Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
1,2,3-Dithiazole chemistry in heterocyclic synthesis Irene C. Christoforou, Panayiotis A. Koutentis,* and Sophia S. Michaelidou Department of Chemistry, University of Cyprus, P.O. Box 20537, 1678 Nicosia, Cyprus E-mail:
[email protected]
Abstract The chemistry of various 5H-1,2,3-dithiazoles is investigated with emphasis on assisted ring opening and ring closure reactions leading to new heterocycles. Thus on treatment with catalytic tetraalkylammonium iodide N-(2-chloropyrid-3-yl)- and N-(4-chloropyrid-3-yl)-4-chloro-1,2,3dithiazol-5H-imines 19 and 20 give thiazolo[5,4-b]pyridine-2-carbonitrile 16 and thiazolo[4,5-c]pyridine-2-carbonitrile 17 respectively. Similar treatment of bisdithiazoles 29 and 30 afford high yielding routes to 1,3,4-thiadiazole-2,5-dicarbonitrile 31 and thiazole-2,4,5tricarbonitrile 32 respectively. N-(Pyrid-3-yl)-4-chloro-1,2,3-dithiazol-5H-imine 36 reacts with secondary alkylamines to give as main product pyrido[2,3-d]pyrimidines 37 and several minor byproducts including a deep green quinoidal 2,2’-bithiazole 40. Dithiazolylidenacetonitriles 43 react with either anhydrous HBr or tetraalkylammonium chloride to afford a series of 3-halo-4substituted-isothiazole-5-carbonitriles 45 and 52 respectively. The reactions of dithiazoles 43 with tetraalkylammonium chloride are complicated owing to the formation of isothiazolopentathiepin-8-carbonitrile 53, isothiazolodithiin-4,5,7-tricarbonitrile 54, tetracyanothiophene 56 and an unidentified compound 55 whose possible structures are proposed. The mechanistic rationales for the formation of the identified products are proposed. Keywords: Heterocycle, dithiazole, isothiazole, thiazole, acetylene
Contents 1. Synthesis of thiazolopyridine-2-carbonitriles 2. Synthesis of percyano thiazole and 1,3,4-thiadiazole 3. Synthesis of fully substituted pyrido[2,3-d]pyrimidines 37: Unexpected byproducts 4. Chemistry of dithiazolylidenacetonitriles 43: Formation of isothiazoles
ISSN 1424-6376
Page 207
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
Introduction 1,2,3-Dithiazole is one of the four possible dithiazole systems all of which are reported in the literature as cations. The salts are planar, 6π and therefore aromatic and can be adequately represented by three canonical forms where the charge (identifying the three most electrophilic sites) is distributed at C-5 or on either of the ring sulfur atoms (Scheme 1). N +S
S
N
1,2,3-
+S
+S
S
1,2,4-
N
N S
+S
S
1,3,2-
1,3,4+
S
S +
+S
N
S
N
S
S
N
Planar 6π Aromatic
1,2,3-
Scheme 1 1,2,3-Dithiazolium salts are commonly prepared from substituted acetonitrile derivatives to afford the 5-substituted-4-chloro-1,2,3-dithiazolium chlorides 1.1,2 One example has appeared in the literature where phenylacetoxime was treated with disulfur dichloride to afford the 5-chloro4-phenyl-1,2,3-dithiazolium chloride 2 and development of this route could provide access to 5chloro-4-substituted-1,2,3-dithiazolium salts.3,4 R R
N
CH2Cl2
+ Cl S S Cl
20 oC, 12 h
Cl
+S N _ S Cl 1 Cl
H3C
Ph NOH
+ Cl S S Cl
Ph
CH2Cl2 o
0 C, C6H5N, 30 %
+S N _ S Cl 2
Our interest focuses on the chemistry of the readily available 4,5-dichloro-1,2,3dithiazolium chloride 3 prepared from chloroacetonitrile and disulfur dichloride.5 Dithiazolium chloride 3 on treatment with nucleophilic species affords neutral 5H-dithiazoles 4 (eg. treatment of dithiazolium 3 with H2O, aniline, diethyl malonate or H2S provides 1,2,3-dithiazolone 5,
ISSN 1424-6376
Page 208
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
dithiazolimine 6, dithiazolylidene 7 and dithiazolethione 8 respectively in good yields (Scheme 2).5 Cl
Cl
X H2X
+
+S N _ S Cl
Cl
Base S
3
S
N
4 EtO EtO
O
Cl
N
Cl
O Cl
S
N
S
Cl
O S
S
N
S
S
N
S
6
5
S 7
S
N
8
Scheme 2 1,2,3-Dithiazoles have uses in both biological and material sciences: N-aryldithiazolimines show interesting antifungal,6 antibacterial,7 and herbicidal8 activities. A search for organic conductors based on neutral radicals has led to the preparation of two 1,2,3dithiazolyl radicals 91 and 109 and also a tetrathiadiazafulvalene analogue 1110 has been prepared and studied. F
F Cl
F
F Cl F S
S
N
.
N
Cl
S Cl
Cl S
S 10
9
N
.
S S N
N S S
Cl 11
Our interest in 1,2,3-dithiazole chemistry revolves around the construction of dithiazole systems that can be converted into new heterocyclic systems via ring transformation. The majority of these ring transformations involve the initial preparation of a neutral dithiazole which supports a potentially nucleophilic side chain or substituent capable of attacking the electrophilic dithiazole at either S-1 (Path A) or at C-5 (Path B) with subsequent ring opening. Dithiazoles, however, can also be ring opened with the use of soft nucleophiles to afford the disulfide intermediate 12 (Path C) (Scheme 3). This disulfide can be a source of both electrophilic and nucleophilic sulfur.
ISSN 1424-6376
Page 209
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
X Y S
E
Cl
S
N
X
Cl
S
N
S
X
X
Path A
Cl
Path B
X
Common
Y
CN
Common
CN
Y S
Y S
S
N
X X
CN
Path C
E
CN
S
S Y
Y
E S Y
E = Electrophilic trap
12
Scheme 3 Various dithiazolylidenes have been prepared specifically with these mechanistic possibilities in mind in order to broaden the capability of dithiazolium chloride 3 as a useful synthetic tool. Emphasis will be made on the in situ generation of disulfides 12 endowed with electrophilic traps for nucleophilic sulfur (Path C).
1. Synthesis of thiazolopyridine-2-carbonitriles Benzo[d]thiazole-2-carbonitrile 13 can be prepared readily from the thermolysis of N-phenyl1,2,3-dithiazol-5H-imine 6 in particular where the aryl substituent is electron rich. N
Cl
Thermolysis
S
S
N
N S8
+
CN
+
Cl
S
N
13
6
N
N
Cl
S S
S
N
N
Cl
S
N
S
H
14
Cl 13
S N
+ N
Cl N
S
CN
_ 14
S
Scheme 4
ISSN 1424-6376
Page 210
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
Where the aryl group is electron deficient the major product is the imidoyl chloride carbonitrile 14.11 The mechanism that has been proposed by Rees is shown in Scheme 4. To our knowledge no examples of the thermolysis of N-heteroaryl-1,2,3-dithiazol-5H-imines have appeared in the academic or patent literature. Thermolysis of N-(pyrid-3-yl)-4-chloro-1,2,3-dithiazol-5H-imine 15 gave thiazolo[5,4-b]pyridine-2-carbonitrile 16 and thiazolo[4,5-c]pyridine-2-carbonitrile 17 in low yields as might be expected based on the mechanism to give benzothiazole 13. Repeating the reaction in the presence of a soft nucleophile benzyltriethylammonium iodide at 132 oC in chlorobenzene gave improved but still moderate yields of both isomers. Under these conditions the dithiazole ring is anticipated to suffer an assisted nucleophilic ring opening-ring closure (ANRORC)12 like mechanism involving the intermediate disulfide 18. The nucleophilic sulfur that is generated (S-1) is trapped at the electrophilic pyridyl C-2 and C-4 positions and a subsequent oxidation restores the aromaticity of the thiazolopyridine systems. N
Cl N
200 oC, Argon
N S
N
S
S8
10 min
CN
+ N
15
+
BnEt3NI (1 equiv.)
S8
PhCl, 132 oC, 40 min
+
CN
S
S
16 (5-11 %)
15
N
N
17 (9-15 %)
+
16 (25 %)
17 (29 %)
Cl N
N
N
CN S 18
S I
Cl
N I
_
N
Cl
N Cl S
S 19
N
S
S
N
20
Introducing a chlorine substituent at either C-2 or at C-4 on the pyridyl ring was expected to assist in directing the ring closure and furthermore adjusts the oxidation level of the starting systems thus avoiding the need for oxidative rearomatisation. The 2-chloro and 4-chloroderivatives 19 and 20 were readily prepared starting from the corresponding aminochloropyridines and dithiazolium chloride 3. Gratifyingly treatment with catalytic iodide (5 mol %) gave the expected single isomers in near quantitative yields.
ISSN 1424-6376
Page 211
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
N
Cl N
BnEt3NI (5 mol %)
N Cl S
S
o
N
Argon, PhCl, 132 C, 24 h
S8
CN
+ N
19
S
16 (98 %)
Cl N
Cl BnEt3NI (5 mol %)
N S
S
o
N
Argon, PhCl, 132 C, 20 h
S8
+
N
N
CN S
20
17 (99 %)
Thermolysis of either the 2-chloro and 4-chloro- derivatives 19 and 20 did not yield regiospecific ring closures but mixtures of products were isolated adding further support to the ANRORC type mechanism proposed above. Cl 19
200 C, 10 min Argon
CN
N
o
S8
+ N
Cl
Cl
+
16 (8-10 %)
+
N
N CN S 13-15 %
74-77 %
We attempted to extrapolate the above success to the preparation of 3,1-benzothiazin-4ones starting from the readily available methyl ester 21 where the electrophilic trap was now the carboxylate group. Initial thermolysis of the ester 21 gave some of the desired benzothiazinone 22 together with the benzothiazolecarbonitrile 23. O 200 oC, 10 min
N S
N
Cl
S 21
N
S
Argon
CO2Me
CO2Me + CN
22 (25 %)
CN S 23 (35 %)
N S BnEt3NI (5 mol %) PhCl, 132 oC, 20 min, Ar
CN NH CO2Me
24 (99 %)
ISSN 1424-6376
Page 212
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
On treatment with various soft nucleophilies the major product was however the Narylcyanothioformamide 24. This could be obtained in near quantitative yield with the use of catalytic benzyltriethylammonium iodide (5 mol %) providing a good route to this useful intermediate. + Ph3P S PPh 3 _ O O
CO2H N S
Cl
S
N
PPh3 (2 equiv.) CH2Cl2, 20 oC, 10 min
22 (96 %)
S N
CN
26
25
Interestingly Besson, Rees et al. reported a quantitative conversion of the iminocarboxylic acid 25 into benzothiazinone 22 on treatment with triphenylphosphine and proposed that the phosphonium salt byproducts help activate the carboxylic acid towards attack (cf. intermediate 26).13 In our case the methyl ester 21 on treatment with triphenylphosphine gave only a moderate yield of the cyanothioformamide 24.
2. Synthesis of percyano thiazole and 1,3,4-thiadiazole The 1,2,3-dithiazole ring can act as both a source of nucleophilic sulfur at S-1 and as electrophilic trap at the ring carbon C-5. On treatment with soft nucleophiles bisdithiazole systems such as 27 can therefore be considered as possible precursors to heterocyclic systems of type 28 (Scheme 5). Two bisdithiazoles of this type compound 2914 and compound 3015 have been reported in the literature and each could provide rapid routes to the corresponding percyano-1,3,4-thiadiazole 31 and percyanothiazole 32 respectively.
ISSN 1424-6376
Page 213
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
Elec C-5 S S N Cl N
X Y
S
Ring Opening
Cl
S Nuc
Nuc S
X Y
Cl
S
N
Ring Closure
NC S
S
_
S
X Y NC
CN
S
Nuc 28
27 S
S
N
N N NC
Cl
CN
S
N N
N 31
S
NC NC
Cl
S
N
N 3
Cl
40 %
CN
CN
S
N
S
H2N NH2
40 %
S 29 S
N
3
Cl
S
32
NC
NH2
NC
NH2
30
Scheme 5 Treatment of either bisdithiazole 29 or 30 with soft nucleophiles such as chloride, bromide or iodide gave the expected percyano-1,3,4-thiadiazole 31 and thiazole 32 systems respectively. These percyano heterocycles suffered hydrolysis during chromatographic isolation to afford a moderate quantity of the corresponding carboxamides 33 and 34 respectively. 29
Reagent
S8
N N
+ NC
S
N N
+ CN
NC
31
S
CONH2
33
Table 1. Transformation of bisdithiazole 29 into 1,3,4-thiadiazoles 31 and 33 Reagent BnEt3NI (1 eqiuv.) BnEt3NI (0.25 eqiuv.) Ph3P (5 eqiuv.) Ph3P – polymer bound (6 mol equiv.)
ISSN 1424-6376
Conditions
Yields (%)
Ar, PhCl, 132 °C, 40 min Ar, PhCl, 132 °C, 6 h CH2Cl2, 20 °C, 5 min CH2Cl2, 20 °C, 10 min
Page 214
31 79 76 Trace 69
33 19 18 0 0
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
NC
NC 30
Reagent
S8
N
+ NC
S
N
+ CN
NC
32
S
CONH2
34
Table 2. Transformation of bisdithiazole 30 into thiadiazoles 32 and 34 Reagent BnEt3NI (1 eqiuv.) BnEt3NBr (0.5 eqiuv.) Ph3P (5 eqiuv.) Ph3P – polymer bound (5 mol equiv.)
Conditions
Yields (%)
Ar, PhCl, 132 °C, 15 min Ar, PhCl, 132 °C, 15 min CH2Cl2, 20 °C, 10 min CH2Cl2, 20 °C, 4h
32 70 68 Trace 76
34 15 20 0 0
Chromatography could be avoided by using polymer bound triphenylphosphine as the soft nucleophile since the polymer resin could be separated by filtration to afford a clean solution of the desired percyano heterocycle. Recrystallisation of this gave pure product free of any impurities or carboxamide. Surprisingly free triphenylphosphine gave only traces of product.
3. Synthesis of fully substituted pyrido[2,3-d]pyrimidines 37: unexpected byproducts The dithiazolimine 36 prepared from the available fully substituted 2-aminopyridine 35 gave on treatment with cyclic secondary amines the expected fully substituted pyrido[2,3-d]pyrimidines 37 in moderate to good yields together with minor quantities of the guanidine 38 and the 4-aminosubstituted dithiazolimine 39 (Scheme 6).
ISSN 1424-6376
Page 215
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
Ph
Ph NC EtO
CN N
NC
3
EtO
NH2
35
Ph
EtO
N
N
NC
Ph CN
+ CN
NC NR2
EtO
N
37 - 40
Cl
Ph
N N
N
R2NH S S N
36 (88 %)
NR2
NC
CN
N
CN
+ EtO
N
N
NR2 37 (50-80 %)
Deep Green Unknown
+
S S N
39 (Trace) R2N
38 (5-20 %)
40 (12-14 %)
Scheme 6 We have shown that the guanidine 38 is a product of the reaction of pyrido[2,3d]pyrimidine 37 with excess amine presumably by nucleophilic attack by R2NH at the electrophilic pyrimidine C-2, with pyrimidine ring opening and release of R2NH from C-4. Surprisingly when acyclic secondary amines such as diethylamine are used in the reaction of dithiazolimine 36 a new previously unobserved green colored byproduct 40 is observed in addition to the byproducts 37-39. Due to very low solubility NMR or MS spectroscopic data were not obtainable and the structure was solved by single crystal X-ray crystallography to afford the unexpected quinoidal 2,2’-bithiazole 40. NC
OEt
Ph
NEt2
N NC
N N
S
S
N
CN
N N
Et2N
Ph EtO
CN
Figure 1. X-Ray structure of 2,2’-bithiazole 40. This is the first time such a product type has been observed in 1,2,3-dithiazole chemistry, which raises new questions about the possibilities of the use of 1,2,3-dithiazoles in synthesis. A possible precursor to this compound is the 4-aminosubstituted dithiazolimine 39 which could
ISSN 1424-6376
Page 216
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
have been cleaved to afford the ring opened intermediate 41. These intermediates have been prepared starting from the 4-aminosubstituted dithiazolimines by Kim using alkali bases in alcohol.16 The two central carbons forming C-2 of the thiazole rings are presumed to be derived from the solvent CH2Cl2. However, attempts to improve the yield of the 2,2’-bithiazole 40 by replacing the solvent with different sources of carbon such at CHCl3, CCl4, CH2ClBr, CH2ClI, CH2Br2, and CHBr3 gave very similar reaction mixtures but surprisingly no trace of the green product, though formation of the green product 40 in the presence of CH2Cl2 was reproducible. OEt NC
OEt S S N
N
Ph
NC
N
N
Ph NR2
CN
CN
39
S N H
40
NH NR2
41
The formation of the pyrido[2,3-d]pyrimidine 37 from dithiazolimine 36 is in itself mechanistically interesting as two possible pathways can be envisaged. In the first (Path A) the nitrile group neighbouring the dithiazolimine is initially attacked by the amine and cyclises onto the dithiazole C-5 carbon which instigates opening of the dithiazole ring. A second possibility (Path B) involves attack by the amine at the dithiazole ring sulfur S-2 to afford the disulfide intermediate 42 which then suffers ring closure to afford either the 4-aminodithiazolimine 39 or the pyrido[2,3-d]pyrimidine 37 depending on which nitrile is attacked preferably by the incoming amine (Scheme 7). Further studies are needed to determine which is the predominant mechanism. HNR2 Ph NC EtO
Ph C N
N
Path A S S N
N
EtO
EtO
36
N
CN NR2
CN
EtO
N
N NR2 38
HNR2 HNR2
CN N
N
NC
37
Ph NC
Ph HNR2
N
Cl
36
NR2
NC
N
S S N Cl
Ph
Ph C N
NC
NR2
NC
CN
S S Path B EtO
N
EtO
N C N
N
N
S S N
39 R2N
42 HNR2
Scheme 7
ISSN 1424-6376
Page 217
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
4. Chemistry of dithiazolylidenacetonitriles 43: Formation of isothiazoles An example where more extensive studies have provided insight into the ring opening mechanisms of dithiazoles is the chemistry of dithiazolylidenacetonitriles 43.17-19 Treatment of dithiazoles 43 with anhydrous HBr afforded the 3-bromoisothiazoles 45 in moderate to good yields. However, with the dithiazolylidenemalononitrile 43 (X = CN) the analogous treatment with anhydrous HCl gives only a trace of the expected isothiazole 44.19 Here we argue that the reaction mechanism proceeds by formation of the imidoyl bromide 46 which then cyclises onto the dithiazole ring sulfur S-1. Anhydrous HCl being a weaker acid than HBr is unable to drive this transformation. X Cl
NC
CN
Cl
Br
HCl (g), PhMe N
CN
S
20 oC, 5 min.
44
Br
S
S
20 oC, 5 min.
N
N
43 X = H, Cl, Br, CN
_
CN Cl
+ HN S
S
N
43 (X = CN)
Y
HBr (g), PhH S
CN
45 X = Y = CN, 83 % X = Y = H, 37 % X = Cl, Y = Br & Cl X = Y = Br, 45 %
CN
Br
Cl
Br
CN +
HN S
S 46
N
N
S
Sulfur
CN
45
Earlier work on dithiazolylidenemalononitrile 43 (X = CN) showed clearly that catalytic chloride (BnEt3NCl 10 mol %) was sufficient to convert the dithiazole 43 (X = CN) into 3-chloroisothiazoledicarbonitrile 44 almost quantitatively. Treatment of dithiazolylidenemalononitrile 43 (X = CN) with secondary amines such as morpholine also afforded the 3-morpholinoisothiazoledicarbonitrile 47 although in moderate yield (60%).17
ISSN 1424-6376
Page 218
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
CN Cl
CN
Cl
NC
R2NH, Cu(I)Cl (1 equiv.)
BnEt3NCl (10 %) N
PhH, 80 oC, 2 d
CN
S
S
PhH, 80 oC, 12 h
N
S
R2N N
CN CN
S
R2NH = Morpholine 43 (X = CN)
44 (98 %)
CN NC S
Cl
Cl
Cl
S
47 (60 %)
CN
_ C N
CN
44
S
N
S Cl
_ 48
Scheme 8 An alternative mechanism was proposed involving the disulfide 48. Nucleophilic attack on the ring sulfur at S-2 generates the disulfide 48 which can then recyclise to afford the isothiazole 44 (Scheme 8). NC CN
CN NC
NC N
PPh3
NC
Cl R2NH (10 equiv.), 1-2 h
PPh3 (5 equiv.) o
PhMe, 110 C, 3 d
N
S
S
N
o
PhMe, 100-110 C
R2N N NC
NR2
R2NH = Morpholine
PPh3 43 (X = CN)
49 (24 %)
CN NC
Cl S
S
N
H+ CN NC CN S S HNR2 R2N
S 50 (30 %) R2NH
R2NH R2N
N
H NC
N N
S
H+
H NC
NH S
R2NH
Scheme 9 During the investigation of this reaction mechanism two unexpected 3H-pyrroles were isolated.18 The first, a deep blue colored 3H-pyrrole 49 whose structure was determined by single crystal X-ray crystallography, was from the reaction of triphenylphosphine with
ISSN 1424-6376
Page 219
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
dithiazolylidenemalononitrile 43 (X = CN). The second, on treatment of the dithiazole 43 (X = CN) with excess morpholine, was the orange colored 3H-pyrrole 50. In light of difficulties in displacing the 4-chloro substituent of neutral 5H-1,2,3-dithiazoles a rational mechanism for the formation of these 3H-pyrroles 49 and 50 required the involvement of the ring opened disulfide [cf. the proposed mechanism of the bismorpholino-3H-pyrrole 50 (Scheme 9)]. As such the dithiazolylidenemalononitrile 43 (X = CN) can ring open to afford an intermediate disulfide 51 that could afford isothiazoles, 3H-pyrroles and even (although this has not been observed yet) 1,2,3-dithiazole ring systems depending on the cyclisation modes of the tricyanovinyl group (Scheme 10). CN
CN Cl
NC
NC CN
S Nuc
Nuc
S
N
C N
S Nuc
_
CN
_
_ Nuc
C N
CN
NC S Nuc
gives isothiazoles
_ Nuc
NC
CN S
51
S
C S
N
S Nuc gives 3H-pyrroles
C S
N
S Nuc gives 1,2,3-dithiazole (not observed)
Scheme 10 In order to investigate the reaction further, a series of substituted dithiazolylidenacetonitriles was prepared and treated with benzyltriethylammonium chloride. Unfortunately for the halo and non-substituted acetonitrile derivatives 43 (X = H, Cl, or Br) the geometry of the attached nitrile group has not been determined (ie. cis or trans with respect to the dithiazole ring sulfur) but in each case only one isomer was observed by NMR and by chromatography. Owing to a strong non-bonding interaction between the ester carbonyl group with the dithiazole ring sulfur S-1 the ethyl carboxylate 43 (X = CO2Et) is known to have a trans geometry for the nitrile group and the ring sulfur,20 however, this system was unreactive to halide.
ISSN 1424-6376
Page 220
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
EtO
CN
EtO Cl
_
O S
S
CN Cl
O +S
N
N
S
43 (X = CO2Et)
The acetonitrile derivatives 43 (X = H, Cl, or Br) gave surprisingly complex reaction mixtures. At best 3,4-dichloroisothiazolecarbonitrile 52 (Y = Cl) could be obtained in a respectable yield (63 %), but it was clear that halide exchange was proceeding from the bromoacetonitrile derivative. X NC
Cl
Cl
Y
BnEt3NCl (10 %) S
S
PhH, 80 oC
N
N
NC
NC
S
S S
CN
S
N
S
CN
52
43 X = H, Cl, Br, CN, CO2Et S S
S
X = CO2Et, NO REACTION X = Y = CN, 98 %, (48 h) X = Y = H, COMPLEX X = Cl, Y = Cl, 63 % (7 h) COMPLEX X = Br, Y = Br & Cl, COMPLEX
NC
53
NC
S
S
N
C8N4S3 Purple
54
CN
NC
55
CN
S 56
Analysis of the complex mixtures from the reaction of dithiazoles 43 (X = H, Cl or Br) and tetraalkylammonium chloride revealed the presence of several unexpected compounds, the isothiazolopentathiepin-8-carbonitrile 53, the isothiazolodithiin-4,5,7-tricarbonitrile 54, an unidentified purple compound 55 and tetracyanothiophene 56. The isothiazolopentathiepin 53 is known to react with dicyanoacetylene to give dithiin 54 and traces of tetracyanothiophene 56.21 These unusual products 53 – 56 are formally composed of one or two units of dicyanoacetylene and sulfur. As such the possibility that dithiazolylidenacetonitrile 43 (X = H) could be unzipped by deprotonation to afford dicyanoacetylene, diatomic sulfur and HCl was investigated. Base H Cl N
CN
S
_ HCl
S
NC
CN
+
S2
43 (X = H)
ISSN 1424-6376
Page 221
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
Treatment of the dithiazolylidenacetonitrile 43 (X = H) with sterically hindered ethyldiisopropylamine (Hünigs base) gave the compounds 53 – 56 but no trace of the monocyclic isothiazole 52. H Cl
CN EtiPr2N (1 equiv.)
N
S
S
+
53 (7 %)
CH2Cl2, 45 oC, 55 min.
54 (34 %) + 55 (4.5 %) +
56 (Trace)
43 (X = H)
The unknown purple colored compound 55 is unstable on silica and hydrolyses during chromatographic isolation to afford a stable purple colored carboxamide 57. Neither compounds 55 and 57 have been identified but each show eight separate carbon resonances in the C-13 NMR and both show the presence of nitriles in the IR spectra. The molecular formulae have been deduced from mass spectrometry (55 C8N4S3 and 57 C8H2N4OS3). Compound 55 formally contains two units of dicyanoacetylene and three sulfur atoms. The highly colored nature of the compound 55 leads to at least four possible structures (Scheme 11) that presumably could arise from a 4 component 10π cycloaddition between two equivalents of diatomic sulfur and two equivalents of dicyanoacetylene to give a fused seven-five heterocycle that then suffers ring contraction through loss of one sulfur atom to give the proposed six-five system. The identity of the purple compounds is being investigated. CN
CN NC
NC
S S
NC
N
CN
CN NC
S
S
S
NC
S S
CN
N S
S
NC
S S
NC
N
N S
S
S
NC
CN
NC
S
S
S
S
N
NC
N
CN
CN
S
S
CN _ 1/8 S8 NC
S
NC
N
S S
Scheme 11 This chemistry looks promising as a possible new route to both pentathiepins and to new methods for the in situ generation of acetylene derivatives. Understanding the mechanisms leading to the formation of these unexpected products will aid in the design of 1,2,3-dithiazoles which can be used to build new heterocyclic compounds.
ISSN 1424-6376
Page 222
©
ARKAT
Issue ICHC-20
ARKIVOC 2006 (vii) 207-223
Acknowledgements The authors wish to thank Professor C. W. Rees for valuable discussions and Dr A. J. P. White for solving the single crystal X-ray structure of the quinoidal 2,2’-bithiazole 40 at Imperial College. The Cyprus Research Promotion Foundation for financial support [Grant No. DRASI/TEXNO/0603/18 (I.C.C.) and PENEK/ENISX/0504/07 (S.S.M.)].
References 1. 2. 3. 4. 5. 6.
7.
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.
Barclay, T.M.; Beer, L.; Cordes, A.W.; Oakley, R.T.; Preuss, K.E.; Taylor, N.J.; Reed R.W. Chem. Commun., 1999, 531. Koutentis, P.A. Molecules, 2005, 10, 346. Gray, M.A.; Rees, C.W.; Williams, D.J. Heterocycles, 1994, 37, 1827. Emayan, K.; Rees, C.W. Bull. Soc. Chim. Belg., 1997, 106, 605. Appel, R.; Janssen, H.; Siray, M.; Knoch, F. Chem. Ber., 1985, 118, 1632. Moore, J.E. U.S. Pat. 4059590, 1977. Appel, R.; Janssen, H.; Haller, I.; Plempel, M. DE 2848221, 1980. Besson, T.; Rees, C.W.; Cottenceau, G.; Pons, A.M. Bioorg. Med. Chem. Lett., 1996, 6, 2343. Cottenceau, G.; Besson, T.; Gautier, V.; Rees, C.W.; Pons, A.M. Bioorg. Med. Chem. Lett., 1996, 6, 529. Thiery, V.; Rees, C.W.; Besson, T.; Cottenceau, G.; Pons, A.M. Eur. J. Med. Chem., 1998, 33, 149. Mayer R.; Foerster E. DD 212387, 1984. Beer, L.; Cordes, A.W.; Haddon, R.C.; Itkis, M.E.; Oakley, R.T.; Reed, R.W.; Robertson, C.M. Chem. Commun., 2002, 1872. Barclay, T.M.; Cordes, A.W.; Oakley, R.T.; Preuss, K.E.; Reed, R.W. Chem. Commun., 1998, 1039. Rees, C.W. J. Heterocyclic Chem., 1992, 29, 639. van der Plas, H.C. Adv. Heterocyclic Chem., 1999, 74, 9. van der Plas, H.C. Adv. Heterocyclic Chem., 1999, 74, 87. Besson, T.; Emayan, K.; Rees, C.W. J. Chem. Soc., Chem. Commun., 1995, 1419. Barclay, T.M.; Beer, L.; Cordes, A.W.; Oakley, R.T.; Preuss, K.E.; Reed, R.W.; Taylor N.J. Inorg. Chem., 2001, 40, 2709. Clarke, D.; Emayan, K.; Rees, C.W. J. Chem. Soc., Perkin Trans. 1, 1998, 77. Choi, S.–H.; Kim, K. Tetrahedron, 1996, 52, 8413. Emayan, K.; English, R.F.; Koutentis, P.A.; Rees, C.W. J. Chem. Soc., Perkin Trans. 1, 1997, 3345. Koutentis, P.A.; Rees, C.W.; White, A.J.P.; Williams, D.J. J. Chem. Soc., Perkin Trans. 1, 1998, 2765. Christoforou, I.C.; Koutentis, P.A.; Rees, C.W. J. Chem. Soc., Perkin Trans. 1, 2002, 1236. Jeon, M.–K.; Kim, K. Tetrahedron, 1999, 55, 9651. Vladuchick S.A. U.S. Pat. 4094985, 1978. Vladuchick S.A.; Fukunaga T.; Simmons H.E.; Webster, O.W. J. Org. Chem., 1980, 45, 5122. Chenard, B.L. Harlow, R.L.; Johnson, A.L.; Vladuchick, S.A. J. Am. Chem. Soc., 1985, 105, 3871.
ISSN 1424-6376
Page 223
©
ARKAT