Unraveling sulfur chemistry in interstellar carbon oxide ices

Formyl radical (HCO•) and hydroxycarbonyl radical (HOCO•) are versatile building blocks in the formation of biorelevant complex organic molecules (COMs) in interstellar medium. Understanding the chemical pathways for the formation of HCO• and HOCO• starting with primordial substances (e.g., CO and CO2) is of vital importance in building the complex network of prebiotic chemistry. Here, we report the efficient formation of HCO• and HOCO• in the photochemistry of hydroxidooxidosulfur radical (HOSO•)–a key intermediate in SO2 photochemistry–in interstellar analogous ices of CO and CO2 at 16 K through hydrogen atom transfer (HAT) reactions. Specifically, 266 nm laser photolysis of HOSO• embedded in solid CO ice yields the elusive hydrogen‑bonded complexes HCO•···SO2 and HOCO•···SO, and the latter undergoes subsequent HAT to furnish CO2···HOS• under the irradiation conditions. Similar photo-induced HAT of HOSO• in solid CO2 ice leads to the formation of HOCO•···SO2. The HAT reactions of HOSO• in astronomical CO and CO2 ices by forming reactive acyl radicals may contribute to understanding the interplay between the sulfur and carbon ice-grain chemistry in cold molecular clouds and also in the planetary atmospheric chemistry.

The authors report on their studies on the formation of HCO and HOCO radicals from UV photochemistry of CO and CO2 ices in the presence of the HOSO radical. These ices serve as analogues for modelling interstellar chemistry on small ice particles. HCO and HOCO radicals are important as nodes in radical reaction networks for forming higher order organic molecules in interstellar ice analogues, e.g., pyruvate and glyoxylate. Inorganic sulfur photochemistry is also important in the context of the atmospheric chemistry of planets and exoplanets, and so the HOSO radical could be important for forming prebiotic carbon feedstocks in the gas phase. The authors utilize HVFP of CHF2S(O)OH to generate the HOSO radical, which they then condense back into the solid-phase for further photochemical investigations at low temperature by laser irradiation at 266 and 365 nm. Products were identified by IR and UV/Vis spectroscopy, aided and abetted by quantum mechanical calculations. The authors conclude that the HCO and HOCO radicals are obtained from hydrogen atom transfer of the HOSO radical under 266 nm irradiation in CO and CO2 ices. The authors further suggest this chemistry could also be relevant to the SO2-rich atmosphere of Venus.
Overall, this is an interesting and well-written paper, and the work will be of interest to prebiotic chemists, both gas-phase and potentially aqueous-phase chemists. The conclusions are well founded for the most part. The reviewer recommends publication after the following comments/criticisms have been addressed.
1) The authors use HVFP of CHF2S(O)OH to generate HOSO radicals presumably as an experimental convenience for their photolysis studies. But, it would be insightful to know if the same HCO and HOCO radicals can be generated from photolysis of an astrochemically realistic icy mixture, perhaps one consisting of SO2, CO and CO2 ices. Could the authors please comment on the plausibility of this type of experiment, and perhaps even include some preliminary data?
2) While the experiments the authors report definitely seem relevant to the photochemistry of dense molecular clouds containing icy grains, the connection to atmospheric chemistry is more speculative. The authors dedicate a significant amount of discussion to potential gas-phase photochemistry, but the current manuscript contains no actual photochemical gas-phase data. The Reviewer thinks the manuscript would be much stronger if they can demonstrate, for example, gas-phase production of glyoxylic or pyruvic acid, by irradiation of, for example, sulfur oxide or carbon oxide gaseous mixtures. Can the authors please comment on whether or not they think such an experiment is feasible and if it can be included in the present manuscript?
3) Carbonyl sulfide is another interesting prebiotic molecule, as it has been demonstrated to be capable of serving as a prebiotic activating agent for amino acid polymerization. See the following reference for an example. https://www.science.org/doi/full/10.1126/science.1102722 The authors might include a couple references to this point and a brief mention in the main text. 4) On page 13, the authors say "The photo-induced hydrogen atom transfer of HOSO also occurs in solid CO2 at 16 K (Fig. 2C), yielding a new molecular complex (Table S1)." Do the authors mean Fig. 3C? Also, the authors should specifically mention how they prepared the CO2 ice experiments in the SI. I assume "CO2-doped Ar-matrix" has the same meaning as "solid CO2", correct? where the HOSO mixing ratio is expected to be high. Furthermore, from astrophysical relevance, broadband UV irradiation is more important than 266 nm irradiation. Also, I agree that the computation of the structures (and spectra) of the complexes is very important, but only those aspects should be discussed in the main manuscript which helps to understand the reaction mechanism. All the other aspects and discussion should be moved to the supporting information.

Reviewer #1
Response: We greatly appreciate the reviewer's insightful comments. As suggested by the reviewer, the main focus of the present work is the discovery of a new pathway for the efficient formation of two astrochemically important acyl radicals HCO• and HOCO• through the hydrogen atom transfer (HAT) reactions of HOSO• in interstellar carbon oxide ices, and the observed HAT reactions may build the link connecting the inorganic chemistry of sulfur oxides and the organic chemistry of carbon oxides in interstellar medium. As described in the Introduction, the two interstellar species HCO• and HOCO• are key building blocks for the formation of complex organic molecules (COMs) in interstellar medium. Previous laboratory studies have found that the two radicals can be generated through the hydrogenation of CO and CO2 with hydrogen-containing molecules such as H2, H2S, CH4, and H2O, and vacuum ultraviolet (VUV) radiations were usually used due to the fact that these We also agree with the Reviewer that broadband UV irradiations are important in astrophysical chemistry. Indeed, we have also applied other light sources (193 laser, 365 nm LED, and high-pressure Hg-lamp) for the study of the photochemistry of HOSO• in CO ice. Formation of HCO• and HOCO• through HAT was also observed, this is also consistent with the broad UV-vis absorption band of HOSO• below 400 nm.
However, the yields for the two acyl radicals and their complexes are much lower since the major absorption of HOSO• locates at around 270 nm ( Figure 2). For clarity, the spectrum for the formation of HCO• during the 365 nm irradiation of HOSO• in CO ice has been added in the revised Supporting Information ( Figure S5), whereas no IR bands for HOCO• were observed which is probably due to its photodissociation to •H and CO2. In the meantime, shorter-wavelength ArF excimer laser (193 nm) irradiation can deplete HCO• and HOCO• and their complexes. Therefore, the more selective 266 nm laser was applied to generate the intermediates (unstable complexes) in the HAT reactions of HOSO• in CO and CO2 ices.  the two bands is about 8 : 1, which means that there is an alternative pathway for the production of CO2. As suggested by the reviewer, the corresponding sentence has been changed to "Assuming a bimolecular reaction of HOCO• and SO for the formation of HOS• and CO2, the observation of more CO2 than HOS• (based on the experimental and calculated IR band intensities) in the photochemistry indicates that there is an alternative pathway for producing CO2.".

Response:
The typo has been changed.
Response: The typo has been corrected. In addition to the experiments in pure CO and CO2 solids, photochemistry of HOSO• in CO-and CO2-doped Ar-matrixes was also performed. To make it clear, a sentence about the experimental details has been added in the supporting information: " For the preparation of the CO2-or CO-doped Ar-matrix, a premix of Ar with CO2 or CO (a ratio of 1:20) was used as the matrix gas.".

Reviewer #1 (Remarks to the Author):
I am satisfied with the answers and the revision. The paper can be published in its present form.

Reviewer #2 (Remarks to the Author):
The reviewer would like to thank the authors for responding well to my first round of comments, but this reviewer agrees with the other reviewer that the introduction still lacks focus. Also, since the authors were unable to get the gas-phase data previously suggested (understandably so), the reviewer strongly feels the introduction and discussion sections need to have some major modifications.
The TOC graphic, abstract and title give the reader the impression that the context of the paper will be focused on the astrochemistry of interstellar space, i.e., ice grain photochemistry, and the experiments are certainly relevant to this context. The introduction, however, hardly even mentions this context. The first two paragraphs of the introduction are almost completely about atmospheric chemistry, which the experiments in the authors' manuscript do not directly relate to. The reviewer strongly suggests that the authors dedicate more space to giving context with respect to interstellar ices, and only briefly mention the potential relevance to planetary atmospheric chemistry.
The same criticism is applied to the discussion section -the relevance of the authors' results to interstellar ice-grain chemistry is virtually completely omitted, and instead the discussion is focused on Venusian atmospheric chemistry. The reviewer feels that the discussion section needs much more focus on interstellar ice-grain chemistry, and much less focus on pure speculation with respect to the organic chemistry of Venus's atmosphere. Figure 6 is especially problematic because the right-hand-side of the scheme is not supported by any data reported by the authors, especially the production of any carboxylic acid derivatives. Even if it were possible to synthesise relatively simple organic acids in this way, it is highly questionable that any other higher order organic molecule could be produced given the harsh conditions of the Venusian atmosphere, namely, the concentrated sulfuric acid which will rather quickly decompose organic molecules. Since the authors were not able to provide this sort of gas-phase data previously requested, understandably because these are very difficult experiments to perform, the reviewer strongly feels the authors need to tone down their speculations in this context, and focus more on the relevance to ice-grain astrochemistry.
In sum, the experiments of the authors are top-notch, but the introduction and discussion still need some work. After the changes suggested above are made, then publication in Nature Communications is recommended.