Unraveling sulfur chemistry in interstellar carbon oxide ices

The study addresses how key acyl radicals (HCO• and HOCO•), crucial intermediates for forming biorelevant COMs in space, can originate from abundant carbon oxides (CO and CO₂) under interstellar conditions. While radical networks involving H-containing species (e.g., CH₃OH, NH₃, CH₄) are well studied for COM formation, the photochemistry of dehydrogenated sulfur oxides (SO, SO₂) and derived radicals (HOS, HOSO) within CO- and CO₂-rich outer ice layers (10–20 K) remains poorly explored. Given SO₂ and SO are abundant in molecular clouds and that matrix photochemistry driven by cosmic UV/X-ray irradiation shapes ice compositions, the authors aim to unravel the photochemical pathways of HOSO• in CO and CO₂ ices at 16 K, test whether HAT from HOSO• to carbon oxides can produce HCO• and HOCO•, identify the resulting weakly bound radical–molecule complexes, and assess implications for interstellar and planetary sulfur–carbon chemistry.

• HCO• and HOCO• are important in atmospheric/combustion chemistry and act as versatile precursors to COMs (e.g., formic, glyoxylic, pyruvic acids) formed at low temperatures in CO/CO₂ ices doped with H-containing species via radical–radical associations.
• HCO has been observed in multiple interstellar sources; HOCO⁺ has been detected in star-forming regions; radical recombination during phase transitions of interstellar CO ice has been documented.
• In contrast, SO/SO₂ and derived radicals (HOS, HOSO) photochemistry within CO/CO₂ interstellar ices has been scarcely investigated, despite SO₂/SO relevance in both ISM and planetary atmospheres (e.g., Venus, Io).
• Modeling suggests SO₂ photochemistry in the presence of water yields HOSO and OH under near-UV/visible light; gas-phase SO₂ photolysis at shorter wavelengths can yield SO, which dimerizes to OSSO implicated in Venus’s near-UV absorption. This context motivates probing HOSO• reactivity with carbon oxides in ices.

Experimental:

• Radical generation: HOSO• produced by high-vacuum flash pyrolysis (HVFP, ~700 °C) of difluoromethylsulfinic acid (CHF₂S(O)OH). ¹⁸O-labeled precursor synthesized via hydrolysis of CHF₂S(O)Cl with H₂¹⁸O to enable isotopic mechanistic probes.
• Matrix isolation: Deposition of pyrolysis stream in solid CO or CO₂ matrices at 16 K (and Ar at 10 K) under high vacuum (~10⁻⁵ Pa) on a gold-plated copper block (IR) or CaF₂ window (UV–vis). Typical precursor:matrix ratios ~1:1000; CO₂/CO-doped Ar matrices (1:20) also prepared.
• Spectroscopy: FT-IR (Bruker 70v; 5000–450 cm⁻¹; 0.5 cm⁻¹ resolution; 200 scans; MCT detector) and UV–vis (Perkin Elmer Lambda 850+; 190–800 nm; 1 nm s⁻¹). Temperature control via cryostats (10 K Ar, 16 K CO/CO₂).
• Photolysis: 266 nm Nd:YAG laser (10 mW) and 365 nm UV lamp (24 W); additional 193 nm irradiation for specific tests (e.g., SO photochemistry). Time-dependent photolysis tracked; some experiments at 532 nm to probe HOCO conformer interconversion.
• Isotopic labeling: ¹⁸O-labeled CHF₂S(O)OH yields a mixture of isotopologues; IR shifts monitored to assign products and probe hydrogen/oxygen transfer routes.

Computational:

• Geometry optimization and vibrational analyses for monomers and complexes at B3LYP-GD3(BJ)/def2-TZVP, followed by higher-level optimizations at UCCSD(T)/aug-cc-pV(T+d)Z (Gaussian 16, MOLPRO 2019.1).
• Noncovalent interaction (NCI) analyses via QTAIM and NCI approaches on CCSD(T)-optimized structures to characterize hydrogen bonding/chalcogen interactions.

Controls/comparators:

• Ar-matrix photolysis of HOSO• (10 K) to establish intrinsic channels (e.g., H + SO₂; OH…OS caged complex; formation of HSO₂) without carbon oxides.
• Separate photochemistry of SO and OSSO in CO ice (365 and 193 nm) to attribute secondary CO₂/OCS formation pathways.
• ¹⁸O-labeling to differentiate CO vs CO₂ hydrogenation and trace oxygen sources in products.

• Isolation and complexation:
  • HOSO• forms hydrogen-bonded complexes with CO and CO₂ in matrices: CO…HOSO• and CO₂…HOSO• identified by characteristic IR red/blue shifts (e.g., ν(OH) shifts: −148.5 cm⁻¹ in CO; −66.6 cm⁻¹ in CO₂ vs Ar; Table 1), consistent with theory (B3LYP/CCSD(T)).
  • UV–vis absorption of HOSO• appears in 350–240 nm range in CO/CO₂ matrices; efficient depletion at 266 nm confirms assignment.
• Photochemistry in CO ice (16 K, 266 nm):
  • Products include CO₂, OCS, HCO, HOCO (t- and c- conformers), H₂CO, HOS, and SO₂.
  • Hydrogen-bonded radical–molecule complexes observed: HCO…SO₂ and HOCO…SO (HOCOSO). In HCO…SO₂, ν(C–H) shifts to 2493.8 cm⁻¹ (vs 2488/2483 cm⁻¹ in CO/Ar), and SO₂ asymmetric/symmetric stretches show small red-shifts (Δν_asym = −8.2 cm⁻¹; Δν_sym = −0.3 cm⁻¹).
  • Secondary HAT: HOCO…SO undergoes H transfer to yield CO₂…HOS (CO₂HOS) under irradiation. CO₂ bending mode splitting (665–650 cm⁻¹) indicates complex formation with HOS.
  • ¹⁸O-labeling: exclusive formation of HCO (no HC¹⁸O) with a 1:2:1 mixture of SO₂/¹⁶OS¹⁶O/S¹⁸O¹⁶O confirms direct HAT from HOSO• to CO. HOCO and H¹⁸OCO• form in 1:1 ratio via association of photogenerated OH/¹⁸OH with CO. No HOCO from direct hydrogenation of CO₂ observed (absence of HSOC¹⁸O/HOCO from CO₂).
  • CO₂ and OCS production traced to photofragmentation of initially generated SO (from HOSO photolysis) and subsequent CO trapping, corroborated by independent SO/OSSO photolysis experiments (365/193 nm) in CO ice.
  • HOCO shows UV sensitivity (300–400 nm); 365 nm irradiation promotes reverse HAT from HOCO to SO₂, reforming HOSO and CO₂.
• Photochemistry in CO₂-doped matrices (16 K, 266 nm):
  • Formation of HOCO…SO₂ complex (ν(C=O) at 1818.9 cm⁻¹; blue-shifted ν(COH) to 1278.7 cm⁻¹; SO₂ stretches red-shifted relative to free SO₂). Additional OCS observed; OSCO not detected.
• Theory (UCCSD(T)/aug-cc-pV(T+d)Z):
  • CO…HOSO and CO₂…HOSO feature strong OH…CO/CO₂ H-bonds with H out of OSO plane (tilts ~33.9° and 30.5% respectively); H-bond lengths 2.168 Å (CO) and 1.990 Å (CO₂); binding energies De ≈ 3.9 and 5.3 kcal mol⁻¹.
  • HOCO retains trans-planar geometry in HOCOSO and HOCOSO₂; H-bond lengths 1.893 and 1.909 Å; binding energies ~6.1 and 5.4 kcal mol⁻¹.
  • HCO…SO₂ forms a 5-membered ring stabilized by CH…OS H-bond (2.590 Å) and CO…SO chalcogen contact (2.978 Å); De ≈ 3.5 kcal mol⁻¹.
  • Energetics: CO…HOSO → HCO…SO₂ (HCOSO₂) and HOCO…SO (HOCOSO) are higher in energy than CO…HOSO by 37.5 and 27.1 kcal mol⁻¹, respectively; secondary HAT in HOCOSO to CO₂…HOS is highly exothermic (−51.5 kcal mol⁻¹). Overall oxidation of CO to CO₂ by HOSO is exothermic by −14.0 kcal mol⁻¹ (cf. OH + CO ≈ −20.0 kcal mol⁻¹). HAT in CO₂…HOSO to HOCO…SO₂ is endothermic by 38.5 kcal mol⁻¹, below the H–O BDE in HOSO (44.1 kcal mol⁻¹).
• Conditions: All primary observations made at 16 K; 266 nm photolysis depletes HOSO rapidly; subsequent 365 nm irradiation modulates product distributions via reverse HAT.

The work demonstrates that HOSO•, a key intermediate in SO₂ photochemistry, can initiate hydrogen atom transfer to CO and CO₂ in cryogenic ices, efficiently generating acyl radicals HCO• and HOCO• alongside stable radical–molecule complexes (HCO…SO₂, HOCO…SO, HOCO…SO₂, CO₂…HOS). These findings address the open question of how dehydrogenated sulfur oxide chemistry couples with carbon oxide chemistry within CO/CO₂-dominated outer ice layers of interstellar grains. Isotopic labeling confirms direct H transfer from HOSO• to CO as the source of HCO, while HOCO primarily arises from association of photogenerated OH with CO and engages in further HAT and complexation chemistry. Theoretical energetics rationalize the feasibility of observed pathways, particularly the exothermic oxidation of CO to CO₂ via HOSO-mediated steps and the strongly stabilizing hydrogen bonds within the complexes. The observed UV sensitivity of HOCO enables reversible HAT cycles, suggesting dynamic photochemical equilibria under interstellar irradiation fields. Collectively, these processes provide mechanistic links connecting sulfur and carbon ice-grain chemistry, offering routes to generate reactive acyl radicals that can subsequently undergo barrierless radical–radical couplings to form COMs. Beyond ISM implications, the chemistry bears relevance to planetary atmospheres (e.g., Venus), where SO₂, SO, CO, and CO₂ coexist and photolytic processes modulate sulfur and carbon speciation.

This study reveals that under interstellar-analog conditions (16 K), UV photolysis (266 nm) of HOSO• embedded in CO and CO₂ ices drives hydrogen atom transfer reactions that produce HCO• and HOCO• and their hydrogen-bonded complexes with sulfur oxides. Spectroscopic identification (IR, UV–vis), isotopic labeling, and high-level quantum chemistry collectively establish mechanistic pathways, including exothermic oxidation of CO to CO₂ mediated by HOSO and reversible HAT involving HOCO. The results elucidate a direct chemical linkage between sulfur oxide photochemistry and carbon oxide hydrogenation, enriching our understanding of acyl radical generation that can seed COM formation in cold molecular clouds. Potential future directions include: extending to mixed ice compositions (e.g., with H₂O, CH₄, CH₃OH), exploring other sulfur radical intermediates (HSO•) and their HAT efficacy, surveying temperature and irradiation dependences, targeting detection of predicted complexes in laboratory astrochemistry and observational spectra, and integrating these pathways into astrochemical and planetary atmospheric models (e.g., Venus).

• The experiments employ matrix isolation in CO/CO₂ (and Ar-doped) ices at 16 K under high vacuum; while astrochemically relevant, matrix effects (cage effects, site heterogeneity) can influence reaction dynamics, tunneling, and product stabilization, potentially differing from amorphous interstellar ice mantles.
• Some intermediates predicted or mechanistically plausible were not directly observed (e.g., OSCO), and certain absorptions (e.g., HCO above 400 nm) were too weak for detection, limiting comprehensive spectroscopic constraints.
• Mechanistic inferences rely on difference spectra and isotopic shifts; while consistent with theory and controls, some steps (e.g., secondary fragmentation of SO) are supported indirectly.
• Studies were conducted at a single low temperature (largely 16 K) and specific photolysis wavelengths/intensities; generalization across broader interstellar irradiation fields and temperatures warrants further work.