Liwen Yang et al. “SIM-HOM (version 1.0): A Mechanistic Module for the formation of highly oxygenated organic molecules from Isoprene, Monoterpene and Sesquiterpene evaluated with ADCHAM (version 1.0)”

Paul Wennberg

This is a very welcome effort to build a mechanistically informed module that can capture the formation of highly oxygenated molecules in the oxidation of biogenic alkenes initiated by OH and ozone. The authors have used MCM and other compiled mechanisms that provide detailed gas phase oxidation insights and added pathways that describe HOM formation from autoxidation and RO2+RO2 chemistry. They use a set of simulations from CLOUD to benchmark the compiled mechanism both with single and mixed alkene starting mixtures. The mechanism has significant skill and will substantially advance our ability to simulate the production of HOMs and the associated aerosol in many regions of the world

I recommend publication in close to the current version. I ask the authors, however, to consider providing a more complete description of what chemistry drives the HOM formation – I suspect that perhaps the ‘top ten’ pathways provide most of the HOM production in each system and it would be helpful for future investigations to have this quantified here.

For the RO2 + RO2 chemistry my sense, based on Murphy et al., 2025 (https://pubs.rsc.org/zh-tw/content/articlepdf/2025/ea/d5ea00106d), is that the dimer formation branching ratio is generally ~10%. Even for simple alkanes (e.g. ethane RO2), this seems the case. To the extent that this is true, the major question is just the rate coefficients for these RO2+RO2 reactions. Given that addition of a functional group (e.g. HOCH2CH2OO vs CH3CH2CH2OO) has such a rate enhancement, I am curious whether such enhancements are additive in your mechanism (e.g. does adding a carbonyl and an alcohol or acid yield 100x rate – see for example Ziola and Ziemann - J Phys Chem A. 2025 Feb 13; 129(6):1688-1703). I would also ask the authors to provide evidence for the statement (ln314) “Likewise, reactions between two non-autoxidizable RO2 are not included due to their low propensity to form accretion products”. Finally, in the monoterpene RO2 section, please add a reference to Kenseth, 2023 where you cite Peräkylä et al., 2023.  [https://www.science.org/doi/10.1126/science.adi0857 ]

To my knowledge, however, there is no evidence that acylperoxy radicals yield such ROOR and I would like to understand what this mechanism suggests as CH3C(O)OO is both one of the most ubiquitous (PAN decomposition) and reacts at 1/10 the collision rate. This is especially important during nighttime with NO3 addition to biogenic alkenes will produce RO2 in a low radical environment.

Finally, I would suggest that although NOx is not considered in the current model it would be helpful to include in the “future work” some commentary on whether NO3 chemistry with biogenics is likely to be important source of HOMs and which alkenes you would recommend be studied in more detail. For example, I think one of the first demonstrations the ROOR formation from NO3 addition is important for SOA was Sally Ng’s study in 2008 that illustrated that this pathway was likely responsible for all the SOA formed in an isoprene chamber study - https://acp.copernicus.org/articles/8/4117/2008/acp-8-4117-2008.html. I’d recommend citing this study in the introduction as well.

General comment

The authors intend to publish this manuscript as a model description paper. However, the model description (Section 2) is incomplete impeding scrutinity and reproducibility of the results. Although additions for autoxidation is well described or referenced, the manuscript lacks specific numbers and formulas for the RO2-RO2 reaction rate constants depending on the structure and size of the radicals. For general and specific numbers one is left to dig into the ADCHAM code uploaded on Zenodo. For instance, it is not obvious where to find the expressions of the relevant rate constants. A presentation, possibly with tables, of a systematic assignment of rate constants and dimers yields must be part of the model description. 

Major comments

1. Dimers from monoterpenes 

Only for monoterpene chemistry (?) I could identify rate constants kRO2_RO2_PRAM and kROOR_PRAM. The former a pseudo-unimolecular one and the latter a bimolecular one. Maybe they are assigned the values 1D-12 and 1D-13*RO2  as could be guessed by looking at the Input file input_PRAM01.txt . In the list of reaction rates in secondRates.f90 these are found in RCONST Array elements multiplied by some chosen integers. Maybe the rationale is described in a previous article describing PRAM but should also be described here. I could not identify rate constants for specific RO2 categories from isoprene and sesquiterpene chemistry. I want to stress that I am very familiar with the MCM and other chemical mechanisms. 

2. Dimers from isoprene and sesquiterpens

For those two compounds is completely obscure (to me) which rate constants and why have they been assigned for the dimer formation. The description of the modifications for autoxidation are described in the text. However, like for the monoterpenes the specifics like the relevant block of reaction list cannot be found anywhere for reference/scrutiny. KPP-generated files for the integration of the associated ODE systems cannot be used for such purpose. I culd not even find the equation (.eqn) and species (.spc) files.

3. RO2 Permutation reactions 

At lines 306-316 it is stated that only cross-reactions between autoxidizable and non-autoxidizable RO2 significantly contribute to the accretion products. For those reactions which rate constants are used and how do they Change with structure and size of the RO2?  What is the yield of ROOR ? Are autoxidizable RO2 ever considered as part of the "RO2 pool" defined by the MCM permutation reactions formalism? 

4. Validity of the RO2 Permutation formalism

This formalism has the fundamental assumption that \chem{CH_3O_2} is the dominant RO2 justifying the estimate of k as the geometric average of two self-reaction rate constants, one of \chem{CH_3O_2} and one of the RO2 at hand. The observations used for model Validation are from Experiments with apparently no nitrogen oxides and likely RO2-RO2 and RO2-HO2 reactions as the dominant RO2 sinks. \chem{CH_3O_2} is not to be expected the dominant RO2. Given that the self-reaction rate constant of \chem{CH_3O_2} is among the lowest ones, how is this inconsistency going to affect the optimal values for the key rate constants affecting dimer formation? Could the authors elaborate on that and perform a sensitivity simulation?  

Minor comments

- There are many instances with no blank space between end of sentence and reference. 

- line 198: where is ozone 10% of isoprene oxidation? Is it a global average? Provide a reference.