not implemented and therefore it is without a single path to HOM RO2 (with or without considering the peroxy alkoxy pathway).
6. The branching ratios you derive for both RO formation and subsequent isomerization are important and really jumped out to me in the abstract. However, the discussion in the text comes late and was not clear. I would recommend putting more of a focus on this in the main text.
Minor comments:
Line 70: Should alcohol product have a radical dot?
Line 90: Add NO2 as product
Line 97: Clarify RO2+NO produces RO or RONO2 as products
Line 201: You mention using SA calibration factor for HOMs and specify a value. But no concentrations are reported in the paper. Why is that?
Line 258: You specify the same formula, C10H17Ox+1, twice. Please clarify.
Line 264-267: This paragraph is generally confusing to read. Please clarify. You point out 3 products coming from 2 precursors and then say respectively. It is not clear what products are from which precursors.
Line 268: What is sufficient NO? Even low NO will be important to peroxy-alkoxy chemistry (Nie, W., et al, Nature Comm, 2023)
Line 417: “Different from C10H15O8 and C10H15O10, formation of C10H15O6 is obviously exclusively initialized by OH oxidation.” I would remove obviously. Could C10H15O6 not been formed from AP + O3 ->C10H15O4 -> C10H15O6 through one generation of autoxidation?
Line 495: “The CO experiment resulted in a clear suppression in the abundance of HOM-RO2 radicals” This is not clear from the data presented in Figure 4 which is normalized. Please show unnormalized data to make this point.
Line 655: “…strong increase in O:C caused by NOx addition is only explainable by impacts of HOM-RO…” I agree that your data supports this conclusion, but I would caution against saying this is the only explanation since there are not measurements for how the less oxygenated RO2 were affected
Additionally, please reread the manuscript to minor typos (extra word in line 491, missing word in line 638, subscripts in line 726)
Citation: https://doi.org/10.5194/egusphere-2025-2772-RC1 -
RC2: 'Comment on egusphere-2025-2772', Anonymous Referee #2, 29 Jul 2025
With chamber oxidation experiments, Kang et al. show that bimolecular reactions of alpha-pinene + OH derived peroxy radicals with RO2 and NO enhance the formation HOM by alkoxy-peroxy steps instead of inhibiting it as previously thought. RO2 + NO reactions appear to be particularly important in this regard, leading to products with up to 15 oxygen atoms, three oxygens higher than experiments without NO. As they highlight in their conclusion, this would imply that the formation of HOM is more favorable under polluted conditions than clean, opposite to what is widely accepted in the community. This work is important, timely and highly relevant to the work of others in the field. I recommend its publication. I have the following comments that I request be addressed.
1. One concern I have is the scarce measurement of the C10H17Ox family of peroxy radicals. These first-generation peroxy radicals from OH addition to alpha-pinene are completely unmeasured, except for C10H17O10, while first-generation peroxy radicals from H-abstraction, a minor channel in comparison, are measured.
- It is clear from Xu et al. 2019 and Berndt 2021 that OH addition to alpha-pinene initiates autoxidation and the formation of peroxy radicals at least up to C10H17O7. Further reactions of some of these are likely the source of the C10H17O10 measured here. Insensitivity of the NO3-CIMS method toward these products is unlikely to be the reason as Figure S5 in this paper shows that C10H15O6-9 are measured above the detection limit.
- The authors attribute their low detection of C10H17Ox compounds partly to their measurement conditions. While some secondary oxidation of pinonaldehyde could explain the C10H15Ox compounds, the authors find primary products from OH H-abstraction to also be at least equally important (Figure S6). This necessitates a more detailed discussion regarding why primary OH-addition products, which should dominate a-pinene + OH reactions, are not measured or measured minimally. Perhaps the additional -OH group in the latter leads more efficiently to termination during autoxidation, and these are then candidates for secondary oxidation by OH?
- Figure S5 shows multiple C10H18Ox measured. As the authors state in line 257 of the manuscript, these products can only form from bimolecular reactions of C10H17Ox and C10H17Ox+1 (correct the typo in the manuscript, you have two instances of C10H17Ox+1). So, the C10H17Ox peroxy radical precursors of C10H18Ox clearly form during these experiments.
2. In addition to the reaction classes described by the authors, some of the products could form from RO2 + OH reaction producing trioxides (ROOOH) (Assaf et al.). This reaction could be particularly important under the elevated OH conditions of the experiments carried out here. The importance of this reaction should perhaps be modelled out.
- Related, could the increase in C10H15O2n+1 signals at elevated levels of OH (line 420) be attributed in some part to the RO2 + OH => ROOOH => RO + HO2 (and not just HOM-RO2 + RO2 as currently stated)?
- This mechanism could also explain in part the CO experiments. In the absence of CO, higher OH concentrations can lead to C10H15O2n+1 products via the ROOOH pathway above, which switches in the presence of CO when OH concentrations are lower. The trioxide does get more stable against decomposition with the increase in carbon chain length, so the contribution of the channel is perhaps minimal.
3. About the C10H16O7 signal in page 10 which dominates the C10H16Ox family in their measurements, the authors state that the contribution to this signal from C10H17Ox is low (line 351). Does the majority of the C10H16O7 signal measured then come from reactions of C10H15Ox? Maybe provide additional details regarding the reactions that are likely involved, whether R2 or R3b or something else. If it’s R2, does [HO2] explain the measured intensity?
4. Lines 415-419: C10H15O6 can form from a-pinene ozonolysis. In fact, Meder et al. 2025 cited here measure multiple isomers of this peroxy radical. Also, consider a different word than “unimportant” in line 419. C10H15O6 is crucial to formation of the next peroxy radical in the autoxidation chain, C10H15O8. It can also react bimolecularly to form the closed-shell C10H14O5 species, as reported in Meder et al.
5. Regarding the effect of CO on the formation of HOM RO, the authors cite Jenkin et al. 2019 to say that the branching to alkoxy radicals from RO2 + HO2 reactions should be low (line 464). However, this is highly dependent on the structure of the R. According to Jenkin et al., there is an almost 50:50 branching towards ROOH and RO for beta-oxo peroxy radicals (Table 8 in Jenkin et al 2019). The authors should discuss the importance of the RO2 + HO2 reaction in the context of the structures of the Rs in their system.
6. How does the decrease in [OH] from CO addition affect the RO2 intensities and distribution? In line 494 the authors put the onus of HOM-RO2 suppression completely on HOM-RO2 + HO2 reactions, but how much of the suppression is due to lower [OH]?
7. The increase in C10H15Ox signals with the increase in NOx: is there a possibility that some ozone is forming from NO2 photolysis? If I understand the method section correctly, UV-A lights are on during these experiments, so won’t NO2 photolysis increase O3 concentrations, explaining at least partly, the observed increase in C10H15Ox signals?
Assaf, E., Schoemaecker, C., Vereecken, L. and Fittschen, C., 2018. Experimental and theoretical investigation of the reaction of RO2 radicals with OH radicals: Dependence of the HO2 yield on the size of the alkyl group. International journal of chemical kinetics, 50(9), pp.670-680
Citation: https://doi.org/10.5194/egusphere-2025-2772-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 288 | 53 | 16 | 357 | 23 | 13 | 21 |
- HTML: 288
- PDF: 53
- XML: 16
- Total: 357
- Supplement: 23
- BibTeX: 13
- EndNote: 21
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1