Reactions of 1,2,3-dithiazoles with halogenated malononitriles

Article abstract of DOI:10.1039/b202038f

Dibromomalononitrile 3a reacts with 4-chloro-1,2,3-dithiazole-5-thione 9 and monobromomalononitrile 3c reacts with 4,5-dichloro-1,2,3-dithiazolium chloride 4 (Appel salt) to give 4-chloro-5H-1,2,3-dithiazole-5-ylidenemalononitrile 1 in 76 and 73% yields,

Full text of DOI:10.1039/b202038f

View Article Online / Journal Homepage / Table of Contents for this issue
Reactions of 1,2,3-dithiazoles with halogenated malononitriles
Irene C. Christoforou,a Panayiotis A. Koutentisa and Charles W. Reesb
a Department of Chemistry, University of Cyprus, P.O. Box 20537, 1678 Nicosia, Cyprus.
E-mail: koutenti@ucy.ac.cy
1
b Department of Chemistry, Imperial College of Science, Technology and Medicine, London,
UK SW7 2AY. E-mail: c.rees@ic.ac.uk
Received (in Cambridge, UK) 27th February 2002, Accepted 4th April 2002
First published as an Advance Article on the web 17th April 2002
Dibromomalononitrile 3a reacts with 4-chloro-1,2,3-dithiazole-5-thione 9 and monobromomalononitrile
3c reacts with 4,5-dichloro-1,2,3-dithiazolium chloride 4 (Appel salt) to give 4-chloro-5H-1,2,3-dithiazole-5-
ylidenemalononitrile 1 in 76 and 73% yields, respectively, thus providing the best available synthesis of this key
intermediate. The reactions were accompanied by the formation of 1-(4-chloro-5H-1,2,3-dithiazol-5-ylideneamino)-
1,2-dihaloethene-2-carbonitriles 5, the yields of which increased with reaction temperature. In refluxing toluene,
dibromomalononitrile 3a and dithiazolethione 9 give directly 3-bromoisothiazole-4,5-dicarbonitrile 12 (59%);
the probable intermediate (dicyanomethylene)dithiazole 1 is readily converted into bromoisothiazole 12 on
treatment with anhydrous gaseous HBr at ca. 20 ЊC (83%). The addition of bromine to dithiazolethione 9 gives
5-bromosulfanyl-4-chloro-1,2,3-dithiazolium bromide 11 almost quantitatively. Mechanisms are proposed for
all these reactions.
The (dicyanomethylene)dithiazole 1 can be readily converted
into the fully functionalised isothiazole 2 in quantitative
yield.1,2 The isothiazole is an important building block for a
variety of potent biocides;3 however, access to isothiazole 2
is limited by the availability of ylidene 1. The preparation of
ylidene 1 from malononitrile and Appel salt gave the ylidene
in only 40% yield and required chromatography1 whilst altern-
ative higher yielding (60 to 70%) preparations from either
Appel salt 4 or dithiazolethione 9 used the expensive reagent
tetracyanoethylene oxide (TCNEO), and both reactions gave
Scheme 1
unexpected by-products.1,2
Table 1 Reaction of halogenated malononitriles 3 with Appel salt 4
(0.50 mmol)
Yields (%)
3 (equiv.) Solvent (3 ml) T /ЊC t/h
5a,bc
5c
1
3a (1)
3a (1)
3a (1)
3a (2)
3a (2)
3a (2)
3a (3)
3a (4)
3b (2)
3b (2)
3b (2)
3b (2)
3b (2)
3c (1)a
3c (2)b
3c (1)
3c (2)
3c (1)
3c (2)
3c (2)
DCM
DCM
PhH
20
40
80
24
24
24
24
24
24
24
24
24
24
24
48
24
12
12
24
24
Trace
Trace
4
11
2
13
16
13
–
–
–
–
–
–
–
–
–
–
–
–
–
Trace
7
13
35
38
56
46
51
0
4
11
5
15
57
73
32
62
45
54
51
In an attempt to overcome these limitations we sought new
routes to ylidene 1 from Appel salt 4 or thione 9 by reaction
with malononitrile derivatives which are mechanistically
equivalent to TCNEO, but are simpler and cheaper. Since
PhH
80
PhMe
PhCl
PhH
PhH
DCM
PhH
PhMe
PhMe
PhCl
DCM
DCM
DCM
DCM
PhH
110
132
80
80
40
TCNEO appears to react through its ring opened stabilized
0
5
1,2,4
ylide form, (NC) C᎐Oϩ–CϪ(CN) ,
we need to generate a
᎐
2
2
80
dicyanomethyl carbanion with a leaving group on the central
carbon atom. In the ylide form of TCNEO the leaving group,
–Oϩ᎐C(CN) , could possibly be replaced by halogen. Bromo-,
110
110
132
20
20
40
40
80
80
132
21
17
17
–
–
–
–
–
–
–
᎐
2
Trace
2
12
22
dibromo- and dichloromalononitrile, which are all readily
prepared from malononitrile in one step, were therefore
investigated as TCNEO substitutes.
2.5 33
1.5 39
0.5 34
PhH
PhCl
Reaction of Appel salt 4 with halogenated malononitriles 3a–c
Treatment of Appel salt 4 with dibromo-, dichloro- and
bromomalononitriles 3a–c did indeed give the desired ylidene 1
together with the dithiazolimines 5a–c (Scheme 1) in moderate
yields (Table 1). The reactions gave only traces of the dithiazo-
lethione 9 and the corresponding dithiazolone (by TLC).
Monobromomalononitrile 3c was by far the most reactive of
the malononitriles 3a–c followed by the dibromo- and finally
a With pyridine (1 equiv.). b With pyridine (2 equiv.). c Predominantly
5b.
the dichloromalononitrile. Higher reaction temperatures gave
better yields of the dithiazolimines 5 and an excess of the
halogenated malononitriles improved the yields of both ylidene
1236
J. Chem. Soc., Perkin Trans. 1, 2002, 1236–1241
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b202038f

View Article Online
1 and imines 5. The dithiazolimines 5 were mixtures of E/Z
isomers and, because of difficulties in their chromatographic
separation, they were isolated and characterized as mixtures. In
the dibromomalononitrile reactions, mixtures of dithiazol-
imines 5a and 5b were obtained. The ylidene obtained from
the reaction of dihalomalononitriles 3a and 3b was dark brown
in colour even after chromatography, which indicated some
unidentified contamination. Using bromomalonitrile 3c the
ylidene was obtained as a bright orange material. The best
conditions for the preparation of pure ylidene 1 (73%) involved
the use of two equivalents of bromomalononitrile 3c with
pyridine (2 equiv.) in DCM at ca. 20 ЊC and these conditions
gave only traces of dithiazolimines 5.
With bromomalononitrile 3c and pyridine, deviation from
an equimolar ratio dramatically reduced the yields of dithiazol-
imines 5. The imines, however, were stable to pyridine in DCM
at ca. 20 ЊC. At reflux in DCM or in PhH no pyridine was
necessary and the yields of dithiazolimine 5 were improved.
With or without pyridine, 3 or 4 equivalents of bromomalono-
nitrile 3c resulted in significant contamination of ylidene 1 with
malononitrile, even after chromatographic work-up; however,
this contamination was less apparent (by 1H NMR) when
pyridine was used. The malononitrile could be washed out
with cold diethyl ether or ethanol. The identification of
malononitrile (by 1H NMR) suggested disproportionation;
however, whilst dibromomalononitrile 3a and malononitrile
disproportionate to give bromomalononitrile in the presence of
BF3 in refluxing EtOH,5 bromomalononitrile 3c did not dis-
proportionate in refluxing DCM or PhH after 48 h. In refluxing
PhMe or PhCl, however, bromomalononitrile 3c did give some
malononitrile and other products. Bromomalononitrile 3c is
known to react with bases to give pentacyanopropenide anions
and bromomalononitrile anions.6 The addition of pyridine at
ca. 20 ЊC to a DCM solution of bromomalononitrile 3c gave a
complex mixture.
Scheme 2
brominating capabilities of dibromomalononitrile.5 With
monobromomalononitrile 3c the hydrogen is more labile than
the bromine and this could account for the improved yields in
the presence of base.
The reaction proceeds either by a concerted or stepwise
thermal loss of dihalogen from adduct 6 and whilst no evidence
is available to differentiate between the possible dehalogenation
pathways, the stepwise process which utilizes the aromatic
dithiazolium intermediate 7 is preferred.
Formation of dithiazolimines 5 can be similarly explained
from the reaction of malononitrile anion now acting as a nucleo-
phile through nitrogen (Scheme 3). The halogen on the ethene
The E- and Z-dithiazolimines 5c obtained from the dichloro-
malononitrile 3b reaction, were identical with the dithiazol-
imines obtained from the reaction of Appel salt with TCNEO.2
The imines obtained from the reactions of monobromo- and
dibromomalononitrile 3c and 3a were deeper red in colour
and less soluble than dithiazolimines 5c; but they all co-run on
silica TLC. Mass spectrometry of the dithiazolimines obtained
from the dibromomalononitrile reaction indicated the imines
were a mixture of bromochloroethene and dibromoethene
substituted materials (m/z 317 and 361). A clean 13C NMR
spectrum was not obtainable from this mixture, but elemental
analysis and MS suggested that the mixture was predominantly
dithiazolimines 5b. We were unable to separate the mixture
of imines 5a and 5b using standard chromatography
techniques. In contrast the dithiazolimines obtained from
monobromomalononitrile 3c and Appel salt gave a very clean
spectrum of only 10 carbon signals in the range 162–80 ppm,
tentatively assigned to the two E/Z isomers of the dithiazol-
imines 5b. This was supported by elemental analysis which gave
a formula of C5BrCl2N3S2 and MS which showed no parent ion
corresponding to imines 5a. The 10 13C NMR peaks were also
identifiable in the complex 13C NMR spectrum of the imines
from the dibromomalononitrile reactions. The two peaks (115.7
and 114.5 ppm), probably derive from the nitrile groups. In a
direct comparison with the data obtained for the dithiazol-
imines 5c the spectra were remarkably similar, and the most
significant differences between these 13C NMR signals and
those of the dithiazolimines 5c was the upfield shift of the
ethylene carbon attached to the nitrile (87.5 and 81.8 ppm in
5b vs. 100.3 and 96.0 ppm in 5c). This upfield shift tentatively
supports the presence of bromine rather than chlorine in this
position, which is also reasonable mechanistically.
Scheme 3
carbon β to the nitrile group in dithiazolimine 5 could be
introduced by addition to the keteneimine 8, a highly reactive
Michael acceptor. The ratio of imines 5 to ylidene 1 tends to
increase at higher reaction temperatures (Table 1).
Reaction of dithiazolethione 9 with halogenated malononitriles 3
Malononitrile does not react with 4-chloro-1,2,3-dithiazole-
5-thione 9 even under reflux (DCM, PhH, PhCl) or with the
addition of bases such as pyridine. Analogous conditions
but using bromomalononitrile 3c gave mainly unreacted
starting thione 9 (85–97%) and traces of the ylidene 1 (1–7%).
Dithiazolethione 9, however, reacts with dihalomalononitriles
3a and 3b to give ylidene 1, dithiazolimines 5a and c, and elem-
ental sulfur (Scheme 4). By using a three- or fourfold excess
of dibromomalononitrile 3a, the reaction could be driven to
completion in refluxing DCM to afford good yields of ylidene
(70–76%) with only minor by-products. Higher temperatures
resulted in the formation of small quantities of dithiazolimines
5. Dichloromalononitrile 3b was significantly less reactive than
dibromomalononitrile and the reaction could not readily be
The mechanism for the formation of ylidene 1 (Scheme 2)
probably involves ionization of the halogenated malononitrile
to the anion. Such a possibility is supported by the strong
J. Chem. Soc., Perkin Trans. 1, 2002, 1236–1241
1237

View Article Online
Scheme 4
Table 2 Reaction of dihalomalononitriles 3a and 3b with dithiazo-
lethione 9 (0.60 mmol) at reflux
Yields (%)
9
5aa
5ca
1a
Scheme 6
3 (equiv.)
Solvent (3 ml)
t/h
1,2-dibromoethene. In contrast with the Appel salt reaction
(Scheme 1), the bromo(chloro)ethene 5b is not expected here
since there is no source of chloride except from the starting
thione 9. This was supported by elemental analysis which gave
a formula of C5Br2ClN3S2 and by MS which showed no parent
ion for dithiazolimines 5b. As proposed in Scheme 3, the
bromomalononitrile anion is an ambident anion which starts
to attack through nitrogen as the temperature is raised.
3a (1)
3a (2)
3a (3)
3a (1)
3b (2)
3b (2)
3b (2)
3b (2)
3b (2)
DCM
DCM
DCM
PhH
DCM
PhH
PhMe
PhMe
PhCl
24
24
16
4
24
24
24
48
24
66
23
Trace
Trace
100
86
31
7
3
0
0
0
19
–
–
–
–
–
54
60
76
76
0
85
64
63
45
0
–
–
–
–
Trace
10
6
24
To gain support for the mechanisms of Scheme 5 and 6, we
a Based on consumed thione 9.
prepared the sulfenyl bromide 10 (X᎐Br) independently and
᎐
treated this with the halomalononitriles 3. The dithiazolethione
9 on treatment with bromine in DCM or in MeCN gave a
yellow–orange crystalline precipitate believed to be 5-bromo-
sulfanyl-4-chloro-1,2,3-dithiazolium bromide 11, mp 162–164
ЊC. Mass spectroscopic analysis (EI) did not identify the parent
ion but indicated a strong ion for the thione (m/z 169, 100%)
and dibromine (m/z 160, 47%); however, the milder chemical
ionization techniques using methane as a proton source (CI/
CH4) did give an ion for the cation (Mϩ, 250, 1%) which
was supported by elemental analysis, giving the formula
C2Br2ClNS3.
driven to completion. Generally both dihalomalononitriles
afforded minor quantities of the dithiazolimines 5 (Table 2).
As expected, the electron-rich dithiazolethione 9 was not
susceptible to nucleophilic attack by either malononitrile or
bromomalononitrile anions. Pure bromomalononitrile 3c was
stable in refluxing PhH and showed no signs of disproportion-
ation to dibromomalononitrile 3a and malononitrile (by
NMR), but it was consumed during the reaction with dithiazole-
thione 9. Reasonable mechanisms to explain the formation of
the products are proposed in Schemes 5 and 6. The halophilic7
The dithiazolium bromide 11 was relatively stable, but on
standing in the atmosphere or on treatment with pyridine it
reverted predominantly to the dithiazolethione 9. The salt was
not sufficiently soluble for the acquisition of its 13C NMR spec-
trum. Similar salt formations involving the addition of molecu-
lar bromine to 1,2-dithiole-3-thiones10,11 and 1,3-dithiole-2-
thiones12 are known.
Bromosulfanyl salt 11 was then treated with malononitrile,
bromomalononitrile 3c, or dibromomalononitrile 3a in reflux-
ing benzene for 24 h to give thione 9, dithiazolimines 5 and
dithiazole ylidene 1 in the yields shown in Table 3. These
reactions of dithiazolium salt 11 with the malononitriles were
sluggish but this was not unexpected since single crystal X-ray
diffraction studies of closely related 4-chloro-5-methylsulfanyl
dithiazolium perchlorate13 indicate that the cationic charge
is evenly distributed between the sulfenyl sulfur and the
dithiazole ring sulfur S-1, making C-5 less electrophilic.
Scheme 5
dithiazolethione 9 could first abstract halogen from the
halogenated malononitriles 3 to form the sulfenyl halide 10
and the malononitrile anion. These two species could combine
to give dithiazole 6 proposed above (Scheme 2), which
subsequently collapses to ylidene 1 and possibly sulfur dihalide.
Both monobromo-8 and dibromomalononitrile are known
brominating agents,5 but the lack of reaction with bromo-
malononitrile suggests that the latter is less susceptible to
nucleophilic attack on bromine. The alternative SN2 attack of
the thione at carbon seems less likely.9
Table 3 Reaction of the sulfenyl bromide 11 (0.30 mmol) with
malononitriles in PhH at 80 ЊC for 24 h
The dithiazolimines 5a and 5c were observed depending
on the starting dihalomalononitrile; a mechanism for their
formation is proposed in Scheme 6. 13C NMR spectroscopy of
Yields (%)
9
3 (2 equiv.)
5a
1
the dibromoethene dithiazolimine 5a (X᎐Br) showed a strong
᎐
CH2(CN)2
CHBr(CN)2
CBr2(CN)2
25
0
0
Trace
6
22
36
54
49
set of five carbon resonances together with a weaker set,
suggesting that one of the E/Z isomers was much preferred;
for steric reasons the major isomer is likely to be the trans
1238
J. Chem. Soc., Perkin Trans. 1, 2002, 1236–1241

View Article Online
Table 4 Preparation of bromoisothiazole 12 from dibromomalononitrile 3a and dithiazolethione 9 (0.60 mmol)
Yields (%)
BnBr
3a (equiv.)
Solvent (3 ml)
T /ЊC
t/h
12
1
5
1
1
1
2
2
2
3
PhH
80
110
110
110
120
120
120
2
2
24
2
2
24
2
0
Nd
78
85
0
0
0
0
37
20
59
Trace
10
7
76
54
40
Trace
56
60
76
11
5
0
Trace
10
2
PhMe
PhMe
PhMe
PhCl
PhCl
PhCl
7
Presumably as thione 9 dehalogenates the malononitriles to
generate the dithiazolium salt and the malononitrile anion, the
close proximity of the paired ions will strongly facilitate the
further reaction.
ylidene 1 with dibromine (2 equiv.) at ca. 20 ЊC, in PhMe gave
bromoisothiazole 12 in good yield. However, the reaction was
complex and the product required repeated recrystallisation to
remove impurities. In general agreement with these mechanistic
proposals, we would expect dibromine to convert ylidene 1 into
isothiazole 12 in PhMe, but not in PhCl, and this proved to be
the case.
Preparation of 3-bromoisothiazole-4,5-dicarbonitrile 12
Somewhat unexpectedly, in PhMe at ca. 110 ЊC dibromo-
malononitrile 3a and dithiazolethione 9 gave 3-bromo-4,5-
dicyanoisothiazole 12, and benzyl bromide 13. In PhCl at
120 ЊC some isothiazole 12 was formed but the reaction was
significantly slower suggesting that the methyl substituent
of the toluene may be involved in the reaction. Prolonged
reaction times diminish the product yields. The analogous
reaction with Appel salt 4 in place of dithiazolethione 9 did
not yield isothiazole 12, in refluxing PhH, PhMe or PhCl,
although bromination of toluene to give benzyl bromide was
observed.
In contrast, it should be noted that the reaction of dithiazo-
lethione 9 with dichloromalononitrile 3b (4 equiv.) in PhMe at
110 ЊC after 24 h gave neither chloroisothiazole 2 nor benzyl
chloride, and treatment of ylidene 1 with anhydrous gaseous
HCl at ca. 20 ЊC gave only a trace of the chloroisothiazole 2 and
almost all of the starting ylidene 1 was recovered. This parallels
the much slower addition of HCl to malononitrile itself
compared to that of HBr.15
The conversion of ylidene 1 into isothiazole 12 on treatment
with HBr at ca. 20 ЊC provides a simple, alternative route to
isothiazoles under mild conditions.
The assignment of structure 12 to the isothiazole product
was initially based on the very close similarity of its 13C NMR,
UV–VIS, MS and IR spectroscopic data with that of known
3-chloroisothiazole-4,5-dicarbonitrile 2.1 The structure was
confirmed by its smooth conversion (75%) into the known
3-morpholinoisothiazole-4,5-dicarbonitrile 141 by heating
under reflux with morpholine in toluene.
In conclusion we have shown that dihalomalononitriles 3
reacted with both 4,5-dichloro-1,2,3-dithiazolium chloride 4
(Appel salt) and 4-chloro-1,2,3-dithiazole-5-thione 9 to give
4-chloro-5H-1,2,3-dithiazole-5-ylidenemalononitrile
1
and
various dithiazolimines 5a–c. Convenient conditions have been
determined for the preparation of ylidene 1 in high yield start-
ing from the readily prepared monobromomalononitrile 3c. A
one-pot preparation of 3-bromoisothiazole-4,5-dicarbonitrile
12 from dibromomalononitrile 3a and dithiazolethione 9 in
refluxing toluene has been discovered and an investigation of
the reaction led to isothiazole formation under mild conditions
by treatment of ylidene 1 with anhydrous HBr.
Presumably ylidene 1, which is formed in the reaction
between dibromomalononitrile 3a and the thione 9, is con-
verted into isothiazole 12 and the reaction is assisted by toluene
in some manner (see Table 4). In the absence of dithiazo-
lethione 9 under analogous reaction conditions, (24 h at 110 ЊC)
dibromomalononitrile 3a did not react with toluene to give
benzyl bromide. Bromination of benzylic positions occurs
readily with dibromine and it is possible that a little dibromine
could have been formed either by disproportionation of the
postulated sulfur dibromide (Scheme 5) or from some decom-
position of dibromomalononitrile by dithiazolethione 9.
The formation of benzyl bromide will be accompanied by the
release of HBr which could react with either nitrile group of
ylidene 1 to generate an E/Z isomeric mixture of imidoyl brom-
ides 15. Closure of the imine N adjacent to the dithiazole ring
onto sulfur S-1 is known to lead to the formation of iso-
thiazoles.14 Ready rotation about the ethene bond in 15 will
result from the push–pull contribution from the dithiazole ring
sulfurs to the cyano and imine bonds.
Experimental
Solvents DCM, PhH, PhMe and PhCl were freshly distilled
from CaH2 under argon. Reactions were protected by CaCl2
drying tubes. Anhydrous sodium sulfate was used for drying
organic extracts, and all volatiles were removed under reduced
pressure. All reaction mixtures and column eluents were
monitored by TLC using commercial glass backed thin-layer
chromatography (TLC) plates (Merck Kieselgel 60 F254). The
plates were observed under UV light at 254 and 365 nm. The
technique of dry flash chromatography was used throughout
for all non-TLC scale chromatographic separations using
Merck Silica Gel 60 (less than 0.063 mm). Melting points were
determined using a Stuart Scientific SMP 1 apparatus. Solvents
used for recrystallisation are indicated after the melting point.
UV spectra were obtained using
a Shimadzu UV-1601
spectrometer and inflections are identified by the abbreviation
“inf”. IR spectra were recorded on a Jasco FT/IR-460 plus
spectrometer and strong, medium and weak peaks are repre-
1
sented by s, m and w respectively. H and 13C NMR spectra
were recorded on Bruker Avance 300 machine (at 300 and 75
MHz respecively). Deuteriated solvents were used for homo-
nuclear locks and the signals are referenced to the deuteriated
solvent peaks. Mass spectra were recorded on a VG Autospec
“Q” mass spectrometer. Microanalyses were performed at the
University of North London. Petrol refers to light petroleum,
bp 40–60 ЊC. Appel salt 4,16 dithiazolethione 9,16 bromo-
Treatment of ylidene 1 with anhydrous gaseous HBr for 5
min in either PhMe or PhCl at ca. 20 ЊC followed by continuous
stirring for 24 h gave bromoisothiazole 12 in 82 and 83% yields,
respectively. This supports the conversion of ylidene 1 into
bromoisothiazole 12 by HBr generated during the bromination
of the toluene methyl group, as proposed above. Treatment of
J. Chem. Soc., Perkin Trans. 1, 2002, 1236–1241
1239

View Article Online
malononitrile 3c,6 dibromomalononitrile 3a,17 and dichlorom-
alononitrile 3b18 were prepared according to literature
procedures.
Reaction of Appel salt 4 with bromomalononitrile 3c with base:
typical procedure
To a stirred suspension of 4,5-dichloro-1,2,3-dithiazolium
chloride 4 (104 mg, 0.50 mmol) in DCM (3 ml) at ca. 20 ЊC,
bromomalononitrile (145 mg, 1.00 mmol) was added. After 10
h pyridine (79 µl, 2 mmol) was added and the mixture was left
to stir for a further 2 h. TLC indicated only one main orange
product and chromatography (petrol–DCM 1 : 1) gave a trace
of an inseparable E/Z isomeric mixture of 1-(4-chloro-5H-1,2,3-
dithiazol-5-ylideneamino)-2-bromo-1-chloroethene-2-carbo-
nitriles 5b (3 mg, 2%) as orange–red crystals, mp 139–141 ЊC
identical with that described above. Further elution (DCM)
gave 2-(4-chloro-5H-1,2,3-dithiazol-5-ylidene)propanedinitrile
1 (73 mg, 73%) as orange crystals, mp 181–182 ЊC identical with
that reported above.
Reaction of Appel salt 4 with dibromomalononitrile 3a: typical
procedure
To a stirred suspension of 4,5-dichloro-1,2,3-dithiazolium
chloride 4 (104 mg, 0.50 mmol) in PhCl (3 ml) at ca. 20 ЊC,
dibromomalononitrile 3a (225 mg, 1.00 mmol) was added and
the mixture was heated to ca. 132 ЊC. After 24 h TLC indicated
two orange products and chromatography (petrol–DCM 1 : 1)
gave an inseparable mixture of the E/Z isomers of 1-(4-chloro-
5H-1,2,3-dithiazol-5-ylideneamino)-1,2-dibromoethene-2-carb-
onitrile 5a (minor) and 1-(4-chloro-5H-1,2,3-dithiazol-5-
ylideneamino)-2-bromo-1-chloroethene-2-carbonitrile
5b
(major) (21 mg, 13% based on 5b m/z 317) as orange–red
crystals, mp 142–145 ЊC (from cyclohexane) (Found: C, 18.6; N,
12.4. C5BrCl2N3S2 requires C, 18.9; N, 13.2 %) νmax(Nujol)/cmϪ1
2213s (CN), 2174m (CN), 1629w, 1541s, 1456s, 1309m, 1205s,
1152w, 1136m, 1118m, 1093s, 1060m, 883s, 844m, 814s, 803s,
770s, 699s, 660s; m/z (EI) 361 (Mϩ, 5%), 317 (Mϩ, 35%), 282
[Mϩ(361) Ϫ Br, & Mϩ(317) Ϫ Cl, 20], 256 [Mϩ(317) Ϫ S2, 5],
236 [Mϩ(361) Ϫ Br2, & Mϩ(317) Ϫ BrCl, 45], 201 (7), 175 (7),
163 (5), 140 (3), 125 (15), 117 (7), 108 (12), 102 (13), 93 (17), 70
(28), 64 (S2ϩ, 100). Further elution (DCM) gave 2-(4-chloro-
5H-1,2,3-dithiazol-5-ylidene)propanedinitrile 1 (56 mg, 56%)
as brown crystals, mp 181–182 ЊC (from 1,2-dichloroethane)
identical with that reported.1
Reaction of dithiazolethione 9 with halogenated malononitriles
3a–c: typical procedure
To a stirred suspension of 4-chloro-1,2,3-dithiazole-5-thione 9
(102 mg, 0.60 mmol) in PhH (3 ml) at ca. 20 ЊC, dibromo-
malononitrile (134 mg, 0.60 mmol) was added and the reaction
was heated to reflux. After 4 h TLC indicated two main orange
products and chromatography (petrol–DCM 1 : 1) gave an E/Z
isomeric mixture of 1-(4-chloro-5H-1,2,3-dithiazol-5-ylidene-
amino)-1,2-dibromoethene-2-carbonitriles 5a (41 mg, 19%) as
red crystals, mp 145–147 ЊC (from cyclohexane) (Found: C,
16.8; N, 11.6. C5Br2ClN3S2 requires C, 16.6; N, 11.6%);
λmax(DCM)/nm 226 (log ε 4.01), 248inf (3.93), 274 (3.95),
367inf (3.70), 431inf (4.27), 440 (4.28), 467inf (4.13), 499inf
(3.68); νmax(Nujol)/cmϪ1 2208s (CN), 1620w, 1544s, 1456s,
1307w, 1202s, 1135m, 1094m, 1078s, 1051m, 1017w, 878s, 812s,
727s, 693s, 657m, 586w; δC(75 MHz; CDCl3) 162.3 (C-5), 159.7
Reaction of Appel salt 4 with dichloromalononitrile 3b
Similar treatment of 4,5-dichloro-1,2,3-dithiazolium chloride
4 with dichloro-malononitrile 3b gave the E/Z isomeric
mixture of 1-(4-chloro-5H-1,2,3-dithiazol-5-ylideneamino)-1,2-
dichloroethene-2-carbonitrile 5c (23 mg, 17%) as orange–red
crystals, mp 144–154 ЊC (from cyclohexane); λmax(DCM)/nm
226 (log ε 3.89), 261inf (3.76), 270 (3.79), 415inf (4.13), 425
(4.15), 445inf (4.03), 473inf (3.66); νmax(Nujol)/cmϪ1 2216m
(CN), 2201m (CN), 1560s, 1530w, 1519m, 1214m, 1157m,
1091m, 892s, 863w, 811m, 785m, 682s; δC(75 MHz; CDCl3)
161.8 (C-5), 159.8 (C-5), 150.2 (C-4), 150.1 (C-4), 147.2
(C-5), 150.3 (C-4), 150.3 (C-4), 143.5 (C᎐CCN), 133.0
᎐
(C᎐CCN), 116.7 (CN), 114.8 (CN), 90.7 (᎐CCN), 82.4 (᎐CCN);
᎐
᎐
᎐
m/z (EI) 361 (Mϩ, 40%), 282 (Mϩ Ϫ Br, 100), 236 (Mϩ Ϫ Br2,
10), 201 (18), 175 (3), 163 (7), 140 (7), 125 (12), 108 (12), 102
(30), 93 (23), 70 (45), 64 (S2ϩ, 95); (Found: Mϩ, 358.7586.
C5Br2ClN3S2 requires M, 358.7589). Further elution (DCM)
gave 2-(4-chloro-5H-1,2,3-dithiazol-5-ylidene)propanedinitrile
1 (92 mg, 76%) as brown crystals, mp 181–182 ЊC identical with
that reported above.
(C᎐CCN), 142.1 (C᎐CCN), 115.1 (CN), 113.7 (CN), 100.3
᎐
᎐
(᎐CCN), 96.0 (᎐CCN), identical with that reported.2 Further
᎐
᎐
5-Bromosulfanyl-4-chloro-1,2,3-dithiazolium bromide 11
elution (DCM) gave 2-(4-chloro-5H-1,2,3-dithiazol-5-ylidene)-
propanedinitrile 1 (15 mg, 15%) as orange–brown crystals,
mp 181–182 ЊC identical with that reported above.
To a stirred solution of 4-chloro-1,2,3-dithiazole-5-thione 9
(500 mg, 2.95 mmol) in DCM (25 ml) at ca. 20 ЊC, was added
a solution of bromine (470 mg, 2.95 mmol) in DCM (5 ml). An
immediate yellow–orange precipitate was formed and the
mixture was left to stir for 10 min. The precipitate was filtered
off, washed (DCM) and dried under vacuum at ca. 20 ЊC to
afford the title compound 11 (923 mg, 95%) as a yellow–orange
powder, mp 162–164 ЊC (decomp.) (Found: C, 7.3; N, 3.9.
C2Br2ClNS3 requires C, 7.3; N, 4.2%); λmax(CH3CN)/nm 235
(log ε 3.78), 269 (4.30), 317inf (3.51), 427 (3.79); νmax(Nujol)/
cmϪ1 1632w, 1426s, 1411m, 1267m, 1241s, 1210s, 1089s, 1052m,
904s, 890w, 818s, 722w, 620m, 594m; m/z (CI/CH4) 250 [(MH Ϫ
Br)ϩ, 1%], 170 [(MH Ϫ Br)ϩ, 100], 134 (5), 118 (5), 108 (15), 93
(12), 81 (HBr, 15), 76 (7), 70 (10), 64 (9).
Reaction of Appel salt 4 with bromomalononitrile 3c (without
base)
Similar treatment of 4,5-dichloro-1,2,3-dithiazolium chloride
4 with bromomalononitrile 3c gave an E/Z isomeric mixture
of 1-(4-chloro-5H-1,2,3-dithiazol-5-ylideneamino)-2-bromo-1-
chloroethene-2-carbonitrile 5b (54 mg, 34%) as orange–red
crystals, mp 139–141 ЊC (from cyclohexane) (Found: C, 19.0; N,
13.2. C5BrCl2N3S2 requires C, 18.9; N, 13.2%); λmax(DCM)/nm
227 (log ε 3.84), 272 (3.92), 356inf (3.50), 434 (4.26), 467inf
(4.07), 493inf (3.67); νmax(Nujol)/cmϪ1 2213s (CN), 2174m
(CN), 1629w, 1541s, 1456s, 1309m, 1205s, 1152w, 1136m,
1118m, 1093s, 1060m, 883s, 844m, 814s, 803s, 770s, 699s, 660s;
δC(75 MHz; CDCl3) 161.9 (C-5), 159.4 (C-5), 150.3 (C-4), 150.2
Reaction of 5-bromosulfanyl-4-chloro-1,2,3-dithiazolium
bromide 11 with malononitrile, bromomalononitrile 3c and
dibromomalononitrile 3a: typical procedure
(C-4), 149.0 (C᎐CCN), 141.9, (C᎐CCN), 115.7 (CN), 114.5
᎐
᎐
(CN), 87.5 (᎐CCN), 81.8 (᎐CCN); m/z (EI) 315 (Mϩ, 25%), 282
᎐
᎐
(Mϩ Ϫ Cl, 10), 256 (Mϩ Ϫ S2, 5), 236 (Mϩ Ϫ BrCl, 53), 201 (5),
175 (10), 163 (5), 125 (15), 108 (8), 102 (13), 93 (17), 70 (25),
64 (S2ϩ, 100); (Found: Mϩ, 314.8082. C5BrCl2N3S2 requires
M, 314.8094). Further elution (DCM) gave 2-(4-chloro-5H-
1,2,3-dithiazol-5-ylidene)propanedinitrile 1 (51 mg, 51%) as
orange crystals, mp 181–182 ЊC identical with that reported
above.
To a stirred suspension of 5-bromosulfanyl-4-chloro-1,2,3-
dithiazolium bromide 11 (100 mg, 0.30 mmol) in PhH (3 ml)
at ca. 20 ЊC, dibromomalononitrile (134 mg, 0.60 mmol)
was added and the reaction mixture was heated to reflux.
After 24 h TLC indicated two main orange products and
chromatography (petrol–DCM 1 : 1) gave an E/Z isomeric mix-
ture of 1-(4-chloro-5H-1,2,3-dithiazol-5-ylideneamino)-1,2-
1240
J. Chem. Soc., Perkin Trans. 1, 2002, 1236–1241

View Article Online
dibromoethene-2-carbonitriles 5a (24 mg, 22%) identical with
that described above. Further elution (DCM) gave 2-(4-chloro-
5H-1,2,3-dithiazol-5-ylidene)propanedinitrile 1 (30 mg, 49%) as
orange–brown crystals, mp 181–182 ЊC, identical with that
reported above.
3-Morpholinoisothiazole-4,5-dicarbonitrile 14
To a stirred solution of 3-bromoisothiazole-4,5-dicarbonitrile
12 (100 mg, 0.47 mmol) in toluene (4 ml) at ca. 20 ЊC, was added
morpholine (327 µl, 3.76 mmol). The reaction mixture was
heated to ca. 110 ЊC for 2 h until no starting material 1
remained (TLC). The mixture was allowed to cool to ca. 20 ЊC
and chromatography (DCM) gave the title compound 14 (78
mg, 75%) as yellow prisms, mp 185–187 ЊC (cyclohexane);
δH(300 MHz; CDCl3) 3.82 (4H, t, CH2O) and 3.67 (4H, t,
CH2N); δC(75 MHz; CDCl3) 165.3, 142.6, 111.4, 108.3, 102.9,
66.0 and 47.9; identical with an authentic sample.1
3-Bromoisothiazole-4,5-dicarbonitrile 12
(i) From dithiazolethione 9 and dibromomalononitrile 3a in
toluene. To a stirred solution of dibromomalononitrile 3a (220
mg, 2.40 mmol) in PhMe (3 ml) at ca. 20 ЊC, 4-chloro-1,2,3-
dithiazole-5-thione 9 (102 mg, 0.60 mmol) was added and
the mixture was heated to reflux. After 2 h TLC indicated the
presence of sulfur, two colourless products, trace amounts of
dithiazole ylidene 1, dithiazolimines 5, and unidentified prod-
ucts. The reaction mixture was cooled to ca. 20 ЊC and chrom-
atography (petrol–DCM 4 : 1) gave benzyl bromide 13 (349 mg,
85%) as a colourless oil, (lachrymator) identical with an
authentic sample. Further elution (petrol–DCM 4 : 1) gave the
title compound 12 (76 mg, 59%) as colourless plates, mp 140–
141 ЊC (from cyclohexane) (Found: C, 28.1; N, 19.6. C5BrN3S
requires C, 28.0; N, 19.6%); λmax(DCM)/nm 247 (log ε 3.81),
256inf (3.73), 295 (3.86), 298inf (3.85); νmax(Nujol)/cmϪ1 2243s
(CN), 2236s (CN), 1641w, 1592w, 1502s, 1360m, 1342s, 1257w,
1186s, 1166s, 1136m, 1006s, 977m, 843s, 800s, 627s, 567s; δC(75
MHz; CDCl3) 142.4, 139.8, 119.3 (CN), 109.0 (CN), 106.7; m/z
(EI) 213 (Mϩ, 100%), 137 (CBrNSϩ, 43), 134 (MHϩ Ϫ Br, 8),
108 (27), 82 (10), 76 (10), 70 (35), 58 (15) (Found: Mϩ, 212.8996
C5BrN3S requires M, 212.9001).
Acknowledgements
We thank the following organisations in Cyprus for generous
donations of chemicals and glassware: the State General
Laboratory, the Agricultural Research Institute and the
Ministry of Agriculture. Furthermore we thank the A. G.
Leventis Foundation for helping to establish the NMR facility
in the University of Cyprus, MDL Information Sytems (UK)
Ltd. for financial support and the Wolfson Foundation
for establishing the Wolfson Centre for Organic Chemistry in
Medical Science at Imperial College.
References
1 K. Emayan, R. F. English, P. A. Koutentis and C. W. Rees, J. Chem.
Soc., Perkin Trans. 1, 1997, 3345.
2 P. A. Koutentis and C. W. Rees, J. Chem. Soc., Perkin Trans. 1, 1998,
2505.
3 L. Assmann, Y. Kitagawa, K. Ishikawa, D. Yamazaki, H. Sawada,
Y. Araki, H. Sakuma, T. Kinbara and K. Imanishi, WO 00/29398,
(2000).
4 W. J. Linn, J. Am. Chem. Soc., 1965, 87, 3665.
5 T. Hata, Bull. Soc. Chem. Jpn., 1964, 37, 547.
6 J. P. Ferris and L. E. Orgel, J. Am. Chem. Soc., 1965, 30, 2365.
7 cf. D. Lloyd and R. W. Millar, Tetrahedron, 1980, 36, 2675.
8 M. Sekiya, K. Ito and K. Suzuki, Tetrahedron, 1975, 31, 231.
9 S. Trofimenko, E. L. Little Jr. and H. F. Mower, J. Am. Chem. Soc.,
1962, 27, 433.
(ii) From ylidene 1 with anhydrous hydrogen bromide. A stirred
solution of 2-(4-chloro-5H-1,2,3-dithiazol-5-ylidene)propane-
dinitrile 1 (100 mg, 0.50 mmol) in PhMe (3 ml) at ca. 20 ЊC, was
vigorously purged for 5 min with anhydrous gaseous HBr. TLC
indicated the presence of sulfur, bromoisothiazoledicarbonitrile
12, and unreacted ylidene. After 24 h no starting material 1 was
observed (by TLC) and chromatography gave the title com-
pound 12 (89 mg, 83%) as colourless plates, mp 140–141 ЊC
identical with that described above.
10 J. L. Adelfang, J. Org. Chem., 1966, 31, 2388.
11 A. Corsaro, G. Perrini, U. Chiacchio, G. Purrello and F. Guerrera,
Phosphorus Sulfur Silicon, 1991, 63, 103.
12 N. Bricklebank, P. J. Skabara, D. E. Hibbs, M. B. Hursthouse and
K. M. Abdul Malik, J. Chem. Soc., Dalton Trans., 1999, 3007.
13 P. A. Koutentis, C. W. Rees, A. J. P. White and D. J. Williams,
J. Chem. Soc., Perkin Trans. 1, in preparation.
14 D. Clarke, K. Emayan and C. W. Rees, J. Chem. Soc., Perkin Trans.
1, 1998, 77.
15 E. Allenstein and P. Quis, Chem. Ber., 1964, 97, 1857.
16 R. Appel, H. Janssen, M. Siray and F. Knoch, Chem. Ber., 1985, 118,
1632.
(iii) From ylidene 1 with dibromine in toluene. To a stirred
solution of 2-(4-chloro-5H-1,2,3-dithiazol-5-ylidene)propane-
dinitrile 1 (100 mg, 0.50 mmol) in PhMe (3 ml) at ca. 20 ЊC,
dibromine (80 mg, 0.50 mmol) was added. After 24 h TLC
indicated the presence of sulfur, benzyl bromide, benzyl
alcohol, benzaldehyde, bromoisothiazole-dicarbonitrile 12 and
unreacted ylidene 1. More dibromine (80 mg, 0.50 mmol) was
added and after an additional 24 h of stirring, chromatography
of the mixture gave the title compound 12 (70 mg, 65%) as
colourless plates, mp 140–141 ЊC identical with that described
above.
17 E. Otto and B. Löpmann, Chem. Ber., 1922, 55, 1259.
18 W. R. Carpenter and P. Armstrong, J. Org. Chem., 1964, 29, 2772.
J. Chem. Soc., Perkin Trans. 1, 2002, 1236–1241
1241