Three sulfur atom insertion into the S-S bond - Pentasulfide preparation

Article abstract of DOI:10.1016/S0040-4039(00)01357-5

Chloro(triphenylmethyl)trisulfide (1) reacts under mild conditions with symmetric primary dialkyl disulfides and aromatic disulfides giving pentasulfides as the main products in good yield and selectivity. A mechanism involving a triphenylmethyl alkyl/phenyl tetrasulfide intermediate is discussed. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)01357-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 7809–7812
Three sulfur atom insertion into the SꢀS
bond—pentasulfide preparation
Yihua Hou, Imad A. Abu-Yousef and David N. Harpp *
a
b
a,
a
Department of Chemistry, McGill University, Montreal, Quebec, Canada H3A 2K6
b
Department of Chemistry, American University of Sharjah, PO Box 26666, Sharjah, United Arab Emirates
Received 6 July 2000; revised 7 August 2000; accepted 8 August 2000
Abstract
Chloro(triphenylmethyl)trisulfide (1) reacts under mild conditions with symmetric primary dialkyl
disulfides and aromatic disulfides giving pentasulfides as the main products in good yield and selectivity.
A mechanism involving a triphenylmethyl alkyl/phenyl tetrasulfide intermediate is discussed. © 2000
Elsevier Science Ltd. All rights reserved.
Keywords: disulfides; pentasulfides; insertion; dithiosulfenyl chloride; sulfur transfer reagent.
1
Pentasulfides occur in nature in a variety of plants and animals. Two pentasulfides,
2
3
diallylpentasulfide and Varacin, possess significant antitumor, antifungal and cytotoxic proper-
4
5
ties. Pentasulfides are also used as antifouling agents and as components of lithium batteries.
Pentasulfides can be synthesized from the reaction of mercaptans or mercaptide salts with
6
6
sulfur halides, by nucleophilic displacement on sulfur or the reaction of methoxycarbonyl
7
trisulfanes with hydrogen sulfide. However, these methods employ reagents that are difficult to
handle or result in the formation of complicated polysulfide mixtures. As a continuation of our
8
study of the reaction of thiosulfenyl chlorides with disulfides, we have discovered that the
reaction of chloro(triphenylmethyl)trisulfide (1) and disulfides 2 (Scheme 1) selectively affords
pentasulfides 3 as the major product in good yield. This reaction can serve as an effective
method to synthesize this difficult class from the more widely available disulfides.
Scheme 1.
*
Corresponding author.
0
040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01357-5

7810
Dithiosulfenyl chloride 1 is a yellow solid prepared from triphenylmethanethiol and sulfur
9
9
monochloride; it is stable in the dark at −15°C for months. Its thermal stability and reactions
1
0
with olefins and dienes have been reported.
The reaction of chloro(triphenylmethyl)trisulfide (1) and a solution of dimethyl disulfide (2a)
in either CH Cl , CHCl , or benzene result in the formation of dimethylpentasulfide (3a) as the
2
2
3
11
main product (Table 1). Dimethyltri-, tetra-, and hexasulfide were obtained as minor products.
In order to inhibit the further reaction of other reaction products with reagent 1, only one
equivalent of 1 is used. This reaction provides a model system to study the influence of solution
concentration, the rate of addition, temperature and solvent.
Table 1
Three sulfur unit insertion into symmetric disulfides
a
b
Entry
R
Sulfide
Time (h)
Product
Selectivity (%)
Yield (%)
1
2
3
4
5
6
7
8
9
CH3
2a
2b
2c
2d
2e
2f
2g
2h
2i
2
2
9
2
3a
3b
3c
3d
86
83
90
87
67
73
74
67
CH CH2
3
PhCH2
HOCH CH2
2
c
2
CH ꢁCHCH
2
d
4-NH Ph
2
4-CH OPh
4-CH Ph
4-NO Ph
i-Propyl
t-Butyl
2
48
3g
3h
71
60
68
44
3
3
e
2
f
1
1
0
1
2j
2k
12
12
3j
86
74
g
3k
a
The percentage of pentasulfides in all polysulfides formed as calculated from the integration of appropriate peaks
1
in the H NMR spectra.
b
Calculated according to peak integration of the crude reaction mixture that includes some starting material and
intermediate 7.
c
Reacted with the double bond.
Dramatic reaction with the NH group.
d
2
e
90% of starting disulfide remained after one week.
f
Underwent disproportionation on the column.
Negligible yield.
g
Solvent polarity is the strongest factor influencing the reaction in that the addition of polar
solvents such as acetic acid or traces of water provide a significant increase in the rate. The
selectivity of the reaction is linked with reaction temperature and solvent. The best procedure
appears to be the following. Dimethyldisulfide (2a) (0.5 mmol, 47 mg) and 1 (0.5 mmol, 187 mg)
were each dissolved in 5 mL of CH Cl . The solution of 2a was quickly added to the solution
2
2
of 1 and the reaction vessel capped. The reaction mixture was stirred at room temperature until
the composition of the reaction mixture remained constant. The solvent was then evaporated
under reduced pressure and the residue was passed through a silica gel (neutral, 230ꢀ400 mesh)
column, using hexane/ethyl acetate as the eluent. The more polar component, triphenylmethyl
chloride (4) hydrolyzed to triphenylmethyl alcohol (5) on the column and was separated cleanly.
However, the complete separation of the main product from the side products by silica gel
column appears impossible. In the case of dibenzyldisulfide (2c), pure pentasulfide product was
obtained by repeated recrystallization.

7811
A strong electronic effect was observed with a series of aryl disulfides with different
substituents at the para position (entries 7–9). Electron-donating groups were found to acceler-
ate the reaction and improve the yield of pentasulfide product while electron-withdrawing
substituents (NO ) kept disulfide 2i essentially inert to the sulfur transfer reagent 1.
2
To explore the limitations of the sulfur-transfer reagent 1, substrates with different functional
groups were examined (entries 4–6). It was discovered that the presence of the OH group in 2d
does not compete with the SꢀS bond to react with 1, but accelerates the reaction. Reagent 1
reacted with the NH group in 2f in a dramatic fashion but the product(s) have not yet been
2
1
identified. When 2e was treated with 1, analysis by H NMR indicated the reaction of 1 with the
double bond took priority over its reaction with the disulfide.
Secondary and tertiary disulfides react more slowly than primary or aromatic disulfides. Three
equivalents of reagent 1 were required to completely react with diisopropyl disulfide (2j). The
pentasulfide 3j that formed underwent rapid disproportionation on a silica gel column. The
reaction of 1 with t-butyl disulfide (2k) provided t-butyl pentasulfide (3k) only as a minor
product even when excess 1 is used in the reaction mixture.
A free radical mechanism can be ruled out due to the strong correlation of reaction rate with
the polarity of the solvent. In addition, adding a radical scavenger to the model reaction did not
alter the rate or outcome of the product mixture. The first step of the reaction is believed to be
8
a nucleophilic cleavage of the SꢀCl bond by the disulfide. This is similar to the reaction of
12
13
arenesulfenyl chlorides with organic disulfides studied by Moore and Pietra. When this attack
takes place with the relatively electron-rich sulfur atom in disulfide 2 engaging the electron-
deficient sulfur in chloro(triphenylmethyl)trisulfide (1), a sulfonium salt 6 is formed as an
intermediate and eventually triphenylmethylalkyl/phenyltetrasulfide (7) and sulfenyl chloride 8
are generated. The subsequent attack of the sulfur atom adjacent to the triphenylmethyl group
in intermediate 7 on the sulfenyl chloride 8 leads to the formation of final product, pentasulfide
9
(Scheme 2). This mechanism is consistent with the electronic effect displayed by the reaction
and the difficulty to induce the same reaction with secondary and tertiary disulfides.
Scheme 2.
There is considerable evidence to support this mechanism. When the reaction of 1 and
1
dimethyldisulfide (2a) was followed by H NMR in CDCl , the appearance and the disappear-
3
ance of peaks relevant to intermediate 7a (R=Me) and the sulfenyl chloride 8a (R=Me) were
observed. It was also noted that when this reaction was performed in anhydrous benzene, the
formation of 7a and 8a proceeded much faster than their further reaction with each other. Thus,
at a certain stage of the reaction (ca. 10–12 h), volatile 8a can be evaporated with the solvent
and 7a left behind. Intermediate 7a was purified by silica gel chromatography followed by
14
recrystallization. Its structure was confirmed by NMR and MS. More convincingly, when pure
a was treated with methanesulfenyl chloride (8a), prepared from dimethyl disulfide and sulfuryl
7
chloride at −15°C, dimethylpentasulfide (3a) was obtained in good yield. In the reaction of cyclic
trisulfide 9 with 1, only two of the three sulfur atoms were ultimately delivered resulting in the

7812
1
5
known pentasulfide 10 in 30% yield (Scheme 3). This apparently reflects the overall stability of
the pentasulfide versus the hexasulfide. We are investigating these effects.
Scheme 3.
Clearly, the reaction of chloro(triphenylmethyl)trisulfide (1) with primary dialkyl disulfides
and aromatic disulfides provides a practical new route to synthesize their pentasulfide homologs.
We are continuing our study of this unusual transformation.
Acknowledgements
The Natural Sciences and Engineering Research Council of Canada supported this work. We
also thank Mr. A. Z. Rys for early insights and helpful discussions.
References
1. Talstikov, G. A.; Shul’ts, E. E.; Tolstikov, A. G. Russ. Chem. Rev. 1997, 66, 813–828.
2. Baba, M.; Takasuka, N.; Onozuka, M.; Masuda, M.; Okuyama, T.; Nishino, H. Wakan Iyagaku Zasshi 1999, 16,
1
02–107.
3
4
5
6
. Ford, P. N.; Davidson, B. S. J. Org. Chem. 1993, 58, 4522–4523.
. Kimura, R.; Hamashima, M. Patent Application, JP 93-132221 19930602; CAN 122268291 AN 1995:522732.
. Naoi, K.; Kawase, K.-I.; Inoue, Y. Proc. Electrochem. Soc. 1996, 96-14, 131–138.
. Picking, T.; Tobolsky, A. V. Sulfur in Organic and Inorganic Chemistry; Marcel Dekker: New York, 1972; Vol.
3
, Chapter 22.
7
8
9
. Mott, A. W.; Barany, G. Sulfur Lett. 1984, 2, 137–142.
. Rys, A. Z.; Harpp, D. N. Tetrahedron Lett. 2000, 41, 7169–7172.
. Williams, C. R.; Britten, J. F.; Harpp, D. N. J. Org. Chem. 1994, 59, 806–812.
1
0. (a) Abu-Yousef, I. A.; Harpp, D. N. J. Org. Chem. 1998, 63, 8654–8660; (b) Abu-Yousef, I. A.; Harpp, D. N.
J. Org. Chem. 1997, 62, 8366–8371; (c) Abu-Yousef, I. A.; Harpp, D. N. Tetrahedron Lett. 1995, 36, 201–204; (d)
Abu-Yousef, I. A.; Harpp, D. N. Tetrahedron Lett. 1994, 35, 7167–7170.
1
1
1
1
1. The compounds are identified according to their reported chemical shifts. See Ref. 7.
2. Moore, C. G.; Porter, M. J. Chem. Soc. 1958, 2890–2892
3. Pietra, F.; Vitali, D. J. Chem. Soc. B. 1970, 4, 623–625.
4. A crystalline white solid (7a) was obtained, mp 92–94°C; H NMR (CDCl ) l=2.48 (s, 3H), 7.31 (m, 15H);
1
13
C
3
+
3
+
NMR (CDCl ) l=23.08, 73.53, 127.23, 127.96, 130.29, 143.03; MS (FAB) m/z 111 [CH S ] (3), 243 [M −MeS ]
3
3
4
+
(
100), 307 [Ph CS ] (5).
3 2
15. Bartlett, P. D.; Ghosh, T. J. Org. Chem. 1987, 52, 4937–4943.
.