A new, unexpected synthesis of 1,3-dithietanones

Article abstract of DOI:10.1039/b302434b

Treatment of a geminal pivaloxy xanthate, prepared by an intermolecular radical addition of a xanthate to vinyl pivalate, gives a 1,3-dithietanone, a little known class of compounds.

Full text of DOI:10.1039/b302434b

A new, unexpected synthesis of 1,3-dithietanones
Béatrice Quiclet-Sire, Graciela Sanchez-Jimenez and Samir Z. Zard*
Laboratoire de Synthèse Organique associé au CNRS, Ecole Polytechnique, 91128 Palaiseau, France.
E-mail: zard@poly.polytechnique; Fax: +33(0)169333851; Tel: +33(0)169334872
Received (in Cambridge, UK) 4th March 2003, Accepted 23rd April 2003
First published as an Advance Article on the web 13th May 2003
Treatment of a geminal pivaloxy xanthate, prepared by an
intermolecular radical addition of a xanthate to vinyl
pivalate, gives a 1,3-dithietanone, a little known class of
compounds.
concomitant formation of ethyl chloride, as depicted in Scheme
2. With a convenient and highly efficient access to the
precursors by the radical addition of a xanthate to vinyl pivalate,
we could easily explore the generality of this transformation. As
indicated by the examples collected in Table 1, both aromatic
and aliphatic structures can be made and groups other than
ketones (e.g. nitrile in 6j) can be present. The reaction was not
clean in the case of adduct 2g containing a simple methyl
ketone. A better result was observed with 2h where the aliphatic
ketone is sterically shielded. The relative sensitivity of
cyclopropyl ketones to Lewis acids may have been one of the
We recently described a practical and flexible synthesis of
pyrroles outlined in Scheme 1.1 It is based on the radical
addition of an acetonyl xanthate 1 to vinyl pivalate2 followed by
heating of the adduct 2 with an amine. Not unexpectedly,
treatment of the 2 with acid in the absence of amine produced
the corresponding furan 4 but this reaction was somewhat
capricious in our hands. Thiophene 5 was also formed in small
amounts in certain cases.
In order to improve the access to either the furans or the
thiophenes, we examined the effect of a Lewis acid in a non-
protic solvent. We reasoned that activation of the ketone by the
Lewis acid would provoke a nucleophilic attack either by the
oxygen of the pivalate to give the furan (path a) or by the sulfur
of the xanthate to furnish ultimately the thiophene (path b).
In the event, exposure of xanthate 2a to the action of titanium
tetrachloride in dichloromethane gave neither the furan nor the
thiophene in significant amounts.† The major product, isolated
in quite good yield (81%), had lost both the xanthate and
pivalate groups and exhibited, in addition to the ketone stretch
band at 1690 cm21, a strong band at 1775 cm21 suggesting the
presence of a strained carbonyl group. All the spectral and
analytical data were consistent with 1,3-dithietanone 6a.
1,3-Dithietanones are very rare and only a few have been
described in the past. Most are halogenated derivatives,
accessible by partial hydrolysis of thiophosgene photodimers.3
4-Alkylidene-1,3-dithietanones are somewhat more common
since these can be made by reacting the dianion of dithio-
carboxylic acids with phosgene.4
Scheme 2 Unexpected formation of 1,3-dithietanones.
Table 1 Synthesis of 1,3-dithietanones 6 from radical adducts 2
Xanthate adduct 2
1,3-Dithietanone 6
This unexpected reaction proceeds presumably through an
initial complexation of the Lewis acid with the carbonyl of the
pivalate group which triggers a nucleophilic attack by the
xanthate sulfur followed by cleavage of the intermediate with
Scheme 1 Synthesis of pyrroles, furans, and thiophenes.
1408
CHEM. COMMUN., 2003, 1408–1409
This journal is © The Royal Society of Chemistry 2003

The solvent was then removed under vacuum and the residue purified by
flash chromatography over silica gel (ethyl acetate–petroleum ether 5 : 95;
a small layer of basic alumina is placed on top of the silica column to remove
any lauric acid present). Adduct 2a was obtained as a yellowish oil (1.40 g;
93%) and was used as such in the next step; 1H NMR (400 MHz; CDCl3):
1.19 (s, 9H), 1.39–1.42 (m, 3H), 2.40–2.44 (m, 2H), 3.12–3.16 (m, 2H),
4.61–4.68 (m, 2H), 6.72 (t, J 6.4, 1H), 7.47 (d, J 6.8, 2H), 7.49–7.57 (m,
1H), 7.95 (d, J 7, 2H); 13C NMR (100 MHz; CDCl3) d (ppm): 13.7, 28.6,
34.2, 43.6, 70.3, 80.3, 128.1, 128.7, 133.4, 136.5, 176.8, 197.9, 210.1; IR
(CCl4, cm21): 1739 (CNO), 1692 (CNO), 1227 (CNS), 1051(C–S); m/z (CI)
385 (M + NH4+).
To a stirred solution of TiCl4 (0.1 mL, 1 mmol) in freshly distilled
dichloromethane (1 mL) was added a solution of adduct 2a (200 mg, 0.5
mmol) in dichloromethane (1 mL). After stirring for 1 hour at room
temperature and under an inert atmosphere, the reaction mixture was cooled
in an ice bath and water was slowly added. The organic layer was separated,
dried over sodium sulfate, and concentrated under reduced pressure. The
residue was purified by flash chromatography over silica gel (ethyl acetate–
petroleum ether 5 : 95) to give 1,3-dithietanone 6a as a white solid (102 mg,
81%); it was recrystallised from dichloromethane–petroleum ether; mp
110–112 °C; 1H NMR (400 MHz; CDCl3) 2.61 (q, J 6.8, 2H), 3.23 (t, J 6.6,
2H), 4.35 (t, J 6.8, 1H), 7.49 (t, J 7.4, 2H), 7.61 (t, J 7.6, 1H), 7.97 (d, J 7.6,
2H); 13C NMR (100 MHz; CDCl3) d (ppm): 26.9, 31.8, 36.2, 128.2, 128.9,
133.8, 136.3, 173.7, 197.8; IR (CCl4, cm21): 1775 (CNO), 1690 (CNO); m/z
(CI) 239 (M + H+). Calc. for C11H10O2S2 (%): C, 48.44; H, 3.33. Found (%):
C, 48.68; H, 3.45.
Scheme 3 Some reactions of 1,3-dithietanones.
causes of the poor yield of 6i. The conditions we have used in
this initial study have not been optimised; nevertheless, the
yields are synthetically useful. Lowering the temperature did
not result in appreciable improvement and replacing titanium
tetrachloride with BF3·OEt2 did not prove very satisfactory:
relatively complex mixtures were obtained containing furan and
thiophene among other compounds. The nature of the Lewis
acid is clearly important but further, more extensive screening is
still needed.
Little is known of the chemistry of the [1,3]dithietanone
group. Heating compound 6d with benzylamine in refluxing
dioxane resulted in a good yield of pyrrole 3d (Scheme 3). In
contrast, heating dithietanone 6a in refluxing dichlorobenzene
for two days gave rise to thiophene 5a in a disappointing 20%
yield. The resistance to heat of the dithietanone motif was
surprising.
1 B. Quiclet-Sire, L. Quintero, Graciela Sanchez-Jimenez and S. Z. Zard,
Synlett, 2003, 75.
2 For reviews on the xanthate transfer reaction, see: S. Z. Zard, Angew.
Chem., Int. Ed. Engl., 1997, 36, 672–685; S. Z. Zard, Xanthates and
Related Derivatives as Radical Precursors, in Radicals in Organic
Synthesis (Eds. P. Renaud, M. P. Sibi), Wiley-VCH, Weinheim, 2001,
Vol. 1, , p. 90.
This preliminary study has provided a simple, flexible, and
convergent access to an unusual class of compounds. Further
work is necessary to better delineate the scope of this approach
and explore its synthetic potential.
3 M. Délépine, L. Labro and F. Lange, Bull. Soc. Chim. Fr., 1935,
1969–1980; G. Diderrich, A. Haas and M. Yazdanbakhsch, Chem. Ber.,
1976, 110, 916–920; R. Schork and W. Sundermeyer, Chem. Ber., 1985,
118, 1415–1420; A. Waterfeld, Chem. Ber., 1990, 123, 1635–1640.
4 K. Dickore and R. Wegler, Angew. Chem., Int. Ed. Engl., 1966, 5, 970; R.
Schulz and A. Schweig, Angew. Chem., Int. Ed. Engl., 1980, 19, 740; R.
Grashey, M. Baumann and R. Hauptrecht, Tetrahedron Lett., 1970,
5083–5085; E. Schaumann and F.-F. Grabley, Liebigs Ann. Chem., 1979,
1715–1733; E. Schaumann, E. Kausch and E. Rossmanith, Liebigs Ann.
Chem., 1978, 1543–1559; E. Schaumann, U. Wriede and G. Adiwidjaja,
Chem. Ber., 1984, 117, 2205–2225; T. Tanaka, T. Hashimoto, K. Lino, Y.
Sugimura and T. Miyadera, J. Chem. Soc., Chem. Commun., 1982,
713–714.
Notes and references
†
Typical example: a solution of xanthate 1a (1, R = Ph; 1.00 g, 4.2
mmol) and vinyl pivalate (1.23 mL, 8.4 mmol) in 1,2-dichloroethane (4.2
mL) was heated under nitrogen for 15 min. A first batch of lauroyl peroxide
(5 mol%) was added followed by another batch (2.5 mol%) after 90 min.
and heating was continued until almost complete consumption of the
xanthate (further amounts of lauroyl peroxide may be added if necessary).
CHEM. COMMUN., 2003, 1408–1409
1409