Simultaneous 1,2-, 1,3- and 1,4-addition of trithiazyl trichloride to a conjugated diene

Article abstract of DOI:10.1039/a802159g

1,4-Diphenylbuta-1,3-diene and trithiazyl trichloride 1 react to give a bi(thiadiazole) 2, an isothiazoloisothiazole 3, a dithiazolothiazine 4 and the thiazinodithiatriazepines 5 and 6, all of which could arise from initial addition of the trimer 1, or it

Full text of DOI:10.1039/a802159g

View Article Online / Journal Homepage / Table of Contents for this issue
Simultaneous 1,2-, 1,3- and 1,4-addition of trithiazyl trichloride to a
conjugated diene
Charles W. Rees* and Tai-Yuen Yue
Department of Chemistry, Imperial College of Science, Technology and Medicine, London, UK SW7 2AY
S
1,4-Diphenylbuta-1,3-diene and trithiazyl trichloride 1 react
Ph
N
N
N
to give a bi(thiadiazole) 2, an isothiazoloisothiazole 3, a
dithiazolothiazine 4 and the thiazinodithiatriazepines 5 and
6, all of which could arise from initial addition of the trimer
1, or its monomer, to the conjugated diene by 1,2-,1,4- and,
for the first time to an all-carbon diene, 1,3-(‘criss-cross’)
cycloaddition reactions.
S
Ph
N
Ph
S
N
N
S
Ph
2
3
Colourless, mp 79–80 °C
Colourless, mp 200–202 °C
We have shown that trithiazyl trichloride (NSCl)3 1, in thermal
equilibrium with its monomer N°SCl,1 converts monoenes and
monoynes directly into 1,2,5-thiadiazoles.2 Furthermore,
1-alkyl-2,5-diphenylpyrroles are similarly converted into the
bi(1,2,5-thiadiazole) 2, thus reacting as masked 1,3-dienes.3 It
would be interesting therefore to explore the reaction of the
reagent 1 with acyclic conjugated dienes where 1,3- and
1,4-cycloaddition, as well as 1,2-cycloaddition, can be envis-
aged, as shown in Scheme 1.
When (E,E)-1,4-diphenylbuta-1,3-diene was treated with a
deficiency of 1 (1 mol) in refluxing CCl4 for 30 min a rapid and
complex reaction ensued, from which five crystalline organic
compounds 2–6 were isolated in low yields† by careful
chromatography and shown to be derivatives of five different,
rare or new heterocyclic ring systems (Scheme 2). The same
five products were obtained in higher yields,† shown in brackets
in Scheme 2, when the diene was treated with 1 (1 mol) in
refluxing toluene for 1 h. With more trimer (3 mol) in refluxing
toluene for 16 h, the reaction was less complex and 2 was the
major product (60%). In spite of their very different structures,
the carbon connectivity of the starting diene is retained in all of
these products.
8% (40%)
1% (3%)
Ph
Ph
+
1
Ph
Ph
Ph
N
S
N
S
Ph
N
S
N
N
S
S
O
N
S
S
N
N
S
N
N
S
Ph
4
6
Ph
Yellow, mp 184–186 °C
Red, mp 190–192 °C
5
10% (10%)
3% (10%)
Brown, mp 178–180°C
7% (11%)
Scheme 2
formed from 4 by loss of O and from 5 by loss of N2. This
suggested the new thiazinodithiatriazepine structure 5 for the
brown product, which agreed with all its spectroscopic
properties, including the dominant loss of N2 in the mass
spectrum. Conversion of 5 into 4 could result from oxidation at
the least hindered sulfur atom, to give 7, followed by extrusion
of N2.
The bi(1,2,5-thiadiazole) 2 was identical with that formed
from N-alkyl-2,5-diphenylpyrroles.3
The isothiazolo[5,4-d]isothiazole structure 3 was suggested
by its high stability and symmetry (1H and 13C NMR) and its
mass spectrum, which gave fragments for PhCN and C2S2. This
product was synthesised independently from the oxime of
5-benzoyl-3-phenylisothiazole4 and disulfur dichloride in DMF
at 100 °C, providing another approach to this rare ring
system.
The new dithiazolo[4,5-d]thiazine S-oxide structure 4 was
based on its spectroscopic properties and confirmed by X-ray
crystallography.5 The sulfoxide group suggested that formation
of 4 had involved oxidation, possibly during isolation; an
examination of the five products 2–6 showed that all were stable
to air except for the brown compound 5 which was oxidised to
4 when adsorbed on silica. Compound 5 was rapidly and cleanly
converted into 4 by MCPBA in CH2Cl2 at room temperature.
Link scan mass spectrometry showed that 4 and 5 gave the same
pattern of daughter ions, the same species (C16H10N2S3) being
Ph
O
N
S
Cl
N
S
S
N
N
S
Ph
3
N
Ph
Ph
Cl
S
N
Ph
7
8
Ph
N
N
9
1,2-addition of NSN
R
Cl
S
The structure of the final product 6, which is isomeric with 5,
was assigned on the basis of spectroscopic properties and a
chemical degradation. Its molecular ion was much stronger than
that of 5 and showed no loss of N2.
The formation of all five products can be explained by initial
1,2-, 1,3- and 1,4-cycloaddition processes (Scheme 1). The
bi(thiadiazole) 2 is probably formed by addition of the trimer, as
R
N
S
N
S
3 NSCl
R
R
Cl
N
Cl
1,3-addition of NS 1,4-addition of NS
Scheme 1
1
Chem. Commun., 1998
1207

View Article Online
Ph
Ph
Ph
Ph
N
Cl
S
N
S
Ph
H
HN
S
Ph
H
N
S
N
S
Ph
S
N
Cl
N
N
Cl
Cl
H
N
Cl
S
N
Cl
S
S
N
Ph
10
S
Cl
SCl
Cl
–HCl
S
Cl
–2 HCl
–SNSCl
Ph
Ph
Ph
H
N
S
N
N
S
N
S
N
S
N
N
–HCl
Ph
Ph
repeat
S
S
2
S
N
Cl
Cl
Ph
Ph
N
N
Cl
S
–HCl
Scheme 3
6
N
S
N
N
S
N
S
an N–S–N unit, across C1–C2 and C3–C4, exactly as for
monoenes2 and pyrroles,3 as shown in Scheme 3.
(3)
S
(1)
air
The structure of the isothiazoloisothiazole 3 suggests the
addition of S–N units, possibly the monomeric species N°SCl,
across C1–C3 and C2–C4, as shown in 8. Subsequent elimination
of HCl and oxidation, possibly via chlorination, would give the
stable aromatic system 3. Such ‘criss-cross’ cycloadditions6
have been reported for azabutadienes, particularly for azines
such as 1,4-diphenyl-2,3-diazabuta-1,3-diene 9,7 but the present
reaction is, we believe, the first example of a criss-cross
cycloaddition to an all-carbon diene.
The 1,2-thiazine ring common to the remaining products
suggests the 1,4-addition of an N–S unit, probably the
monomer, to the diene. Since at least four molecules of
monomer are required to form compounds 5 and 6, it is possible
that the Diels–Alder process is followed by reaction with a
molecule of trimer, with overall elimination of 4 HCl units to
give the intermediate 11 (Scheme 4). Initial Diels–Alder
reaction, elimination of HCl and a hydrogen shift would give
3,6-diphenyl-1,2-thiazine 10, which is nucleophilic at C4.
Attack on the trimer 1 at nitrogen, followed by a second
nucleophilic attack by C5 at sulfur and elimination of the
remaining HCl, could give the key intermediate 11. This is
formally a 16p antiaromatic system which could become 14p
by extrusion of any one of the sulfur atoms. Extrusion of S3 or
S1 would then lead to the observed products 5 and 6,
respectively, and 4 has been shown to arise by oxidation and N2
extrusion from 5.
5
4
Ph
11
Scheme 4
It is hoped that similar reactions of unsymmetrical and
functionalised conjugated dienes will prove to be more selective
and hence of synthetic as well as mechanistic value.
We gratefully acknowledge financial support from the
Commonwealth Scholarship Commission and the British Coun-
cil (T-Y. Y.) and from MDL Information Systems (UK) Ltd,
Professor D. J. Williams for the X-ray structure determination,
the Wolfson Foundation for establishing the Wolfson Centre for
Organic Chemistry in Medical Science at Imperial College, and
a referee for valuable comments.
Notes and References
† Yields are based on the minimum amount of trimer 1, which is in
deficiency, required to give each product.
1 Gmelin Handbook of Inorganic Chemistry, ed. B. Heibel, Sulfur-
Nitrogen Compounds, Part 2, Springer-Verlag, Berlin, 8th edn., 1985,
p. 92.
2 X.-G. Duan, X.-L. Duan, C. W. Rees and T.-Y. Yue, J. Chem. Soc.,
Perkin Trans. 1, 1997, 2597.
3 X.-G. Duan and C. W. Rees, J. Chem. Soc., Perkin Trans. 1, 1997,
3189.
Thus the initially puzzling array of products from this one
diene can be explained by invoking all three cycloaddition
modes of the diene with the reactive trimer 1 and its more
reactive monomer. At present the only product produced in high
yield is the bi(1,2,5-thiadiazole) 2, under more vigorous
conditions; the analogous di-p-tolyl compound was also the
main product from the same reaction of the corresponding
butadiene.
4 X.-L. Duan, R. Perrins and C. W. Rees, J. Chem. Soc., Perkin Trans. 1,
1997, 1617.
5 D. J. Williams, Imperial College, unpublished results.
6 1,3-Dipolar Cycloaddition Chemistry, ed. A. Padwa, Wiley, Chichester,
1984, vol. 1.
7 T. Wagner-Jauregg, Synthesis, 1976, 349; S. Ra´dl, Aldrichim. Acta, 1997,
30, 97.
Received in Liverpool, UK, 18th March 1998; 8/02159G
1208
Chem. Commun., 1998