The near-UV absorber OSSO and its isomers

Zhuang Wua, Huabin Wana, Jian Xua, Bo Lua, Yan Lua, André K. Eckhardtb, Peter R. Schreiner*b, Changjian Xiec, Hua Guoc and Xiaoqing Zeng*a
aCollege of Chemistry, Chemical Engineering and Materials Science, Soochow University, Suzhou 215123, China. E-mail:
bInstitute of Organic Chemistry, Justus-Liebig University, Heinrich-Buff-Ring 17, Giessen 35392, Germany. E-mail:
cDepartment of Chemistry and Chemical Biology, University of New Mexico, Albuquerque, NM 87131, USA

Received 5th February 2018 , Accepted 27th February 2018

First published on 27th February 2018

Disulfur dioxide, OSSO, has been proposed as the enigmatic “near-UV absorber” in the yellowish atmosphere of Venus. However, the fundamentally important spectroscopic properties and photochemistry of OSSO are scarcely documented. By either condensing gaseous SO or 266 laser photolysis of an S2⋯O2 complex in Ar or N2 at 15 K, syn-OSSO, anti-OSSO, and cyclic OS([double bond, length as m-dash]O)S were identified by IR and UV/Vis spectroscopy for the first time. The observed absorptions (λmax) for OSSO at 517 and 390 nm coincide with the near-UV absorption (320–400 nm) found in the Venus clouds by photometric measurements with the Pioneer Venus orbiter. Subsequent UV light irradiation (365 nm) depletes syn-OSSO and anti-OSSO and yields a fourth isomer, syn-OSOS, with concomitant dissociation into SO2 and elemental sulfur.

Diatomic sulfur monoxide, SO, is a key intermediate in the oxidation of elemental sulfur and thus plays a significant role in the Earth's sulfur cycle.1 Sulfur monoxide can be generated through UV photolysis of SO2 (→ SO + O), a reaction believed to be responsible for the sulfur isotope mass-independent fractionation (S-MIF) in the Earth's atmosphere.2 The existence of SO in the atmosphere of both Io3 and Venus4 has already been confirmed, with SO being the second most abundant sulfur oxide in the Venusian atmosphere (7–30 ppb for SO and 30–500 ppb for SO2 at 64 km altitude near the equator).5 The rich SO chemistry involving the interconversion with other sulfur oxides such as S2O, SO2, and SO3 and the formation of sulfur/sulfate aerosols on the Earth and other interstellar media has been extensively studied in the recent decades.1,2,6

Chemically, SO is highly reactive at room temperature, not only via facile oxidation by molecular oxygen,7 but also through fast reactions with dienes8 and N-heterocyclic carbenes9 in the synthesis of episulfoxides and sulfines, respectively. As a ground electronic state triplet species, SO dimerizes in the gas phase.10 Hence, SO dimers are always present during the production of SO. Typically, 5% of the side-on dimer OSSO forms in the microwave discharge of SO2, together with 20–30% SO and 5% S2O.11 The presence of planar OSSO in the mixture has been confirmed by microwave,12 photoelectron,13 and millimeter-wave spectroscopy.14 The results from rotational spectroscopy suggest a syn-conformation with C2v symmetry for OSSO, whereas the C2h-symmetric anti-conformer could not be detected since it has no dipole moment.

Recently, both syn- and anti-OSSO have been proposed as the enigmatic “near-UV absorber” in the atmosphere of Venus,15 mainly based on the coincidence of the computed (LR-CC2/aug-cc-pV(T+d)Z) transitions at 313 (syn-OSSO) and 370 nm (anti-OSSO) with the recorded near-UV spectrum (320–400 nm) for the yellowish cloud in the atmosphere of Venus.16 However, as pointed out by a more recent photochemical model, this identification remains tentative since the agreement is not completely satisfactory.17 Therefore, an experimental determination of the UV/Vis spectrum of OSSO and the study of its photochemistry are of vital importance for unveiling the sulfur-chemistry in the SO-rich Venusian atmosphere.

Computationally, the molecular structures, energies, and vibrational data of more than ten S2O2 isomers on the singlet and triplet potential energy surfaces have been explored.15,18 According to the latest computations (Fig. 1),15 the end-on (I–IV) and side-on complexes (V) are true minima, all except anti-OSOS (III) should be viable candidates for experimental detection because of the large activation barriers for mutual isomerization, bond dissociation into two SO, and rearrangement to the global minimum S[double bond, length as m-dash]SO2 (VI). As shown in Fig. S5 (ESI), the two highest occupied molecular orbitals MO-24 and MO-23 of the syn and anti OSSO conformers correspond to the out-of-plane π and in-plane σ-bonding between the two OS fragments, whereas the two terminal sulfur p orbitals adopt an anti-bonding configuration. Hence, one would expect partial double bond characters in these two conformers. Experimentally, the S–S bond length in syn-OSSO [2.0245(6) Å determined by microwave spectroscopy;12 2.011(3) Å determined by millimeter wave spectroscopy14] is very close to the presumably typical S–S single bond in HSSH [2.0564(1)].14b

image file: c8cc00999f-f1.tif
Fig. 1 Energy profile of S2O2 isomers (I–VI) computed at the fc-CCSD(T)/cc-pV(T+d)Z and MRCI/cc-pV(T+d)Z (in parentheses) levels.15

Herein, we report the first IR and UV/Vis spectroscopic characterization of four S2O2 isomers, namely, syn-OSSO (I), anti-OSSO (II), syn-OSOS (IV), and cyclic OS([double bond, length as m-dash]O)S (V) and their multistep photo-induced isomerization in cryogenic matrices.

The generation of the monomers and dimers of SO was achieved by high-vacuum flash pyrolysis (HVFP) of ethylene episulfoxide7a in Ar or N2 (1[thin space (1/6-em)]:[thin space (1/6-em)]1000) at ca. 1000 K. The resulting mixture was then condensed at 15 K as a solid matrix for spectroscopic characterization. A typical IR spectrum of the pyrolysate is shown in Fig. 2A. Decomposition of the episulfoxide (a) is evident through the formation of C2H4 (b)19 and SO (c),20 along with new species with a strong IR band at 1105.5 cm−1 (f) and other weaker bands at 1252.3 and 795.6 cm−1 (g). Given the theoretically predicted (TD-B3LYP/6-311+G(3df)) symmetry-forbidden n → π* transition (1A21A1) at around 500 nm for the most likely candidate OSSO (syn-OSSO: 495 nm; anti-OSSO: 542 nm), the matrix was irradiated first with yellow light (570 ± 20 nm). The resulting IR difference spectrum (Fig. 2B) shows the selective depletion of the carrier of the very strong IR band at 1105.5 cm−1 (f). Another weaker IR band at 2219.7 cm−1 also vanishes, which very likely belongs to the overtone of the strong band. In addition, a weak side band at 1098.9 cm−1 for the naturally abundant 34S can be discerned, the corresponding 34S isotopic shift (6.6 cm−1) is very close to that of the S[double bond, length as m-dash]O stretching vibration in SO (c, 1138.5 cm−1, Δνexpt(32/34S) = 6.2 cm−1). Referring to the computed IR spectra of the various S2O2 isomers (Tables S2–S5, ESI), the band at 1105.5 cm−1 can be assigned to anti-OSSO, for which only one band at 1115.8 cm−1νcalc(32/34S) = 6.2 cm−1, B3LYP/6-311+G(3df), 1122.4 cm−1, CCSD(T)-F12b/VTZ-F12) with a non-zero IR intensity (300 km mol−1) above 200 cm−1 was computed. According to the computed vibrational displacement vector for C2h-symmetric anti-OSSO, this band corresponds to the antisymmetric combination of the two S[double bond, length as m-dash]O stretching vibrations.

image file: c8cc00999f-f2.tif
Fig. 2 (A) IR spectrum of matrix isolated HVFP (1000 K) products of ethylene episulfoxide (a) in Ar (1[thin space (1/6-em)]:[thin space (1/6-em)]1000) at 15 K. (B) The IR difference spectrum reflecting the change of the same matrix upon yellow light irradiation (570 ± 20 nm, 50 min).

Upon yellow light irradiation, dissociation of anti-OSSO (f) results in the formation of SO2 (d) and elemental sulfur (S2, vide infra). Concomitantly, two other species with IR bands at 1159.3/1112.5 cm−1 (e) and 1252.3/795.6 cm−1 (g) form. In fact, these two species are also present in the HVFP products of ethylene episulfoxide (Fig. 2A). Remaining anti-OSSO (f) and e in the matrix can be completely depleted with subsequent UV light irradiation (365 nm) within 5 min (Fig. 3A). As a result, d, g, and another new species h with IR bands at 1209.3 and 764.7 cm−1 appear.

image file: c8cc00999f-f3.tif
Fig. 3 (A) IR difference spectrum reflecting the depletion of e and f and the formation of d, g, and h upon UV light irradiation (365 nm, 5 min). (B) IR difference spectrum reflecting the depletion of h and the formation of d, e, and f upon green light irradiation (532 nm, 10 min). (C) IR difference spectrum reflecting the depletion of d and g and the formation of e and f upon UV light irradiation (266 nm, 8 min).

Consistent with the computed intense near-UV transition at 350 nm for syn-OSSO, the IR bands at 1159.3 and 1112.5 cm−1 (e) can reasonably be assigned. The observed band positions agree with the computed (fc-CCSD(T)/cc-pV(Q+d)Z)14 anharmonic fundamentals for syn-OSSO at 1171 and 1120 cm−1 (1192.4 and 1139.0 cm−1, B3LYP/6-311+G(3df) and 1173.4 and 1122.8 cm−1, CCSD(T)-F12b/VTZ-F12) for symmetric and antisymmetric combination of the two S[double bond, length as m-dash]O stretching vibrations, respectively. The simultaneous depletion of anti-OSSO (f) with 365 nm light also coincides with the computed intense transition at 400 nm.

In addition to the photodecomposition (e and f, Fig. 3A), the 365 nm photolysis of OSSO yields two species exhibiting IR bands at 1252.4/795.5 cm−1 (g) and 1209.3/764.7 cm−1 (h). Further successive irradiations with green light (532 nm, Fig. 3B) and a UV laser (266 nm, Fig. 3C) selectively deplete h and g and reform both e and f, respectively. Upon comparison with the computed IR spectra of the various S2O2 isomers (Tables S2–S5, ESI), it was found that the two pairs of IR bands could be associated with syn-OSOS (h, 1213.3/744.0 cm−1, B3LYP/6-311+G(3df) and 1236.9/739.6 cm−1, CCSD(T)-F12b/VTZ-F12) and cyclic OS([double bond, length as m-dash]O)S (g, 1271.9/804.5 cm−1, B3LYP/6-311+G(3df) and 1266.1/803.1 cm−1, CCSD(T)-F12b/VTZ-F12). The isotopic shifts for the naturally abundant 34S for the two bands at 1209.3 cm−1νexpt(32/34S) = 12.3 cm−1, Δνcalc(32/34S) = 12.5 cm−1) and 1252.4 cm−1νexpt(32/34S) = 12.9 cm−1, Δνcalc(32/34S) = 13.2 cm−1) can be clearly identified. Due to low IR intensities, a third band for each of these two species computed above 500 cm−1 (g: 559.4 cm−1; h: 520.1 m−1) was not observed in the available spectral range (4000–500 cm−1). Note that the decomposition of these S2O2 isomers to SO2 and sulfur always occurs under the applied irradiation conditions.

The formation of some of the S2O2 isomers was also observed when a mixture of S2 and O2 was photolyzed with a 266 nm laser. In contrast to the previously observed formation of SO2 and S2O from an initial S2⋯O2 complex upon broad-band irradiation (>235 nm),21 the 266 nm laser photolysis of the matrix-isolated S2⋯O2 complex, generated through HVFP of 5-methyl-1,3,4-oxthiazol-2-one in an O2/Ar mixture (1[thin space (1/6-em)]:[thin space (1/6-em)]50[thin space (1/6-em)]:[thin space (1/6-em)]1000), yields SO2 and various S2O2 isomers. This alternative synthetic route enabled us to prepare doubly-18O labeled S2O2 isomers using 18O2. As expected, all IR bands at around 1200 cm−1 for the S[double bond, length as m-dash]O stretching vibrations in the isomers exhibit isotopic shifts of about 40 cm−1.

The photointerconversion of the four S2O2 isomers in the solid N2-matrix was also followed by UV/Vis spectroscopy (Fig. 4). In line with the IR spectroscopic observation, the two broad absorptions centered (λmax) at 517 and 390 nm for anti-OSSO (f), which correspond to the n → π* and π → π* transitions, respectively, vanish completely upon yellow light irradiation (579 nm). Subsequent 365 nm irradiation depletes the broad band at 375 nm, which, by correlation to the changes in the IR spectrum (Fig. 3A), should belong to syn-OSSO (e). In the meantime, the weak absorption at 517 nm reappears, which is likely to be associated with syn-OSOS (h), since further green light irradiation (532 nm) results in complete disappearance of this band. As for the strong transition at 287 nm, the contribution from S2 is evident from the characteristic vibrational structure,22 additional contributors include cyclic OS([double bond, length as m-dash]O)S (g) with the computed lowest transition at 294 nm and other decomposition species (SO2, C2H4, and SO) in the matrix. A weak band with superimposed vibrational structures at 458 nm remains almost unchanged during irradiation; its assignment is currently unclear.

image file: c8cc00999f-f4.tif
Fig. 4 UV/Vis spectra of the HVFP products of ethylene episulfoxide in N2 (1[thin space (1/6-em)]:[thin space (1/6-em)]1000) at 12 K.

Apparently, the experimentally obtained UV/Vis spectrum of OSSO (anti: 517 and 390 nm; syn: 375 nm) match the characteristic near-UV absorption (320–400 nm) found for the Venus's yellowish cloud by the Pioneer Venus orbiter.23 Consistent with the previously observed strong correlation of the unknown “near-UV absorber” with the most abundant sulfur oxide SO2 in the Venusian atmosphere,24 OSSO molecules isolated in cryogenic matrices not only undergo isomerization to OSOS and cyclic OS([double bond, length as m-dash]O)S but also decompose to SO2 and elemental sulfur upon irradiation, in which the putative isomer S[double bond, length as m-dash]SO2 (VI, Fig. 1) might also be involved. Subsequently, highly reactive sulfur atoms aggregate and furnish sulfur particles for the formation of cloud aerosols, the existence of which in the atmosphere of Venus has been suggested based on the data from the Venus Monitoring Camera on board Venus Express.25 SO2 can be further photolyzed to reform SO and an oxygen atom as part of the sulfur cycle.2

In summary, three isomers, syn-OSSO, anti-OSSO, and cyclic OS([double bond, length as m-dash]O)S, were identified as the dimerization products of diatomic sulfur monoxide in the gas phase. Upon irradiation of matrix-isolated OSSO, a fourth dimer OSOS formed in the more favorable syn-conformation. The identification of these four isomers with IR and UV/Vis spectroscopy was supported by quantum chemical computations. The potential role of OSSO in the sulfur-cycle of the Venusian atmosphere was unveiled according to the observed near-UV absorptions and the observed photochemistry. The obtained IR and UV/Vis spectral data serve as fingerprints for the identification of OSSO and isomers in the atmosphere of Earth and other planets.

This research was supported by the National Natural Science Foundation of China (21422304 and 21673147 to X. Q. Z.), and the project of scientific and technologic infrastructure of Suzhou (SZS201708). Partial support from the U.S. Department of Energy (DE-SC0015997 to H. G.) is also acknowledged. A. K. E. was supported through a fellowship of the Fonds der Chemischen Industrie.

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. M. Kumar and J. S. Francisco, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 864–869 CrossRef CAS PubMed , and references therein.
  2. For examples, see: (a) S. Ono, Annu. Rev. Earth Planet. Sci., 2017, 45, 301–329 CrossRef CAS; (b) A. R. Whitehill, C. Xie, X. Hu, D. Xie, H. Guo and S. Ono, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 17697–17702 CrossRef CAS PubMed; (c) S. Hattori, J. A. Schmidt, M. S. Johnson, S. O. Danielache, A. Yamada, Y. Ueno and N. Yoshida, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 17656–17661 CrossRef CAS PubMed; (d) S. Ono, A. R. White and J. R. Lyons, J. Geophys. Res., 2013, 118, 2444–2454 Search PubMed.
  3. C. T. Russell and M. G. Kivelson, Science, 2000, 287, 1998–1999 CrossRef CAS PubMed.
  4. X. Zhang, M.-C. Liang, F. Montmessin, J.-L. Bertaus, C. Parkinson and Y. L. Yung, Nat. Geosci., 2010, 3, 834–837 CrossRef CAS.
  5. (a) C. Y. Na, L. W. Esposito and T. E. Skinner, Geophys. Res. Lett., 1990, 95, 7485–7491 CrossRef; (b) E. S. Barker, Geophys. Res. Lett., 1979, 6, 117–120 CrossRef CAS.
  6. For example, see: (a) F. P. Mills and M. Allen, Planet. Space Sci., 2007, 55, 1729–1740 CrossRef CAS; (b) X. Zhang, M. C. Liang, F. P. Mills, D. A. Belyaev and Y. L. Yung, Icarus, 2012, 217, 714–739 CrossRef CAS; (c) R. Hu, S. Seager and W. Bains, Astrophys. J., 2013, 769, 6–19 CrossRef.
  7. (a) Z. Wu, B. Lu, R. J. Feng, J. Xu, Y. Lu, H. B. Wan, A. K. Eckhardt, P. R. Schreiner, C. J. Xie, H. Guo and X. Q. Zeng, Chem. Commun., 2018, 54, 1690–1693 RSC; (b) A. R. Whitehill, B. Jiang, H. Guo and S. Ono, Atmos. Chem. Phys., 2015, 15, 1843–1864 CrossRef.
  8. For examples, see: (a) R. S. Grainger, B. Patel, B. M. Kariuki, L. Male and N. Spencer, J. Am. Chem. Soc., 2011, 133, 5843–5852 CrossRef CAS PubMed; (b) J. Nakayama, Y. Tajima, X.-H. Piao and Y. Sugihara, J. Am. Chem. Soc., 2007, 129, 7250–7251 CrossRef CAS PubMed.
  9. (a) L. E. Longobardi, V. Wolter and D. W. Stephan, Angew. Chem., Int. Ed., 2015, 54, 809–812 CrossRef CAS PubMed; (b) D. W. Stephan, J. Am. Chem. Soc., 2015, 137, 10018–10032 CrossRef CAS PubMed; (c) D. W. Stephan, Science, 2016, 354, aaf7229-1 CrossRef PubMed.
  10. E. A. Evans, A. B. Scott and J. L. Huston, J. Am. Chem. Soc., 1952, 74, 5525–5527 CrossRef CAS.
  11. (a) R. Steudel, Top. Curr. Chem., 2003, 231, 203–230 CrossRef CAS; (b) P. W. Schenk and R. Steudel, Angew. Chem., Int. Ed., 1965, 4, 402–409 CrossRef.
  12. (a) F. J. Lovas, E. Tiemann and D. R. Johnson, J. Chem. Phys., 1974, 60, 5005–5010 CrossRef CAS; (b) R. Steudel, Angew. Chem., Int. Ed., 1975, 14, 655–664 CrossRef.
  13. B.-M. Cheng and W.-C. Hung, J. Chem. Phys., 1999, 110, 188–191 CrossRef CAS.
  14. (a) M. A. Martin-Drumel, J. Van Wijingaarden, O. Zingsheim, F. Lewen, M. E. Harding, S. Schlemmer and S. Thorwirth, J. Mol. Spectrosc., 2015, 307, 33–39 CrossRef CAS; (b) J. Behrend, P. Mittler, G. Winnewisser and K. M. T. Yamada, J. Mol. Spectrosc., 1991, 150, 99–119 CrossRef CAS.
  15. B. N. Frandsen, P. O. Wennberg and H. G. Kjaergaard, Geophys. Res. Lett., 2016, 43, 11146–11155 CrossRef.
  16. For example, see: J. B. Pollack, O. B. Toon, R. C. Whitten, R. Boese, B. Ragent, M. Tomasko, L. Esposito, L. Travis and D. Wiedman, J. Geophys. Res., 1980, 85, 8141–8150 CrossRef.
  17. V. A. Krasnopolsky, Icarus, 2018, 299, 294–299 CrossRef CAS.
  18. For examples, see: (a) A. Ramírez-Solís, F. Jolibois and L. Maron, Chem. Phys. Lett., 2011, 510, 21–26 CrossRef; (b) M. Goodarzi, M. Vahedpour and F. Nazari, Chem. Phys. Lett., 2010, 494, 315–322 CrossRef CAS; (c) T. G. Clements, H.-J. Deyerl and R. E. Continetti, J. Phys. Chem. A, 2002, 106, 279–284 CrossRef CAS; (d) C. D. Paulse, R. A. Poirier and R. W. Davis, Chem. Phys. Lett., 1990, 172, 43–48 CrossRef CAS; (e) K. Takano, H. Hosoya and S. Iwata, J. Am. Chem. Soc., 1982, 104, 3998–4005 CrossRef CAS; (f) C. J. Marsden and B. J. Smith, Chem. Phys. Lett., 1990, 141, 335–353 CAS; (g) Y. Murakami, S. Onishi, T. Kobayashi, N. Fujii, N. Isshiki, K. Tsuchiya, A. Tezaki and H. Matsui, J. Phys. Chem. A, 2003, 107, 10996–11000 CrossRef.
  19. E. Pytter and D. M. Gruen, Spectrochim. Acta, 1979, 35A, 199–207 Search PubMed.
  20. F. Salama and H. Frei, J. Phys. Chem., 1989, 93, 1285–1292 CrossRef CAS.
  21. G. D. Brabson, A. Citra, L. Andrews, R. D. Davy and M. Neurock, J. Am. Chem. Soc., 1996, 118, 5469–5473 CrossRef CAS.
  22. (a) X. Q. Zeng, H. Beckers, H. Willner and J. S. Francisco, Angew. Chem., Int. Ed., 2012, 51, 3334–3339 CrossRef CAS PubMed; (b) L. Brewer, G. D. Brabson and B. Meyer, J. Chem. Phys., 1965, 42, 1385–1389 CrossRef CAS.
  23. J. B. Pollack, B. Ragent, R. Boese, M. Tomasko, J. Blamont, R. G. Knollenberg, L. W. Esposito, A. I. Stewart and L. Travis, Science, 1979, 205, 76–79 CAS.
  24. L. W. Esposito, Science, 1984, 223, 1072–1074 CAS.
  25. For examples, see: (a) E. Marcq, J.-L. Bertaux, F. Montmessin and D. Belyaev, Nat. Geosci., 2013, 6, 25–28 CrossRef CAS; (b) L. W. Esposito, Nat. Geosci., 2013, 6, 20–21 CrossRef CAS.


Electronic supplementary information (ESI) available: Experimental and computational details, IR spectra, computed data, and other electronic format. See DOI: 10.1039/c8cc00999f

This journal is © The Royal Society of Chemistry 2018