Bis[(methylsulfanyl)carbonyl]disulfane

The title compound, C4H6O2S4, was prepared by repeating, with subtle improvements, a multi-step route originally described by Mott & Barany [J. Chem. Soc. Perkin Trans. 1 (1984) ▶, pp. 2615–2621]. The title compound was obtained for the first time as a crystalline material. The two [(methylsulfanyl)carbonyl]sulfenyl moieties are essentially perpendicular to each other, each approximately planar (r.m.s. deviations of 0.02 and 0.01 Å) and with a C—S—S—C torsion angle = 90.99 (6)°, which compares well with the theoretical value of 90°.

The title compound, C 4 H 6 O 2 S 4 , was prepared by repeating, with subtle improvements, a multi-step route originally described by Mott & Barany [J. Chem. Soc. Perkin Trans. 1 (1984), pp. 2615-2621. The title compound was obtained for the first time as a crystalline material. The two [(methylsulfanyl)carbonyl]sulfenyl moieties are essentially perpendicular to each other, each approximately planar (r.m.s. deviations of 0.02 and 0.01 Å ) and with a C-S-S-C torsion angle = 90.99 (6) , which compares well with the theoretical value of 90 .
Supplementary data and figures for this paper are available from the IUCr electronic archives (Reference: GW2119). While bis(alkoxysulfanylcarbonyl)polysulfanes, [RO(C═S)] 2 S n (n = 1, 2, 3, 4) have been known for a long time due to their derivation from readily formed xanthate salts (Reid, 1962), much less information is available about the isomeric bis[alkyl(sulfanylcarbonyl)]polysulfanes, [RS(C═O)] 2 S n (Mott & Barany, 1984). In 1984, we reported methodology for the preparation of reasonably pure (>95%) exemplars in the latter family for R = Me; the trisulfane and tetrasulfane of the series were obtained as crystals and their structures were solved by X-ray diffraction (Mott & Barany, 1984). Beyond the successful synthetic routes reported therein, a number of alternative methods were tested, which seemed rather straightforward and were well precedented for analogous compounds, but failed for the series under investigation. For the present studies, we carried out more careful experimental work aimed at the disulfane (Fig. 3), and obtained it in crystalline form for the first time. The structure of the disulfane was solved by X-ray crystallography, and compared to the structures determined earlier.
All bond distances and angles are within expected ranges. Because the disulfane is adjacent to carbonyl groups, the S-S bond length of 2.03 Å is slightly shorter than the 2.07 Å reported for the S-S bond length in S 8 . This phenomenon is quite general and has been observed for related compounds. In all, a search of the Cambridge Database for compounds of the formula R(C═O)SS(C═O)R provided four different compounds for comparison [CSD refcodes: BOWGAV (Bereman et al., 1983), DBZOSS01&03 (Rout et al., 1983;Paul & Srikrishnan, 2004), UDALER (Li et al., 2006), and 880326 (Schroll et al., 2012)]. The most noteworthy feature of our newly reported structure is that the torsion angle about the disulfane was 90.99°, which comes closer to the theoretical optimum of 90° (Pauling, 1960) than any other of the comparison compounds with the exception of bis(N,N-dicyclohexylsulfanylcarbamoyl)disulfane, where this angle is 89.7° (Li et al., 2006).

Note regarding nomenclature:
The title compound is named in a manner that is consistent with our prior publications and modern conventions. The related compound studied by Li et al. (2006) was named bis(N,N-dicyclohexylsulfanylcarbamoyl)disulfide by those authors. (1). First, potassium t-butyl xanthate was prepared by adding carbon disulfide (37.0 ml, 0.61 mol) over 15 min to a well stirred suspension of potassium t-butoxide (69.7 g, 0.62 mol) in p-xylene (450 ml) at 80°C. The reaction mixture was allowed to cool to 25°C, and the orange-yellow precipitate which had formed was collected on a Buchner funnel, and washed with ethyl ether (1.7 l total, gravity filtration) over a 1 h period. After the final ether wash, the solid was air-dried by suction applied to the filtration apparatus from a water aspirator, and then dried further in a desiccator (20 mm) for 72 h (time required to achieve constant mass). Yield of xanthate salt: 80.3 g (70%). A portion of the xanthate salt (23.9 g, 127 mmol) was suspended in ether (200 ml), and neat iodomethane (12.0 ml, 193 mmol) was added, under magnetic stirring, at 25°C over a 2 min period. After 24 h, the reaction mixture was filtered and  (2). Neat sulfuryl chloride (1.3 ml, 16.2 mmol) was added, with stirring at 4°C, over 1 min, to a solution of compound 1 (5.0 g, 30.4 mmol) in CHCl 3 (20 ml). After completion of addition, stirring continued for a further 5 min, following which the homogeneous reaction mixture was concentrated in vacuo to provide the crude title product as a yellow oil [3.4 g, nominally quantitative, but comprising primarily desired 2 and starting 1 in a molar ratio of 7:4, along with smaller amounts of other impurities that were not identified further]. A portion (2.0 g) of the crude oil was purified by silica gel chromatography, eluted with hexanes-ethyl acetate (6:1). The purest fractions were placed under petroleum ether at -20°C, whereupon white crystals (0.27 g, >99% pure, 14% yield), m.p. 31-32°C, formed within 24 h. 1 H NMR (CDCl 3 , 300 MHz) δ 2.48 [s, 6H, SCH 3 ].

Refinement
All of the H atoms were positioned geometrically (C-H = 0.96 Å) and refined as riding with U iso (H) = 1.2U eq (C).

Computing details
Data collection: SMART (Bruker, 1998); cell refinement: SMART (Bruker, 1998); data reduction: SAINT (Bruker, 1998    Chemistry used to prepare bis[(methylsulfanyl)carbonyl]disulfane (experimental procedures are provided herein).  Refinement. Refinement of F 2 against all reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on all data will be even larger.