New regiospecific isothiazole C-C coupling chemistry

Article abstract of DOI:10.1039/b607442a

Regioselective palladium catalysed coupling reactions are achieved in good to high yields, starting from either 3,5-dichloro- or 3,5-dibromoisothiazole-4-carbonitriles 1 and 2, providing 3-halo-5-(hetero/aryl, alkenyl and alkynyl)isothiazoles 3, 4, 6-9 fr

Full text of DOI:10.1039/b607442a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
New regiospecific isothiazole C–C coupling chemistry†‡
Irene C. Christoforou and Panayiotis A. Koutentis*
Received 25th May 2006, Accepted 27th July 2006
First published as an Advance Article on the web 29th August 2006
DOI: 10.1039/b607442a
Regioselective palladium catalysed coupling reactions are achieved in good to high yields, starting from
either 3,5-dichloro- or 3,5-dibromoisothiazole-4-carbonitriles 1 and 2, providing 3-halo-5-(hetero/aryl,
alkenyl and alkynyl)isothiazoles 3, 4, 6–9 from Stille couplings, 3-halo-5-(hetero/arylethynyl)-
isothiazoles 14–19 from Sonogashira and 5,5ꢀ-bi(3-chloroisothiazole-4-carbonitrile) (13) from an
Ullmann type coupling. 3,5-Dibromoisothiazole-4-carbonitrile 2 is more reactive than the
dichloroisothiazole-4-carbonitrile 1 and effective enough for Stille, Negishi and Sonogashira couplings.
5,5-Bi(3-chloroisothiazole-4-carbonitrile) (13) is prepared by a palladium catalysed Ullmann coupling
from 3-chloro-5-iodoisothiazole-4-carbonitrile (11). A variety of 3-substituted isothiazoles
(3-substituents = Cl, Br, OMs, OTs and OTf) are less reactive and fail to give successful Suzuki
couplings at the isothiazole C-3 position. The 3-iodo-5-phenyl-isothiazole-4-carbonitrile (28), prepared
via Sandmeyer iodination, participates successfully in Suzuki, Ullmann type, Stille, Negishi and
Sonogashira coupling reactions. All products are fully characterized.
Isothiazoles display wide-ranging biological activity. Commer-
R
cially important isothiazoles include the Kathonꢀ preservatives,
the artificial sweetener saccharin and the antibacterial sulfa
drug, sulfasomizole. Isothiazoles are being studied as potential
anticancer agents.1 The chemistry and biological uses of the
system has been reviewed extensively.2–6 Most synthetic strate-
gies involve construction of the isothiazole systems with the
desired carbon substituents in place and the routes are therefore
product specific.5 3,5-Dichloroisothiazole-4-carbonitrile (1)7 is
readily available and, in combination with the development of
C–C coupling methodology,8,9 is a potentially useful starting
material for the construction of more complex isothiazoles. Sur-
prisingly little, however, has appeared on isothiazole C–C coupling
chemistry.10 Palladium catalysed cross-coupling reactions were
Scheme 1
reported for several bromo- and iodoisothiazoles to give alkenyl
and alkynyl substituted isothiazoles.11,12 Negishi and Suzuki
cross couplings have also been demonstrated on 3-benzyloxy-
5-iodoisothiazoles.13 Furthermore, an attempted Heck coupling
between 5-bromo-3-methylisothiazole and styrene gave the 3,3ꢀ-
dimethyl-5,5ꢀ-biisothiazole in almost quantitative yield.14 We have
already shown15 that 3,5-dichloro- and 3,5-dibromoisothiazole-4-
carbonitrile (1 and 2)7 react regiospecifically with either phenyl-
boronic acid or phenyltrifluoroborate to give in high yields
3-halo-5-phenylisothiazole-4-carbonitriles 3 and 4, respectively.
Under analogous reaction conditions both the 3-chloro- and
the 3-bromo-5-phenylisothiazole-4-carbonitrile (3 and 4) failed
to give the desired 3,5-diphenylisothiazole-4-carbonitrile (5)16
(Scheme 1).15
As a logical extension of our earlier work we now present
successful regioselective Stille,17 Negishi,18 Sonogashira19 and
Ullmann type20 C–C coupling reactions for 3,5-dihaloisothiazoles
and furthermore the work is extended to show for the first time
Suzuki,21 Stille, Negishi, Ullmann and Sonogashira couplings at
the C-3 position of 3-halo-5-phenylisothiazole-4-carbonitrile.
Stille coupling reaction at C-5
Initial attempts at Stille coupling of 3,5-dichloroisothiazole-4-
carbonitrile (1) and tributylphenyltin in toluene with Pd(OAc)2 as
catalyst could not be driven to completion and most of the starting
isothiazole was recovered. The use of more polar solvent such as
THF or MeCN at reflux improved the product ratio (by TLC)
but even with excess organotin reagent, complete consumption
of starting isothiazole was not achieved. However, complete
consumption of the starting isothiazole was possible with the use
of tributylphenyltin (1 equiv.) in DMF at 100 ◦C for 72 h and
gave 3-chloro-5-phenylisothiazole-4-carbonitrile (3) in moderate
yield (68%). The use of excess tributylphenyltin (2 equiv.) under
Department of Chemistry, University of Cyprus, P.O. Box 20537, 1678
Nicosia, Cyprus. E-mail: koutenti@ucy.ac.cy; Fax: 00357 22 892809;
Tel: 00357 22 892783
† Electronic supplementary information (ESI) available: Experimental
procedures and full characterization data for all new compounds. See
DOI: 10.1039/b607442a
‡ Dedicated to Professor Charles W. Rees on the occasion of his 79th
birthday.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3681–3693 | 3681
©

View Article Online
Table 1 Stille coupling reaction of 3,5-dihaloisothiazole-4-carbonitriles
1 and 2 with Pd(OAc)2 (5 mol%) at 20 ◦C with heating to 100 ◦
(1 equiv.) led to incomplete consumption of starting isothiazole;
the use of 2 equiv., however, did not affect the product yields or
reaction times. As with the Stille and Suzuki reactions no trace of
the 3,5-diphenylisothiazole-4-carbonitrile (5) was observed. The
synthesis of derivatives was not attempted since arylzinc halides
have limited commercial availability and their preparation was less
attractive.
C
Ullmann type homocoupling at C-5
Ha1
R
RSnBu3 (equiv.) Solvent Time/min Yield (%)
Only one example of the 4,4ꢀ-biisothiazole11 and one of the
5,5ꢀ-biisothiazole14 systems have appeared in the literature as
byproducts of Heck cross coupling reactions. Attempts to prepare
the 5,5ꢀ-biisothiazole starting from either 3,5-dichloro- or 3,5-
dibromoisothiazole-4-carbonitrile (1 and 2) using palladium ac-
etate gave predominantly unreacted starting material. Aryl iodides
are known to be more reactive towards homocoupling reactions.9
Several 5-iodoisothiazoles are known and have been prepared
from isothiazoles not substituted at C-5 using both butyllithium
and iodine24 or periodic acid and iodine,25 or by nucleophilic dis-
placement of 5-bromoisothiazoles26 or 5-hydrazinoisothiazoles27
using NaI. Sandmeyer iodination of the readily available 5-
amino-3-chloroisothiazole-4-carbonitrile (10)7 was not, however,
reported. Diazotization28 of 3-, 4- and 5-aminoisothiazoles has
been reported using nitrosyl tetrafluoroborate in a 1 : 1 mixture
of acetic and propionic acids or by treating the amine with
sodium nitrite and concentrated acids.29 Sandmeyer iodination
has also been achieved for 4-aminoisothiazoles using standard
diazotization conditions.30 The successful Sandmeyer iodination
of 3-amino-5-phenylisothiazole-4-carbonitrile (29) using isoamyl
nitrite in iodine saturated MeCN (see below) was modified to
achieve Sandmeyer iodination at C-5. Treatment of an iodine
saturated MeCN solution of the 5-aminoisothiazole 10 with
isoamyl nitrite at ca. 20 ◦C gave the 3-chloro-5-iodo-isothiazole-4-
carbonitrile 11 (55%) together with a trace of 3-chloroisothiazole-
4-carbonitrile 12. The yield of the 5-iodoisothiazole 11 was
significantly improved if the reaction was conducted at higher
temperatures in either refluxing MeCN (ca. 80 ◦C, 79% yield) or
in refluxing nitromethane (ca. 100 ◦C, 83% yield) (Scheme 2).
Cl
Cl
Br
Br
Br
Br
Br
Br
Br
Br
Ph
Ph
Ph
Ph
2-Furyl
2-Furyl
2-Thienyl
Vinyl
Propynyl
Bu3Sn
1
2
1.2
1.2
1
1.2
1
1.2
1.2
1
DMF
DMF
MeCN 20
DMF
MeCN
MeCN 20
MeCN 10
MeCN 45
MeCN 30
MeCN 24 h
72 h
23 h
3 (68)
3 (84)
4 (93)
4 (90)
6 (86)
6 (100)
7 (93)
8 (94)
20
6 h
9 (86)
a
a No reaction after 24 h.
the same conditions gave a shorter reaction time (23 h) and an
improved yield (84%). 3,5-Dibromoisothiazole-4-carbonitrile (2),
however, was more reactive and complete consumption of starting
isothiazole was observed in both MeCN and DMF. The reaction
time and the product yield were also improved. Using MeCN
as solvent, Pd(OAc)2 as catalyst and a variety of commercially
available organotin reagents aryl, heteroaryl, vinyl and propynyl
C-5-substituted isothiazoles 3, 4, 6–9 were prepared (Table 1). No
trace of the 3,5-disubstituted isothiazoles was observed by TLC.
It is worthy of note that the 5-(thien-2-yl)isothiazole (7) was
synthesised in high yield while, according to our previous work,15
Suzuki methodology using 2-thienylboronic acid failed to intro-
duce the 2-thienyl substituent at the C-5 position. These results
demonstrate that the Stille reaction can be used for regioselective
synthesis of C-5-substituted isothiazoles but there are drawbacks;
organotin reagents and their residues are toxic,22 and these residues
could not be removed easily by chromatography. Attempts to
extract the organotin side products with pentane or hexane23
failed because the isothiazole products were also soluble. The
organotin residues were finally removed after recrystallisation of
the isothiazole products from cyclohexane. Another disadvantage
of the Stille coupling reaction was the failure to synthesise the
5,5ꢀ-biisothiazole 13 using the bis(tributyltin) reagent. This 5,5ꢀ-
biisothiazole 13 was prepared using palladium catalysed Ullmann
type homo-coupling (see below).
Negishi coupling reaction at C-5
The organozinc reagent used in Negishi couplings is non-toxic
and product contamination does not occur readily, and therefore
the Negishi coupling reaction is potentially a cleaner alternative
to the Stille coupling. Negishi couplings have been conducted
with 5-iodoisothiazoles13 but there are to date no reports of
coupling reactions with the less reactive chloro- or bromoiso-
thiazoles. Treatment of 3,5-dichloro- and 3,5-dibromoisothiazole-
4-carbonitrile (1 and 2) with phenylzinc chloride (1.5 equiv.)
and bis(triphenylphosphine)palladium (II) dichloride (5 mol%) in
refluxing THF (25 min) gave the desired products 3 and 4 in 84
and 90% yield, respectively. The use of less phenylzinc chloride
Scheme 2 Reagents and conditions: (i) I2 (3 equiv.), MeNO2, i-amylONO
(4 equiv.), 120 ◦C, 1 h; (ii) Pd(OAc)2 (5 mol%), DMF, 140 ◦C, 27.5 h, 86%,
Ar; or Pd(OAc)2 (5 mol%), (o-Tolyl)3P (5 mol%), DMF, 140 ◦C, 9 h, 72%,
Ar; or Pd(OAc)2 (1 equiv.), DMF, 140 ◦C, 2.5 h, 85%, Ar.
With catalytic Pd(OAc)2 (5 mol%) in DMF at 140 ◦C
the 5-iodoisothiazole (11) could be converted into
3682 | Org. Biomol. Chem., 2006, 4, 3681–3693
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
5,5ꢀ-bi(3-chloroisothiazole-4-carbonitrile) (13) in 86% yield
although the reaction required heating for over 27 h. The addition
of catalytic amount of tri-o-tolylphosphine as ligand significantly
reduced the reaction time (9 h) but gave a slightly lower yield
(72%). A faster reaction time (2.5 h) was observed when 1
equivalent of Pd(OAc)2 was used and this had no adverse effect
on the product yield (Scheme 2).
was consumed rapidly (1 h) to give only 3-chloroisothiazole-4-
carbonitrile (12) as major product (38%). The structure of 3-
chloroisothiazole-4-carbonitrile (12) was confirmed by thermal
decarboxylation of the kn◦own 3-chloro-4-cyanoisothiazole-5-
carboxylic acid (21)31 at 200 C (Scheme 3).
Sonogashira coupling reaction at C-5
The reaction of 3,5-dichloroisothiazole-4-carbonitrile (1) with
phenylacetylene, triethylamine (2 equiv.), bis(triphenylphos-
phine)palladium(II) dichloride (5 mol%) and copper iodide (10
mol%) was investigated in several solvents. MeCN and DMF
were suitable with the former giving slightly improved reac-
tion times and yields. In toluene at least 2 equivalents of
phenylacetylene were required to drive the reaction with the 3,5-
dichloroisothiazole-4-carbonitrile (1) to completion. The more
reactive 3,5-dibromoisothiazole-4-carbonitrile (2), however, could
be converted completely to the 5-(phenylethynyl)isothiazole (14)
even in toluene with only 1.2 equivalents of phenylacetylene.
Starting with 3,5-dibromoisothiazole-4-carbonitrile (2) the 3-
thienyl, ferrocenyl and trimethylsilyl derivatives 16, 18 and 19 of
5-ethynylisothiazoles were synthesized in good yields, although
the trimethylsilyl derivative 19 suffered some desilylation to afford
3-bromo-5-ethynylisothiazole-4-carbonitrile (20; Table 2).
Scheme 3 Reagents and conditions: (i) sealed tube, 200 ◦C, 15 min, 76%.
When the reaction of 3-chloro-5-iodoisothiazole-4-carbonitrile
(11) and 2-pyridylacetylene (2 equiv.) was performed in toluene
the desired product was obtained in moderate yield together
with 3-chloroisothiazole-4-carbonitrile (12). A similar result was
obtained when phenylacetylene (2 equiv.) was used. Increasing the
equivalents of 2-pyridylacetylene reduced the reaction time but
did not significantly change the product yields (Table 3).
Clearly under the reaction conditions the 5-iodoisothiazole 11
was labile to reductive hydrodeiodination however, the equivalent
Sonogashira reactions at the C-3 position (see Table 8) did not
show any hydrodeiodination and gave high yields of the desired
3-ethynylisothiazole.
The reaction involving the 3,5-dibromoisothiazole-4-carbo-
nitrile (2) with 2-pyridinylacetylene (1.2 equiv.) failed to reach
completion even with additional equivalents of 2-pyridinyl-
acetylene (up to 3 equiv.) after 24 h. The reaction was therefore
repeated with 3-chloro-5-iodoisothiazole-4-carbonitrile (11), but
when run in either DMF or MeCN the starting isothiazole
Coupling at the less reactive isothiazole C-3
An earlier attempt to achieve the Suzuki coupling reaction at
the C-3 position of 3-chloro- and 3-bromo-5-phenylisothiazole-4-
carbonitrile (3 and 4) resulted in the unexpected synthesis of 3-
phenoxy-5-phenylisothiazole-4-carbonitrile and gave no isolable
trace of the desired 3,5-diphenylisothiazole-4-carbonitrile (5).15
This result indicated the need for more reactive 3-substituted
isothiazoles. For this purpose, the 3-mesylate, tosylate, and triflate
isothiazoles 23, 24 and 25 were synthesized (Table 4) starting from
the readily available 3-hydroxy-5-phenylisothiazole-4-carbonitrile
(22).32 In addition 2-mesyl, and 2-tosylisothiazol-3-one (26 and
27) were isolated as secondary minor byproducts.
Table 2 Sonogashira coupling reaction of 3,5-dihaloisothiazole-4-
carbonitriles 1 and 2 with (PPh3)2PdCl2 (5 mol%), Et3N (2 equiv.) CuI
(10 mol%), 20 ◦C with heating to 100 ◦
C
Table 3 Reaction of 3-chloro-5-iodoisothiazole (11; 0.113 mmol) with
either 2-pyridylacetylene or phenylacetylene in PhMe (2 ml) with Et3N
(2 equiv.), (PPh3)2PdCl2 (5 mol%), CuI (10 mol%) at 20 ◦C with heating to
Hal
R
R–≡ (equiv.) Solvent Time/h
Yields (%)
100 ◦
C
Cl
Cl
Cl
Cl
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Ph
Ph
Ph
Ph
Ph
Ph
Ph
3-Thienyl
2-Pyridyl
2-Pyridyl
2-Pyridyl
1.2
2.0
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
PhMe
PhMe
MeCN
DMF
PhMe
MeCN
DMF
MeCN
PhMe
MeCN
DMF
MeCN
PhMe
PhMe
MeCN
DMF
24
5.25
1
1.5
1
1
1
0.5
24
24
24
14a
14 (76)
14 (91)
14 (80)
15 (86)
15 (90)
15 (86)
16 (77)
17a
17a
17a
R (equiv.)
Time/min
Yields (%)
Ferrocenyl 1.2
1.5
24
0.25
18 (88)
TMS
TMS
TMS
TMS
1.2
1.5
1.2
1.2
19a
19 (69)b
19a
a
2-Pyridyl (1.2)
2-Pyridyl (2)
2-Pyridyl (3)
Ph (2)
24
24
30
15
35
17 (54)
17 (49)
14 (50)
12 (40)
12 (43)
12 (36)
19a
a Incomplete reaction. b 3-Bromo-5-ethynylisothiazole-4-carbonitrile (20)
was also isolated in 14% yield.
a Incomplete reaction after 24 h.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3681–3693 | 3683
©

View Article Online
Table 4 Reaction of 3-hydroxy-5-phenylisothiazole-4-carbonitrile (22)
with particular reagents (2 equiv.) in DCM at 0–10 ◦
been prepared by treating phenylmethylenemalononitrile with
hydrazine.35 Whilst the analogous transformation of isoxazoles
into pyrazoles is well documented,36,37 only one similar report
has appeared on the transformation of isothiazoles into pyrazoles
using arylhydrazines.38
C
Halogens at the isothiazole C-3 position are known to be
labile to alkylamines.39 Treatment of 3-chloroisothiazole 3 with
stoichiometric or excess benzylamine in a variety of solvents
(PhMe, PhCl, DCM, THF, DMF, EtOH) gave only a trace of the
3-benzylamino-5-phenylisothiazole-4-carbonitrile (32), but when
the reaction was repeated in neat benzylamine at 80 ◦C, the
desired product was isolated in 90% yield (Scheme 5). At higher
Yields (%)
Reagent (2 equiv.) Et3N (equiv.) Time/h
R
23–25
26–27
Ms2O
TsCl
Tf2O
1
2
1
0.3
1
0.5
Ms 23 (78)
26 (16)
27 (12)
—
◦
temperatures (150 C) the yield of 32 was reduced and a second
Ts
Tf
24 (84)
product 3,3ꢀ-bis(4-cyano-5-phenylisothiazole-4-carbonitrile) (33)
was observed indicating partial ring cleavage of the isothiazole
possibly initiated by nucleophilic attack of benzylamine on the
ring sulfur atom.
25 (86)a
a Based on recovered 3-hydroxyisothiazole 22 (8%).
Early studies on the acylation and sulfonylation of 3-
hydroxyisothiazoles show a kinetic preference for the formation
of the acyloxy and sulfonyloxyisothiazole due to steric reasons; on
standing, however, the acyl (but not the sulfonyl) group migrates
to the thermodynamically more stable N-acylisothiazolone.33
The O-sulfonylated isothiazoles 23–25 showed no tendency to
isomerisation to their sulfonamide isothiazolone derivatives, but
were unfortunately hydrolytically labile in MeCN (and more so
in DMF) to a variety of Suzuki reaction conditions, affording
the starting 3-hydroxyisothiazole 22. The compounds were sta-
ble to hydrolysis in 1,4-dioxane but unfortunately gave mainly
unreacted starting material after 24 h at reflux. Therefore, the
synthesis of the previously unknown 3-iodo-5-phenylisothiazole-
4-carbonitrile (28) was targeted.
Scheme 4
3-Iodoisothiazoles have, to our knowledge, not been reported in
the literature. Attempts to convert the 3-hydroxyisothiazole 22 into
the 3-iodoisothiazole 28 using neat HI, excess of KI–I2 in refluxing
THF and Ph3P–I2 in DMF at 50 ◦C all failed and gave only
unreacted 3-hydroxyisothiazole 22. The Sandmeyer iodination
route was then investigated but this required the synthesis of
3-amino-5-phenylisothiazole-4-carbonitrile (29) which has been
previously prepared from the reaction of potassium 2,2-dicyano-
1-phenylethenethiolate with chloramine.34 Our strategy, which
attempts to avoid product-specific synthetic routes, required the
use of the prepared 3-halo-5-phenylisothiazoles 3 or 4. Unlike
the halogen at C-5, which can be displaced readily by anhydrous
ammonia in refluxing THF,7 nucleophilic displacement of the
halogen at C-3 requires vigorous conditions. Treatment of 3-
chloro-5-phenylisothiazole-4-carbonitrile (3) with aqueous am-
monia, anhydrous ammonia, potassium phthalimide or sodamide
with various solvents and temperatures either failed to react or
gave very complex mixtures. An attempt to displace the 3-chloro
substituent with hydrazine hydrate or anhydrous hydrazine, in or-
der to prepare the 3-hydrazino-5-phenylisothiazole-4-carbonitrile
(30), gave instead 3-amino-5-phenyl-1H-pyrazole-4-carbonitrile
(31) in quantitative yield (Scheme 4). Pyrazole 31 has previously
Scheme 5
Dimeric isothiazoledisulfides have been previously observed as
major products during electrosynthesis of isothiazoles starting
from 3-aryl-2-phenylsulfonylpropenonitriles.40 Disulfide forma-
tion was proposed to arise from the oxidative dimerisation of
the analogous isothiazole-3-thiolate and it is possible that this
also occurs here. Extraction of the benzylamine reaction mixture
with hot dichloromethane (DCM) afforded a trace of a new
compound, bis(isothiazol-3-ylthio)methane (34), and this could
have arisen from reaction of the proposed isothiazole-3-thiolate
with the DCM.
Cleavage of the benzyl group of 3-(N-benzylamino)isothiazole
(32) was not possible using mild reductive conditions (H2, Pd/C)
or with strong mineral acids possibly due to the amidine nature
of the 3-benzylamino nitrogen. A recent publication on the
debenzylation of amides using NBS and catalytic AIBN offered
an alternative method.41 With the 3-benzylaminoisothiazole 32,
however, a complex reaction was observed. Replacing the NBS
3684 | Org. Biomol. Chem., 2006, 4, 3681–3693
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
with dibromine gave a much cleaner reaction and the desired 3-
aminoisothiazole 29 was isolated in high yield together with traces
of 3-(N-benzamido)-5-phenylisothiazole-4-carbonitrile (35) and
benzaldehyde (Scheme 6).
with dry and degassed DMF did, however, afford for the first time
3,5-diphenylisothiazole-4-carbonitrile (5) in 34% yield together
with a minor amount of 3,3ꢀ-bi(5-phenylisothiazol-4-carbonitrile)
(38).
In light of this promising result the reaction was optimized
first with respect to base, then with respect to boronic acid
and finally with respect to catalyst in DMF. Strong bases, such
as KOH and Cs2CO3, led to decomposition of the starting
isothiazole 28 whilst Li2CO3, KHCO3 and KF required longer
reaction times. Na2CO3 and K2CO3 both afforded reasonable
yields of the diphenylisothiazole 5 in less than 24 h. Of the
two carbonates K2CO3 gave the higher yields and was chosen
for further optimization studies. It was found that freshly and
finely powdered (pestle and mortar) K2CO3 greatly reduced the
reaction times but also the product yields. Similar surface area
effects on the reaction rate have been reported for Buchwald–
Hartwig aminations with Cs2CO3.44 The equivalents of K2CO3
were then carefully screened. With phenylboronic acid (2 equiv.),
Pd(OAc)2 (5 mol%), DMF at 140 ◦C for 2.5 h under an argon
atmosphere with 1.5 equivalents of powdered K2CO3 gave a
consistent 70% yield of the diphenylisothiazole 5 and traces of
the biisothiazole 38. More than (up to 3.5 equiv.) or less than
(down to 0.5 equiv.) 1.5 equivalents of powdered K2CO3 was
detrimental to both yield and reaction time. The equivalents of
phenylboronic acid were then screened; with 3 equivalents of
PhB(OH)2 the reaction time was reduced to 50 min and yields
of the diphenylisothiazole 5 raised to 80%. A further increase in
phenylboronic acid (4 equiv.) did not change either reaction times
or product yields. The DMF was then examined to determine
tolerances in water, air and heating. Wet DMF gave marginally
lower yields but the reaction time was unaffected (50 min). Dry but
non-degassed DMF gave similar results. A significant reduction
of the reaction time to 10 min was observed when the oil bath was
preheated to 140 ◦C, although the product yields were marginally
reduced (70%). In light of these results with the iodo compound
the 3-chloro- and 3-bromo-5-phenylisothiazole-4-carbonitriles 3
and 4 were reinvestigated; however only the 3-bromoisothiazole
4 showed any of the 3,5-diphenylisothiazole 5 (by TLC) and the
reaction was slow and could not be driven to completion.
Scheme 6 Reagents and conditions: (i) Br2 (1.5 equiv.), AIBN (0.2 equiv.),
PhH–H2O (2 : 1), 0.5 h, 80 ◦C.
Diazotization of the 3-aminoisothiazole 29 could not be
achieved using sodium nitrite with a variety of acids (H2SO4,
AcOH, HI) and, similarly, nitrous acid (HNO3–Na2S2O5) also
failed to affect diazotization. In nearly all cases the 3-
aminoisothiazole 29 decomposed or was converted into mainly
polar products, e.g. 3-hydroxy-5-phenylisothiazole-4-carbonitrile
(22). The use of basic solvents such as pyridine or quinoline has
been shown to assist in such cases,42 but the 3-aminoisothiazole
29 was isolated unchanged. Nitrosyl tetrafluoroborate in a 1 : 1
mixture of acetic and propionic acids gave only traces of the
desired 3-iodoisothiazole 28 (by TLC). Similar difficulties in
Sandmeyer reactions with the 3-aminoisothiazole 29 have been
observed.32 Iodination of arylamines is known to proceed well
under aprotic diazotization conditions using isoamyl nitrite in
the presence of iodine43 and similar conditions gave the 3-iodo-5-
phenylisothiazole-4-carbonitrile (28) in good yield together with
two minor by-products the methylenemalononitrile 36 and the
triazene 37 (Scheme 7). The reaction was optimized with respect
to iodine and isoamyl nitrite and required dropwise addition of
an MeCN solution of the 3-aminoisothiazole 29 into an iodine-
saturated MeCN solution of isoamyl nitrite.
Finally a range of commercially available catalysts was investi-
gated; interestingly three catalysts, (PPh3)2PdCl2, (PPh3)4Pd and
(dppf)PdCl2 showed no formation of 3,3ꢀ-bi(5-phenylisothiazole-
4-carbonitrile) (38) but gave significantly lower yields of the
diphenylisothiazole 5 (61–68%) and longer reaction times.
(MeCN)2PdCl2, (PhCN)2PdCl2, and PdCl2 gave both long re-
action times (3–4 h) and reduced yields (60–61%) of the
diphenylisothiazole 5 together with significant traces of the
biisothiazole 38 (3–5%) whilst (dba)3Pd2 gave comparable yields
to Pd(OAc)2 but at a cost of reaction time (2.5 h).
Having partially optimized the reaction conditions for Suzuki
coupling reaction at the isothiazole C-3 position a variety of 3-
arylsubstituted isothiazoles 5, 39–48 were synthesized (Table 5).
The electron deficient 3-nitrobenzeneboronic acid gave a relatively
low yield (58%) of the desired isothiazole 39 despite the addition of
further boronic acid. Electron rich arylboronic acids (MeOC6H4,
and 3-thienyl) gave higher yields. Sterically hindered boronic
acids such as 2-tolyl- and 2-chloro-benzeneboronic acids gave
either a low yield of the desired isothiazole or could not be
driven to completion within 24 h. As with the C-5 coupling
Scheme 7 Reagents and conditions: (i) i-AmylONO (4 equiv.), I2
(2.5 equiv.), MeCN, 5 ◦C, 0.5 h.
Suzuki coupling reactions at C-3
The synthesis of 3-iodo-5-phenylisothiazole-4-carbonitrile (28)
provided a new opportunity to attempt the Suzuki reaction at the
isothiazole C-3 position. The conditions [PhMe, KF, 18-Crown-6,
Pd(OAc)2] that were successfully used for Suzuki couplings at C-
515 did not work with the 3-iodoisothiazole 28. Replacing toluene
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3681–3693 | 3685
©

View Article Online
Table 5 Reaction of 3-iodo-5-phenylisothiazole-4-carbonitrile (29) with
RB(OH)2 (3–4 equiv.), powdered K2CO3 (1.5 equiv.), Pd(OAc)2 (5 mol%),
in dry degassed DMF at 20 ◦C and heated to 140 ◦C under Ar
Table 6 Reaction of 3-iodo-5-phenylisothiazole-4-carbonitrile (28) with
Pd(OAc)2, in dry DMF, under Ar
Pd(OAc)2 (equiv.)
Temp. (◦C)
Time/h
Yield 38 (%)
a
0.05
0.5
1
1
1
20–140
20–140
20–140
140c
24
5d
16.5
5
12 min
5 min
52b
76
74
57
56
170c
1
200c
a Trace of dimer 38 (by TLC) after 24 h. b Yield of 38 is based on recovered
3-iodoisothiazole 28 (6%). c In microwave reactor, 255 W, ramp time 10 s.
Yields (%)
R
RB(OH)2 (equiv.) Time/h
5, 39–48 38
Replacing the copper catalyst with Pd(OAc)2 (5 mol%) in dry
degassed DMF failed to give more than a trace of the 3,3ꢀ-
biisothiazole 38. The use of triarylphosphine ligands Ph3P or (o-
toly)3P, reductive conditions (Zn–H2O), basic conditions (Hu¨nig’s
base) or the introduction of either Et4NBr or CuI did not improve
the yield of the 3,3ꢀ-biisothiazole 38. The reaction times and
product yields were improved with additional Pd(OAc)2 and after
16.5 h with 1 equiv. of Pd(OAc)2 the desired product 38 was
obtained in 76% yield (Table 6). A significant decrease in the
reaction time was observed under microwave conditions at 140 ◦C
although the yield remained unchanged. Harsher microwave
Ph
3
1
5 (80)
1
6
5
6
3-NO2C6H4
4-MeOC6H4
3-MeOC6H4
2-MeOC6H4
4-Tol
3-Tol
2-Tol
4-ClC6H4
3-ClC6H4
2-ClC6H4
3-Thienyl
2-Thienyl
Me
3.5
3.5
3
1.17
1.75
1
0.34
0.42
0.5
39 (58)
40 (95)
41 (84)
42 (95)
43 (75)
4
5
3.5
3.5
4
3.5
3.5
3.5
3
Ndc
44 (91)
4
a
a
24
0.34
0.34
0.34
0.5
24
24
45 (82)
46 (75)
47 (58)
4
3
7
48 (91)
2
b
b
3
3
b
b
◦
conditions, at 170 or 200 C, provided a further decrease in the
a Incomplete reaction after 24 h. b No reaction. c Nd = no data.
reaction time but the product yield dropped to 56–57%.
Several other commercial palladium catalysts were compared
with Pd(OAc)2 but no appreciable benefits could be discerned.
(MeCN)2PdCl2, (PhCN)2PdCl2 and PdCl2 (1 or 2 equiv.) led to
complete consumption of the starting iodoisothiazole 28 but gave
lower yields (32–71%) compared to Pd(OAc)2 and a complex
reaction mixture was observed with (PPh3)2PdCl2 while (dba)3Pd2
gave no trace of biisothiazole 38; with (dppf)PdCl2 no reaction
was observed.
chemistry the 2-thienyl-boronic acid failed to react presumably
due to protodeboronation of the boronic acid. However unlike the
coupling at C-5, methylboronic acid failed to react at C-3. 3,3ꢀ-
Biisothiazole 38 was formed in trace quantities in all reactions.
Ullmann type homocoupling reaction at C-3
Whilst the monocyclic 4,4ꢀ- and 5,5ꢀ-biisothiazoles have been
reported,11,14 only the 3,3ꢀ-bibenzoisothiazole moiety has appeared
in the literature.45 Therefore an independent synthesis of the
monocyclic 3,3ꢀ-biisothiazole 38 was attempted via the tradi-
tional copper catalysed Ullmann reaction starting from the 3-
iodoisothiazole 28. Treatment of the 3-iodoisothiazole 28 with
either stoichiometric or excess copper powder in refluxing MeCN,
benzene, toluene or xylene gave slow reactions and only traces of
product (by TLC). The reaction went to completion when DMF
was used as solvent but the dimer 38 was isolated in moderate yield
(31%) together with an unexpected isomeric by-product 49 (34%)
which could only arise from degradation of another isothiazole
(Scheme 8). A similar 3-[(Z)-(2-cyano-2-phenylethenyl)thio]-5-
phenylisothiazole was isolated during electrosynthesis of isoth-
iazoles starting from 3-aryl-2-phenylsulfonylpropenonitriles.40
Stille coupling reaction at C-3
The readiness of 3-chloro-, 3-bromo- and 3-iodoisothiazoles 3, 4,
and 28 to participate in Stille coupling reaction was investigated.
Treatment of either 3-chloro- or the 3-bromo-5-phenylisothiazole-
4-carbonitrile (3 or 4) with tributylphenyltin (up to 3 equiv.)
gave only incomplete reactions (by TLC) even after 24 h. On
the other hand addition of tributylphenyltin (1 equiv.) and
Pd(OAc)2 (5 mol%) to 3-iodo-5-phenylisothiazole-4-carbonitrile
(28) in DMF at 100 ◦C gave the desired 3,5-diphenylisothiazole 5
in 94% yield together with traces of the 3,3ꢀ-biisothiazole 38. Under
these reaction conditions heteroaryl, including the 2-thienyl, vinyl
and propynyl isothiazole derivatives 50–53, were synthesised in
high yields (Table 7). In all the reaction mixtures the 3,3ꢀ-
biisothiazole 38 was observed in trace quantities. Recrystallisation
(cyclohexane) of the products isolated by chromatography was
sufficient to remove the toxic organotin residues.
Negishi coupling reaction at C-3
A comparison of the reactivity of the 3-halo-5-phenylisothiazole-
4-carbonitriles 3, 4 and 28 with respect to the Negishi coupling
reactions showed that in the presence of bis(triphenyl-
phosphine)palladium(II) dichloride (5 mol%) in dry degassed
DMF the 3-chloroisothiazole 3 failed to react completely even
Scheme 8
3686 | Org. Biomol. Chem., 2006, 4, 3681–3693
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 7 Reaction of 3-halo-5-phenylisothiazole-4-carbonitriles 3, 4 and
4-carbonitrile (11), both the TMS and 2-pyridyl derivatives 55 and
57 were prepared in excellent yield.
28 with RSnBu3 and Pd(OAc)2 (5 mol%) in dry degassed DMF at 20 ◦
C
heated to 100 ◦C, under Ar
Yields (%)
Summary
Hal
R
RSnBu3 (equiv.) Time/h
5, 50–53 38
Regioselective Stille, Negishi, Sonogashira, and Ullmann
type C–C coupling reactions were demonstrated with 3,5-
dihaloisothiazole-4-carbonitriles 1 and 2 in good to high yields
at C-5 position of the isothiazole ring. The analogous couplings
at C-3 were much less readily achieved and the preparation of
the more reactive 3-iodo-5-phenylisothiazole-4-carbonitrile (28)
was required. This, the first 3-iodo substituted isothiazole was
achieved via a Sandmeyer iodination of the 3-amino precursor
29. 3-Iodo-5-phenylisothiazole-4-carbonitrile (28) was sufficiently
reactive to undergo Suzuki, Stille, Negishi, Sonogashira and
Ullmann type coupling reactions at the C-3 position. The re-
activity of haloisothiazoles towards the coupling methodology
followed the anticipated order I > Br > Cl. The work described
demonstrates the synthetic usefulness of the readily available
3,5-dichloroisothiazole-4-carbonitrile (1) when combined with
powerful palladium catalysed C–C coupling.
a
Cl
Br
I
I
I
I
I
Ph
Ph
Ph
2-Thienyl
2-Furyl
Vinyl
Propynyl
3
3
1
1
1
1
1
24
24
1.34
0.5
1.67
24
24
5a
a
5a
5 (94)
Trace
Trace
50 (87)
51 (91)
52 (96)
53 (73)
Trace
Trace
Trace
a Incomplete reaction (by TLC) after 24 h.
when an excess of phenylzinc chloride (3 equiv.) was used.
Under analogous conditions both the 3-bromoisothiazole 4
and 3-iodoisothiazole 28 afforded the 3,5-diphenylisothiazole-4-
carbonitrile 5 in 74 and 78% yields, respectively, in only 20 min.
The use of less than 3 equivalents of phenylzinc chloride resulted
in incomplete reactions in all cases.
Experimental
Sonogashira coupling reaction at C-3
Solvents PhH and PhMe were freshly distilled from CaH2 under
argon. DMF was azeotropically distilled with PhH then redistilled
The three 3-haloisothiazoles 3, 4 and 28 were investigated with
respect to the Sonogashira reaction using an alkyne, triethyl-
amine (2 equiv.), bis(triphenylphosphine)palladium(II) dichloride
(5 mol%) and copper iodide (10 mol%) in dry degassed DMF and
the order of reactivity was determined to follow I > Br > Cl.
3-Chloro-5-phenylisothiazole-4-carbonitrile (3) gave incomplete
reactions, even when an excess of phenylacetylene (3 equiv.) was
used. In contrast, 3-bromo- and 3-iodoisothiazoles 4 and 28 gave
complete reactions with 2 and 1 equivalent of phenylacetylene, re-
spectively. Several 3-substituted-acetyleneisothiazoles 54–58 were
prepared from the 3-iodoisothiazole 28 in good yields (Table 8).
Surprisingly, and in contrast with the 3-chloro-5-iodoisothiazole-
˚
under vacuum from anhydrous MgSO4 and stored over 4 A molec-
ular sieves under argon. THF was freshly distilled from potassium
under argon. Anhydrous K2CO3 was freshly powdered using an
agate pestle and mortar before use. Reactions were protected by
CaCl2 drying tubes or performed under an argon atmosphere.
Anhydrous MgSO4 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). Mi-
crowave mediated chemistry was performed with a CEM Discover
Microwave Reactor and reaction temperatures were controlled
using standard IR thermometry. Melting points were determined
using a PolyTherm-A, Wagner & Munz, Koefler-Hotstage Micro-
scope apparatus. Solvents used for recrystallization are indicated
after the melting point. UV spectra were obtained using a Perkin-
Elmer Lambda-25 UV/vis spectrophotometer and inflections are
identified by the abbreviation “inf”. IR spectra were recorded on a
Shimidazu FTIR-NIR Prestige-21 spectrometer with a Pike Mira-
cle Ge ATR accessory and strong, medium and weak peaks are rep-
resented by s, m and w respectively. 1H and 13C NMR spectra were
recorded on a Bruker Avance 300 machine (at 300 and 75 MHz,
respectively). Deuterated solvents were used for homonuclear lock
and the signals are referenced to the deuterated solvent peaks.
Low resolution (EI) mass spectra were recorded on a Shimadzu
Q2010 GCMS with a direct inlet probe whilst high resolution
spectra were recorded on a VG Autospec “Q” mass spectrometer.
3,5-Dichloroisothiazole-4-carbonitrile (1),7 3,5-dibromoisothia-
zole-4-carbonitrile (2),7 3-chloro-5-phenylisothiazole-4-carbonit-
rile (3),15 3-bromo-5-phenylisothiazole-4-carbonitrile (4),15
Table 8 Reaction of 3-halo-5-phenylisothiazole-4-carbonitriles 3, 4 and
28 with Et3N (2 equiv.), (PPh3)2PdCl2 (5 mol%), CuI (10 mol%) in dry
degassed DMF at 20 ◦C heated to 100 ◦C, under Ar
Hal
R
R–≡ (equiv.)
Time/h
Yield (%)
Cl
Cl
Cl
Br
I
I
I
Ph
Ph
Ph
Ph
1.2
2
3
2
1
1.5
1.2
1.2
2
24
24
24
2.85
1.25
4.25
1.5
0.5
0.5
54a
54a
54a
54 (87)
54 (92)
55 (70)
56 (91)
57 (92)
58 (100)
Ph
TMS
3-Thienyl
2-Pyridinyl
Ferrocenyl
I
I
a Incomplete reaction after 24 h, mainly isothiazole 3 (TLC).
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3681–3693 | 3687
©

View Article Online
5-amino-3-chloroisothiazole-4-carbonitrile (10),7 3-chloro-4-
cyanoisothiazole-5-carboxylic acid (21)31 and 3-hydroxy-5-phenyl-
isothiazole-4-carbonitrile (22)32 were prepared according to
literature procedures.
CDCl3) 158.4 (C-5), 151.2, 111.0, 109.7; dC (75 MHz; coupled
CDCl3) 158.4 (d, J 192.3, C-5), 151.2 (d, J 12.3, C-3), 111.0 (d,
J 2.9, C≡N or C-4), 109.7 (d, J 3.4, C-4 or C≡N); dC (75 MHz;
DEPT 90, CDCl3) 158.3 (CH); m/z (EI) 146 (M+ + 2, 37%), 144
(M+, 100), 108 (M+–HCl, 1), 93 (CClNS+, 40), 83 (C3HNS+, 92), 82
(23), 58 (6), 51 (13) (found: M+, 143.9549, C4HClN2S requires M,
143.9549).
3-Chloro-5-phenylisothiazole-4-carbonitrile (3; typical Stille
coupling conditions at C-5: Table 1)
A stirred mixture of 3,5-dichloroisothiazole-4-carbonitrile (1;
30 mg, 0.168 mmol), tributylphenyltin (109.7 ll, 0.336 mmol,
2 equiv.) and Pd(OAc)2 (1.9 mg, 5 mol%) in DMF (2 ml) protected
with a CaCl2 drying tube, was heated to ca. 100 ◦C, until no
starting material remained (TLC). The mixture was allowed to
cool to ca. 20 ◦C, diluted with DCM (15 ml) and washed with
H2O (4 × 10 ml). The organic layer was separated, dried and
the volatiles evaporated. The residue obtained was absorbed on
silica and chromatography (hexane–DCM, 7 : 3) and gave the ti◦tle
compound 3 (31.1 mg, 84%) as colourless needles, mp 87–88 C
(from cyclohexane) identical to an authentic sample.
3-Chloroisothiazole-4-carbonitrile (12) from
3-chloro-4-cyanoisothiazole-5-carboxylic acid (21)
A thick walled glass pressure tube was charged with 3-chloro-
4-cyanoisothiazole-5-carboxylic acid (21; 100 mg, 0.531 mmol),
sealed and heated in a preheated Woods metal bath to ca. 200 ◦C
for 15 min. The residue was allowed to cool to ca. 20 ◦C and
absorbed on silica. Chromatography (hexane–DCM, 5 : 5) gave
the title compound 12 (58.3 mg, 76%) as colourless needles, mp
50–51 ◦C (from pentane) identical to that described above.
5,5ꢀ-Bi(3-chloroisothiazole-4-carbonitrile) (13)
3-Chloro-5-phenylisothiazole-4-carbonitrile (3; typical Negishi
coupling conditions at C-5)
A stirred mixture of 3-chloro-5-iodoisothiazole-4-carbonitrile (11;
30 mg, 0.11 mmol) and Pd(OAc)2 (24.7 mg, 0.11 mmol) in DMF
(2 ml) under an argon atmosphere, was heated to ca. 140 ◦C until
no starting material remained (TLC). The mixture was allowed to
cool to ca. 20 ◦C, diluted with DCM (15 ml) and washed with H2O
(4 × 10 ml). The organic layer was separated, dried and absorbed
on silica. Chromatography (hexane–DCM, 1 : 4) gave the title
compound 13 (27 mg, 86%) as colourless needles, mp 244–245 ◦C
(from PhH); (found: C, 33.5; N, 19.5. C8Cl2N4S2 requires C, 33.5;
N, 19.5%); kmax (DCM)/nm 295 (log e 3.03); mmax/cm−1 2234w
(C≡N), 1634w, 1468s, 1356w, 1341s, 1285w, 1240w, 1065s, 887w,
818s, 741s; dC (75 MHz; DMSO-d6) 160.8, 149.8, 111.1, 110.0; m/z
(EI) 290 (M+ + 4, 17%), 288 (M+ + 2, 75), 286 (M+, 100), 251 (3),
240 (4), 225 (14), 207 (3), 190 (4), 187 (4), 146 (3), 126 (5), 108 (6),
93 (37), 82 (13), 70 (9), 64 (9) (found: M+, 285.8944, C8Cl2N4S2
requires M, 285.8941).
A stirred mixture of 3,5-dichloroisothiazole-4-carbonitrile (1;
30 mg, 0.168 mmol), phenylzinc chloride (504 ll, 0.252 mmol,
0.5M in THF, 1.5 equiv.) and (PPh3)2PdCl2 (5.9 mg, 5 mol%)
in dry and degassed THF (2 ml) under an argon atmosphere,
was heated to ca. 60 ◦C, until no starting material remained
(TLC). The mixture was allowed to cool to ca. 20 ◦C and the
volatiles were evaporated. The residue was absorbed on silica
and chromatography (hexane–DCM, 7 : 3) and gave the ti◦tle
compound 3 (34.5 mg, 93%) as colourless needles, mp 87–88 C
(from cyclohexane) identical to an authentic sample.
3-Chloro-5-iodoisothiazole-4-carbonitrile (11)
To a stirred and heated (ca. 110 ◦C) mixture of iodine (238.5 mg,
0.94 mmol, 3 equiv.) and isoamyl nitrite (168 ll, 1.25 mmol,
4 equiv.) in nitromethane (2 ml) was added dropwise a ni-
tromethane (1 ml) solution of 5-amino-3-chloroisothiazole-4-
carbonitrile (10; 50 mg, 0.313 mmol). The reaction mixture was
kept at ca. 110 ◦C until no starting material remained (TLC)
and then allowed to cool to ca. 20 ◦C and absorbed on silica.
Chromatography (hexane–DCM, 7 : 3) gave the title compound
11 (70 mg, 83%) as colourless needles, mp 117–118 ◦C (from
pentane); (found: C, 17.8; N, 10.3. C4ClIN2S requires C, 17.8;
N, 10.4%); kmax (DCM)/nm 267 (log e 3.00); mmax/cm−1 2232 w
(C≡N), 1479m, 1371w, 1360w, 1325s, 1221w, 1076w, 1070w, 953w,
810s, 781m; dC (75 MHz; CDCl3) 151.3, 118.8, 112.8, 111.5; m/z
(EI) 272 (M+ + 2, 35%), 270 (M+, 100), 209 (M+–CClN, 28), 177
(M+–CClNS, 3), 143 (M+–I, 14), 127 (I+, 24), 108 (M+–ClI, 15), 93
(CClNS+, 4), 82 (C3NS+, 67), 70 (3), 56 (4) (found: M+, 269.8510,
C4ClIN2S requires M, 269.8516). Further elution (hexane–DCM,
3 : 2) gave 3-chloroisothiazole-4-carbonitrile (12; 7 mg, 16%) as
colourless needles, mp 50–51 ◦C (from pentane); (found: C, 33.3;
H, 0.7; N, 19.4. C4HClN2S requires C, 33.2; H, 0.7; N, 19.4%);
kmax (DCM)/nm 263 (log e 3.02); mmax/cm−1 3109w and 3098w
(CH), 2241w (C≡N), 1497m, 1368w, 1356w, 1335s, 1207w, 1153w,
1144w, 1061m, 1047m, 866m, 841m, 829m, 822m, 816m, 731w;
dH (300 MHz; CDCl3) 9.23 (1H, s, H-5); dC (75 MHz; decoupled
3-Chloro-5-(phenylethynyl)isothiazole-4-carbonitrile (14; typical
Sonogashira conditions at C-5: Table 2)
A stirred mixture of 3,5-dichloroisothiazole-4-carbonitrile (1;
30 mg, 0.168 mmol), CuI (3.2 mg, 10 mol%), (PPh3)2PdCl2 (5.9 mg,
5 mol%), ethynylbenzene (22.1 ll, 0.202 mmol, 1.2 equiv.) and
triethylamine (46.8 ll, 0.336 mmol, 2 equiv.) in DMF (2 ml) was
◦
heated to ca. 100 C until no starting material remained (TLC).
The mixture was allowed to cool to ca. 20 ◦C, diluted with DCM
(15 ml) and washed with H2O (4 × 10 ml). The organic layer
was separated, dried and absorbed on silica. Chromatography
(hexane–DCM, 7 : 3) gave the title compound 14 (32.9 mg, 80%) as
colourless needles, mp 74–75 ◦C (from pentane); (found: C, 58.8;
H, 2.0; N, 11.3. C12H5ClN2S requires C, 58.9; H, 2.1; N, 11.5%);
kmax (DCM)/nm 230 (log e 3.69), 301 inf (372), 320 (3.85), 336
(3.88); mmax/cm−1 2236w (C≡N), 2207m (C≡C), 1512m, 1483w,
1443w, 1393w, 1348s, 1279w, 1258w, 1070w, 1026w, 1001w, 993m,
961w, 926w, 876m, 818m, 762s; dH (300 MHz; CDCl3) 7.63–7.59
(2H, m, Ph H), 7.53–7.40 (3H, m, Ph H); dC (75 MHz; CDCl3) (1
peak missing) 156.7, 149.9, 132.2 (Ph CH), 131.0 (Ph CH), 128.7
(Ph CH), 119.9, 111.4, 110.7, 75.0 (C≡C); dC (75 MHz; DEPT 90,
CDCl3) 132.2 (Ph CH), 131.0 (Ph CH), 128.7 (Ph CH); m/z (EI)
3688 | Org. Biomol. Chem., 2006, 4, 3681–3693
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
246 (M+ + 2, 79%), 244 (M+, 100), 209 (13), 183 (10), 165 (38), 151
(19), 145 (8), 139 (13), 124 (7), 117 (3), 113 (3), 93 (11), 75 (4), 63
(5) (found: M+, 243.9871, C12H5ClN2S requires M, 243.9862).
7.72 (2H, m, Ph H), 7.61–7.50 (3H, m, Ph H), 7.41 (2H, d, J 8.5,
Tol H), 2.48 (3H, s, CH3); dC (75 MHz; CDCl3) 176.1, 158.8, 146.6,
132.3 (Ar CH), 132.1, 130.0 (Ar CH), 129.8 (Ar CH), 129.0 (Ar
CH), 127.6, 127.1 (Ar CH), 111.3 (C≡N), 97.0 [C(C≡N)], 21.8
(CH3); dC (75 MHz; DEPT 90, CDCl3) 132.3 (Ar CH), 130.0 (Ar
CH), 129.8 (Ar CH), 129.0 (Ar CH), 127.1 (Ar CH); m/z (EI) 356
(M+, 0.2%), 292 (M+–SO2, 20), 155 (47), 127 (8), 100 (4), 91 (100),
77 (5), 65 (24) (found: M+, 356.0283, C17H12N2O3S2 requires M,
356.0289). Further elution (hexane–Et2O, 1 : 4) gave 4-cyano-5-
phenyl-2-(4-tosyl)isothiazol-3-one (27; 21 mg, 12%) as colourless
needles, mp 190–191 ◦C (from t-BuOMe); (found: C, 57.4; H,
3.3; N, 7.8. C17H12N2O3S2 requires C, 57.3; H, 3.4; N, 7.9%); kmax
(DCM)/nm 230 (log e 2.86), 296 (3.01); mmax/cm−1 3065w (Ar
CH), 2957w, 2922w and 2855 (CH3), 2228w (C≡N), 1730w, 1701s,
1593w, 1551w, 1487w, 1447w, 1377m, 1329w, 1288w, 1175s, 1123w,
1094w, 1080m, 984w, 928w, 907w, 810w, 800w, 772m, 762m, 743m,
702w; dH (300 MHz; CDCl3) 8.05 (2H, d, J 8.4, Tol H), 7.77–7.73
(2H, m Ph H), 7.70–7.64 (1H, m, Ph H), 7.61–7.53 (2H, m, Ph
H), 7.42 (2H, d, J 8.1, Tol H), 2.48 (CH3); dC (75 MHz; CDCl3)
168.7, 161.8, 147.3, 134.1 (Ar CH), 132.5 (Ar C), 130.2 (Ar CH),
130.1 (Ar CH), 129.1 (Ar CH), 127.0 (Ar CH), 126.9 (Ar C),
111.9 (C≡N), 95.9 [C(C≡N)], 21.9 (CH3); dC (75 MHz; DEPT
90, CDCl3) 134.1 (Ar CH), 130.2 (Ar CH), 130.1 (Ar CH), 129.2
(Ar CH), 127.0 (Ar CH); m/z (EI) 356 (M+, 0.1%), 292 (M+–SO2,
17), 155 (53), 128 (4), 127 (7), 100 (4), 91 (100), 77 (6), 65 (27)
(found: M+, 356.0288, C17H12N2O3S2 requires M, 356.0289).
3-Methanesulfonyloxy-5-phenylisothiazole-4-carbonitrile (23)
To
a stirred solution of 3-hydroxy-5-phenylisothiazole-4-
carbonitrile (22; 100 mg, 0.495 mmol) and triethylamine (69 ll,
0.495 mmol, 1 equiv.) in DCM (2 ml) cooled to ca. 0 ◦C was added
in one portion methanesulfonic anhydride (172.5 mg, 0.99 mmol,
2 equiv.). The reaction mixture was kept at ca. 0 ◦C until no starting
material remained (TLC). Chromatography (hexane–DCM, 5 : 3)
gave the title compound 23 (108 mg, 78%) as colourless needles,
mp 104–105 ◦C (from cyclohexane); (found: C, 47.2; H, 2.7; N, 9.9.
C11H8N2O3S2 requires C, 47.1; H, 2.9; N, 10.0%); kmax (DCM)/nm
279 (log e 3.17); mmax/cm−1 2236w (C≡N), 1558w, 1535w, 1495w,
1449w, 1429w, 1387s, 1329w, 1188s, 1126s, 1082w, 978m, 908w,
866m, 781s, 770s, 731m, 714s; dH (300 MHz; CDCl3) 7.80–7.76
(2H, m, Ph H), 7.64–7.53 (3H, m, Ph H), 3.56 (3H, s, CH3); dC
(75 MHz; CDCl3) 176.6, 159.0, 132.6 (Ph CH), 129.9 (Ph CH),
127.4 (Ph C), 127.1 (Ph CH), 111.2 (C≡N), 97.0 [C(C≡N)], 40.4
(CH3); dC (75 MHz; DEPT 90, CDCl3) 132.5 (Ph CH), 129.9 (Ph
CH), 127.1 (Ph CH); m/z (EI) 280 (M+, 50%), 216 (M+–SO2,
4), 202 (100), 187 (7), 173 (3), 159 (13), 146 (11), 142 (24), 128
(51), 121 (12), 114 (6), 100 (22), 88 (4), 79 (32), 77(13), 63 (6), 51
(11) (found: M+, 279.9977, C11H8N2O3S2 requires M, 279.9976).
Further elution (hexane–t-BuOMe, 1 : 4) gave 4-cyano-2-mesyl-
5-phenylisothiazol-3-one (26; 22 mg, 16%) as colourless needles,
mp 182–183 ◦C (from t-BuOMe); (found: C, 47.1; H, 2.8; N, 9.9.
C11H8N2O3S2 requires C, 47.1; H, 2.9; N, 10.0%); kmax (DCM)/nm
295 (log e 3.08); mmax/cm−1 3030w and 3011w (Ar CH), 2930 (CH3),
3-Trifluoromethanesulfonyloxy-5-phenylisothiazole-4-carbonitrile
(25)
To
a stirred solution of 3-hydroxy-5-phenylisothiazole-4-
=
2232w (C≡N), 1701s (C O), 1593w, 1545w, 1489w, 1447w, 1416w,
carbonitrile (22; 100 mg, 0.495 mmol) and triethylamine (69 ll,
0.495 mmol, 1 equiv.) in DCM (2 ml) cooled to ca. 0 ◦C was added
dropwise trifluoromethanesulfonic anhydride (167 ll, 0.99 mmol,
2 equiv.). The reaction mixture was kept at ca. 0 ◦C until no starting
material remained (TLC). Chromatography (hexane–DCM, 5 : 3)
gave the title compound 25 (142 mg, 86%) as colourless needles,
mp 67–68 ◦C (from cyclohexane); (found: C, 39.5; H, 1.5; N, 8.2.
C11H5F3N2O3S2 requires C, 39.5; H, 1.5; N, 8.4%); kmax (DCM)/nm
281 (log e 3.05); mmax/cm−1 2236w (C≡N), 1541w, 1497w, 1450w,
1414m, 1224s, 1165w, 1134m, 1109m, 1101m, 1032w, 1001w, 951w,
910m, 862m, 791m, 770m, 762m, 692m, 687m; dH (300 MHz;
CDCl3) 7.82–7.78 (2H, m, Ph H), 7.67–7.56 (3H, m, Ph H);
dC (75 MHz; CDCl3) (1 peak missing) 177.7, 155.6, 132.9 (Ph
1368s, 1335m, 1290w, 1171s, 1099m, 1005w, 964s, 939w, 908w,
773s, 758w, 741m; dH (300 MHz; CDCl3) 7.80–7.77 (2H, m, Ph
H), 7.73–7.68 (1H, m, Ph H), 7.63–7.58 (2H, m, Ph H), 3.59
(3H, s, CH3); dC (75 MHz; CDCl3) 169.4, 162.9, 134.4 (Ph CH),
130.2 (Ph CH), 128.3, 127.1 (Ph CH), 126.8, 111.7 (C≡N), 95.8
[C(C≡N)], 42.0 (CH3); dC (75 MHz; DEPT 90, CDCl3) 134.4 (Ph
CH), 130.2 (Ph CH), 127.1 (Ph CH); m/z (EI) 280 (M+, 73%), 215
(6), 202 (M+–CH2O2S, 100), 187 (9), 173 (5), 159 (16), 146 (13),
142 (23), 128 (59), 121 (11), 114 (5), 100 (16), 88 (4), 79 (CH2O2S+,
21), 77 (11), 69 (3), 63 (5), 51 (10), 46 (11) (found: M+, 279.9973,
C11H8N2O3S2 requires M, 279.9976).
1
CH), 130.0 (Ph CH), 127.2 (Ph CH), 118.5 (1C, q, JCF 319.5,
3-(4-Toluenesulfonyloxy)-5-phenylisothiazole-4-carbonitrile (24)
CF3), 110.3, 96.8; dC (75 MHz; DEPT 90, CDCl3) 132.9 (Ph
CH), 130.0 (Ph CH), 127.2 (Ph CH); m/z (EI) 334 (M+, 78%),
270 (M+–SO2, 45), 201 (4), 196 (100), 186 (7), 176 (8), 159 (10),
To
a stirred solution of 3-hydroxy-5-phenylisothiazole-4-
carbonitrile (22; 100 mg, 0.495 mmol) and triethylamine (69 ll,
0.495 mmol, 1 equiv.) in DCM (2 ml) cooled to ca. 0 ◦C was added
in one portion 4-toluenesulfonyl chloride (188.7 mg, 0.99 mmol,
2 equiv.). The reaction mixture was kept at ca. 0 ◦C until no starting
material remained (TLC). Chromatography (hexane–DCM, 5 : 3)
gave the title compound 24 (148 mg, 84%) as colourless needles,
mp 94–95 ◦C (from cyclohexane); (found: C, 57.4; H, 3.3; N, 7.8.
C17H12N2O3S2 requires C, 57.3; H, 3.4; N, 7.9%); kmax (DCM)/nm
214 (log e 4.87), 278 (4.01); mmax/cm−1 2234w (C≡N), 1597w,
1539m, 1495w, 1449w, 1379s, 1294w, 1217w, 1194m, 1180s, 1123m,
1088m, 1038w, 1016w, 999w, 955w, 910w, 862s, 814m, 800w, 768m,
743s, 712w; dH (300 MHz; CDCl3) 8.00 (2H, d, J 8.4, Tol H), 7.76–
+
146 (8), 127 (39), 114 (3), 100 (8), 84 (5), 77 (C6H5 , 8), 69 (58),
63 (4), 51(6) (found: M+, 333.9695. C11H5F3N2O3S2 requires M,
333.9694). Further elution gave 3-hydroxy-5-phenylisothiazole-4-
carbonitrile (8 mg, 8%) as colourless needles, mp 233–234 ◦C (from
PhH) (lit.,32 235–236 ◦C), identical to an authentic sample.
3-Amino-5-phenylpyrazole-4-carbonitrile (31)
A stirred mixture of 3-chloro-5-phenylisothiazole-4-carbonitrile
(3; 100 mg, 0.454 mmol) in 80% hydrazine hydrate (2 ml) was
heated to ca. 80 ◦C for 1 h. The mixture was poured onto
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3681–3693 | 3689
©

View Article Online
crushed ice (50 g) to form a white precipitate. Filtration gave
the t◦itle compound 31 (84 mg, 100%) as white powder, mp 194–
195 C (from H2O–EtOH) (lit.,35 mp 200 ◦C); kmax (EtOH)/nm
205 (log e 4.39), 234 (4.21), 253 inf (4.11); mmax/cm−1 3348w,
3304w, 3184m, 3169m, 3129w, 3098w, 3049w, 3019w, 2978w,
2953w, 2909w, 2835w, 2232s (C≡N), 1649w, 1580w, 1568w, 1535m,
1501m, 1493m, 1443w, 1422w, 1350w, 1171w, 1140w, 1078m,
1026w, 966w, 916w, 816w, 768s, 725m; dH (300 MHz; DMSO-d6)
12.18 (1H, br s, NH), 7.82–7.80 (2H, m, Ph H), 7.49–7.37 (3H, m,
Ph H), 6.45 (2H, br s, NH2); dC (75 MHz; DMSO-d6) (Ph CH peak
missing) 154.9 (C-3), 150.3 (C-5), 132.1 (Ph C), 129.0 (Ph CH),
125.9 (Ph CH), 116.4 (C≡N), 69.9 (C-4); dC (75 MHz; DEPT 90,
DMSO-d6) (Ph CH peak missing) 129.0 (Ph CH), 125.9 (Ph CH);
m/z (EI) 184 (M+, 100%), 167 (1), 155 (10), 142 (10), 128 (13), 121
(3), 115 (3), 106 (11), 102 (4), 91 (25), 77 (15), 65 (4), 63 (3), 51(9),
(found: M+, 184.0749, C10H8N4 requires M, 184.0749).
C20H10N4S4 requires M, 433.9788). If the reaction mixture is
initially extracted with hot DCM then chromatography (hexane–
DCM, 7 : 3) of the extracts gave in addition to the above
products 3,3ꢀ-methylenebis(sulfanediyl)bis(5-phenylisothiazole-4-
carbonitrile) (34; 0.5 mg, 1%) as colourless needles, mp 134–135 ◦C
(from THF); (found: C, 56.1; H, 2.5; N, 12.4. C21H12N4S4 requires
C, 56.2; H, 2.7; N, 12.5%); kmax (DCM)/nm 230 (log e 3.24),
287 (3.34); mmax/cm−1 30098w, (Ar CH), 2222w (C≡N), 1582w,
1510w, 1481s, 1441w, 1381w, 1327m, 1250w, 1223m, 1180w, 1155w,
1101w, 1078w, 1051s, 999w, 961w, 920w, 837s, 785w, 758s, 739m;
dH (300 MHz; CDCl3) 7.77–7.74 (2H, m, Ph H), 7.60–7.50 (3H,
m, Ph H), 5.16 (1H, s, CH2); dC (75 MHz; CDCl3) 175.8, 163.7,
131.7 (Ph CH), 129.5 (Ph CH), 127.5 (Ph C), 127.3 (Ph CH),
112.2 (C≡N), 103.1 [C(C≡N)], 32.9 (CH2); dC[75 MHz; DEPT
135, CDCl3 + Cr(acac)3] (Ph CH), 129.5 (Ph CH), 127.3 (Ph CH),
32.9 (CH2); m/z (EI) 448 (M+, 38%), 415 (M+–HS, 4), 401 (7),
284 (3), 265 (8), 231 (100), 219 (11), 187 (8), 181 (3), 163 (7), 159
(7), 144 (17), 139 (9), 135 (12), 121 (55), 109 (33), 87 (12), 58 (13);
(found: M+, 447.9961, C21H12N4S4 requires M, 447.9945).
3-(Benzylamino)-5-phenylisothiazole-4-carbonitrile (32)
A stirred solution of 3-chloro-5-phenylisothiazole-4-carbonitrile
(3; 50 mg, 0.227 mmol) in benzylamine (2 ml) was heated to ca.
80 ◦C until no starting material remained (TLC). The mixture
was diluted with DCM (15 ml) and was washed with 10% aq.
HCl (4 × 10 ml) followed by saturated aq. Na2S2O5 (4 × 10 ml).
The organic layer was separated, dried and absorbed on silica.
Chromatography (hexane–DCM, 7 : 3) gave the title compound
32 (55 mg, 90%) as colourless needles, mp 127–128 ◦C (from
cyclohexane); (found: C, 70.2; H, 4.4; N, 14.4. C17H13N3S requires
C, 70.1; H, 4.5; N, 14.4%); kmax (DCM)/nm 229 (log e 3.12), 280
(3.05), 338 (2.44); mmax/cm−1 3381m (NH), 3058w and 3025w (Ph
CH), 2217m (C≡N), 1556s, 1538m, 1496w, 1458w, 1449w, 1419w,
1349m, 1300w, 1196w, 1160w, 1106w, 1083w, 1065w, 1033w,
1027w, 1011w, 1001w, 956w, 912w, 875w, 771s; dH (300 MHz;
CDCl3) 7.75–7.72 (2H, m, Ph H), 7.55–7.47 (3H, m, Ph H), 7.42–
7.30 (5H, m, Ph H), 5.29 (1H, br s, NH), 4.65 (2H, s, CH2); dC
(75 MHz; CDCl3) 174.3, 164.3, 138.0 (Ph C), 131.3 (Ph CH),
129.5 (Ph CH), 128.7 (Ph C), 128.6 (Ph CH), 127.8 (Ph C), 127.7
(Ph CH), 127.7 (Ph CH), 127.1 (Ph CH), 114.0 (C≡N), 92.2
[C(C≡N)], 47.0 (CH2); dC (75 MHz; DEPT 135, CDCl3) 131.3
(Ph CH), 129.5 (Ph CH), 128.7 (Ph CH), 127.8 (Ph CH), 127.7
(Ph CH), 127.1 (Ph CH), 47.0 (CH2); m/z (EI) 291 (M+, 100%),
290 (30), 275 (3), 258 (3), 218 (4), 214 (5), 186 (5), 159 (3), 155 (3),
3-Amino-5-phenylisothiazole-4-carbonitrile (29)
A
stirred mixture of 3-(benzylamino)-5-phenylisothiazole-4-
carbonitrile (32; 50 mg, 0.185 mmol), bromine (14.2 ll,
0.278 mmol, 1.5 equiv.), AIBN (6 mg, 0.037 mmol, 0.2 equiv.) and
◦
PhH–H2O (2 : 1, 3 ml) was heated to ca. 80 C until no starting
material remained (TLC). The organic layer was separated, dried
and then absorbed on silica. Chromatography (hexane–DCM,
4 : 1) gave the title compound 29 (33 mg, 90%) as colourless
needles, mp 127–128 ◦C (from cyclohexane) (lit.,34 128.5 ◦C); kmax
(DCM)/nm 228 (log e 3.68), 280 (3.83), 325 (3.35); mmax/cm−1
3436w (NH), 3288w and 3193w (Ph CH), 2218m (C≡N), 1618s,
1549s, 1501s, 1465w, 1440w, 1411s, 1337w, 1306w, 1290w, 1270w,
1197w, 1159w, 1100w, 1066w, 1031w, 1002w, 957w, 901w, 844s,
761m; dH (300 MHz; CDCl3) 7.73–7.70 (2H, m, Ph H), 7.53–7.46
(3H, m, Ph H), 5.02 (2H, br s, NH2); dC (75 MHz; CDCl3) 174.5,
164.6, 131.4 (Ph CH), 129.4 (Ph CH), 128.4 (Ph C), 127.0 (Ph CH),
113.9 (C≡N), 92.7 [C(C≡N)]; dC (75 MHz; DEPT 90, CDCl3)
131.4 (Ph CH), 129.5 (Ph CH), 127.0 (Ph CH); m/z (EI) 201 (M+,
100%), 174 (M+–HCN, 10), 159 (15), 153 (13), 128 (49), 114 (7),
100 (11), 88 (5), 77 (18), 74 (41), 63 (5), 51 (13). Further elution
(DCM) gave 3-benzoylamino-5-phenylisothiazole-4-carbonitrile
(35; 1 mg, 1%) as colourless needles, mp 169–170 ◦C (from
cyclohexane); (found: C, 66.7; H, 3.5; N, 13.7. C17H11N3OS
requires C, 66.9; H, 3.6; N, 13.8%); kmax (DCM)/nm 275 (log
+
146 (3), 141 (3), 128 (5), 121 (7), 106 (BnNH+, 37), 91 (PhCH2 ,
80), 77 (C6H5 , 9), 65 (11), 51 (5) (found: M+, 291.0829. C17H13N3S
+
requires M, 291.0830). Further elution (hexane–DCM, 7 : 3) gave
3,3ꢀ-bis-(4-cyano-5-phenylisothiazole)disulfide (33; 0.5 mg, 1%) as
colourless needles, mp 138–139 ◦C (from EtOH); (found: C, 55.2;
H, 2.3; N, 12.8. C20H10N4S4 requires C, 55.3; H, 2.3; N, 12.9%);
kmax (DCM)/nm 230 (log e 3.20), 287 (3.30); mmax/cm−1 2224m
(C≡N), 1514m, 1483s, 1443m, 1387w, 1325m, 1290w, 1269w,
1240w, 1190w, 1103w, 1076w, 1045m, 1026w, 999w, 955w, 916w,
826m, 766s, 760s; dH (300 MHz; CDCl3) 7.79–7.72 (4H, m, Ph
H), 7.59–7.50 (6H, m, Ph H); dC (75 MHz; CDCl3) 177.0, 162.1,
132.0 (Ph CH), 129.7 (Ph CH), 127.5 (Ph C), 127.4 (Ph CH), 112.3
(C≡N), 105.1 [C(C≡N)]; dC (75 MHz; DEPT 90, CDCl3) 132.0
(Ph CH), 129.7 (Ph CH), 127.4 (Ph CH); m/z (EI) 434 (M+, 68%),
401 (M+–HS, 52), 369 (M+–HS2, 15), 337 (7), 249 (7), 218 (100), 190
(13), 185 (10), 159 (26), 141 (13), 128 (77), 121 (34), 114 (9), 100 (7),
e 3.32); mmax/cm−1 3246w (NH), 2232w (C≡N), 1672s (C O),
=
1537s, 1503m, 1472w, 1443w, 1427w, 1381s, 1285m, 1273m, 1254w,
1179w, 1153w, 1101w, 1076w, 1026w, 1001w, 961w, 935w, 920m,
883w, 841m, 795w, 762s, 716s; dH (300 MHz; CDCl3) 8.85 (1H, br s,
NH), 7.98–7.95 (2H, m, Ph H), 7.80–7.77 (2H, m, Ph H), 7.64–
7.49 (6H, m, Ph H); dC (75 MHz; CDCl3) 175.7, 165.0, 157.3, 133.0
(Ph CH), 132.4 (Ph C), 131.8 (Ph CH), 129.7 (Ph CH), 128.9 (Ph
CH), 127.9 (Ph C), 127.7 (Ph CH), 127.4 (Ph CH), 112.9 (C≡N),
99.5 [C(C≡N)]; dC (75 MHz; DEPT 90, CDCl3) 133.0 (Ph CH),
131.8 (Ph CH), 129.7 (Ph CH), 128.9 (Ph CH), 127.7 (Ph CH),
127.4 (Ph CH); m/z (EI) 305 (M+, 16%), 277 (M+–CO, 7), 218 (1),
201 (2), 184 (3), 127 (3), 105 (PhCO+, 100), 84 (5), 77 (C6H5 ,
50), 51 (11) (found: M+, 305.0646, C17H11N3OS requires M,
305.0623).
+
+
90 (22), 77 (C6H5 , 30), 69 (6), 63 (5), 51 (19) (found: M+, 433.9790.
3690 | Org. Biomol. Chem., 2006, 4, 3681–3693
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
3-Iodo-5-phenylisothiazole-4-carbonitrile (28)
3 equiv.), powdered K2CO3 (33.2 mg, 0.24 mmol, 1.5 equiv.) and
Pd(OAc)2 (1.8 mg, 5 mol%) in dry and degassed DMF (2 ml)
under an argon atmosphere, was heated to ca. 140 ◦C, until no
starting material remained (TLC). The mixture was allowed to
cool to ca. 20 ◦C, diluted with DCM (15 ml) and washed with
H2O (4 × 10 ml). The organic layer was separated, dried and
absorbed on silica. Chromatography (hexane–DCM, 3 : 2) gave
the title compound 5 (34 mg, 80%) as whit◦e needles, mp 146–
147 ◦C (from cyclohexane) (lit.,16 mp 149–150 C); (found: C, 73.2;
H, 3.7; N, 10.7. C16H10N2S requires C, 73.3; H, 3.8; N, 10.7%);
kmax (DCM)/nm 262 (log e 3.18); mmax/cm−1 3061w and 3030w
(Ph CH), 2226w (C≡N), 1518w, 1481m, 1445m, 1410w, 1364m,
1074w, 1032w, 1001w, 966w, 912m, 839m, 770m, 760m, 718s; dH
(300 MHz; CDCl3) 8.08–8.03 (2H, m, Ar H), 7.83–7.79 (2H, m, Ar
H), 7.57–7.52 (6H, m, Ar H); dC (75 MHz; CDCl3) 176.8, 168.9,
132.9 (Ph C), 131.5 (Ph CH), 130.4 (Ph CH), 129.6 (Ph CH),
128.8 (Ph CH), 128.1 (Ph C), 127.9 (Ph CH), 127.7 (Ph CH),
114.9 (C≡N), 103.6 [C(C≡N)]; dC (75 MHz; DEPT 90, CDCl3)
131.5 (Ph CH), 130.4 (Ph CH), 129.6 (Ph CH), 128.8 (Ph CH),
127.9 (Ph CH), 127.7 (Ph CH); m/z (EI) 262 (M+, 100%), 261 (6),
229 (M+–HS, 4), 218 (6), 159 (3), 135 (7), 134 (4), 131 (7), 121
◦
To a stirred and cooled (ca. 0–5 C) mixture of iodine (158 mg,
0.623 mmol, 2.5 equiv.) and isoamyl nitrite (134 ll, 0.996 mmol,
4 equiv.) in MeCN (2 ml) was added dropwise an MeCN (1 ml)
solution of 3-amino-5-phenylsothiazole-4-carbonitrile (29; 50 mg,
◦
0.249 mmol). The reaction mixture was kept at ca. 0–5 C until
no starting material remained (TLC), allowed to warm to ca.
20 ◦C and absorbed on silica. Chromatography (hexane–DCM,
3 : 7) gave the title compound 28 (66 mg, 85%) as colourless
needles, mp 123.5–124.5 ◦C (from cyclohexane); (found: C, 38.5;
H, 1.6; N, 8.9. C10H5IN2S requires C, 38.5; H, 1.6; N, 9.0%);
kmax (DCM)/nm 229 (log e 2.95), 285 (3.07); mmax/cm−1 3032w
(Ph CH), 2232m (C≡N), 1508w, 1477s, 1445w, 1377m, 1327m,
1233m, 1188w, 1107w, 1080w, 1036m, 1016w, 997m, 962w, 923w,
816s, 770s, 756s; dH (300 MHz; CDCl3) 7.77–7.73 (2H, m, Ph
H), 7.60–7.50 (3H, m, Ph H); dC (75 MHz; CDCl3) 175.7, 132.1,
(Ph CH), 129.8 (Ph CH), 127.4 (Ph CH), 126.8, 114.4, 114.1,
113.8; dC (75 MHz; DEPT 90, CDCl3) 132.1, (Ph CH), 129.8
(Ph CH), 127.4 (Ph CH); m/z (EI) 312 (M+, 100%), 185 (M+–
I, 25), 158 (10), 153 (3), 141 (28), 127 (4), 121 (5), 114 (8), 100
(3), 84 (4), 77 (C6H5 , 17), 63 (3), 51 (12) (found: M+, 311.9209,
+
+
(4), 103 (PhCN+, 6), 77 (C6H5 , 13), 51 (8) (found: M+, 262.0558,
C10H5IN2S requires M, 311.9218). Further elution (hexane–DCM,
3 : 2) gave 1,1-dicyano-2-iodo-2-phenylethene (36; 1 mg, 1%) as
yellow needles, mp 114–115 ◦C (from cyclohexane); (found: C,
42.9; H, 1.7; N, 10.0. C10H5IN2 requires C, 42.9; H, 1.8; N, 10.0%);
kmax (DCM)/nm 282 inf (log e 2.85), 319 (3.00); mmax/cm−1 3030w
(Ph CH), 2224w (C≡N), 1591w, 1574w, 1537m, 1508m, 1485w,
1441w, 1377w, 1327w, 1233m, 1179w, 1157w, 1076w, 1036w, 999w,
924w, 878w, 833w, 816w, 770w, 752s; dH (300 MHz; CDCl3) 7.63–
7.58 (2H, m, Ph H), 7.57–7.45 (3H, m, Ph H); dC (75 MHz;
CDCl3) 142.4, 138.7, 133.1 (Ph CH), 129.0 (Ph CH), 129.0 (Ph
CH), 115.3 (C≡N), 112.2 (C≡N), 96.2 [C(CN)2]; dC (75 MHz;
DEPT 90, CDCl3) 133.1 (Ph CH), 129.0 (Ph CH), 129.0 (Ph CH);
m/z (EI) 280 (M+, 39%), 153 (M+–I, 100), 126 (16), 100 (7), 77
C16H10N2S requires M, 262.0565). Further elution gave 3,3ꢀ-bi(5-
phenylisothiazole-4-carbonitrile) (38; 0.3 mg, 1%) as colourless
needles, mp 151–152 ◦C (from cyclohexane–PhH); (found: C, 64.6;
H, 2.8; N, 15.2. C20H10N4S2 requires C, 64.8; H, 2.7; N, 15.1%); kmax
(DCM)/nm 278 (log e 3.35); mmax/cm−1 3053w (Ph CH), 2232m
(C≡N), 1508w, 1476s, 1443m, 1373m, 1339m, 1331m, 1233w,
1188w, 1105w, 1080w, 1030m, 993m, 962w, 914w, 835s, 764s, 733w;
dH (300 MHz; CD2Cl2) 7.87–7.84 (4H, m, Ph H), 7.63–7.59 (6H,
m, Ph H); dC (75 MHz; CD2Cl2) 177.5, 160.8, 132.2 (Ph CH),
130.1 (Ph CH), 128.8 (Ph CH), 127.9 (Ph C), 113.8 (C≡N), 105.6
[C(C≡N)]; dC (75 MHz; DEPT 90, CD2Cl2) 132.2 (Ph CH), 130.1
(Ph CH), 128.8 (Ph CH); m/z (EI) 370 (M+, 91), 369 (80), 344
(M+–CN, 2), 337 (M+–HS, 4), 305 (2), 290 (5), 274 (2), 242 (2), 211
(4), 205 (2), 185 (M2+, 14), 177 (5), 159 (11), 141 (2), 133 (20), 127
(9), 121 (10), 115 (9), 103 (11), 89 (48), 87 (25), 77 (13), 73 (52), 59
(16) (found: M+, 370.0354 C20H10N4S2 requires M, 3700347).
(C6H5 , 24), 75 (8), 63 (5), 51 (12) (found: M+, 279.9500, C10H5IN2
+
requires M, 279.9498). Further elution (DCM) gave 3,3ꢀ-(triaz-
1-ene-1,3-diyl)bis(5-phenylisothiazole-4-carbonitrile) (37; 0.5 mg,
◦
1%) as pale yellow needles, mp 196–197 C (from cyclohexane);
(found: C, 57.9; H, 2.7; N, 23.6. C20H11N7S2 requires C, 58.1; H,
2.7; N, 23.7%); kmax (DCM)/nm 296 (log e 3.55), 335 inf (3.26);
mmax/cm−1 2226w (C≡N), 1584m, 1574m, 1537w, 1522w, 1493w,
1479w, 1423s, 1385m, 1260w, 1227s, 1186w, 1119w, 1080w, 1032w,
1001w, 974w, 883w, 872w, 853w, 764m, 727m, 714m; dH (300 MHz;
DMSO-d6) 14.67 (1H, br s, NH), 7.86–7.81 (4H, m, Ph H), 7.67–
7.64 (6H, m, Ph H); dC (75 MHz; DMSO-d6) (C-4 peak is missing)
176.1, 131.9 (Ph CH), 129.7 (Ph CH), 128.2 (Ph C), 127.5 (Ph
CH), 127.5 (Ph C), 113.1 (C≡N); dC (75 MHz; DEPT 90, DMSO-
d6) 131.9 (Ph CH), 129.7 (Ph CH), 127.5 (Ph CH); m/z (EI) 384
(M+–HN2, 100), 352 (5), 312 (5), 308 (2), 275 (2), 242 (8), 213 (25),
201 (27), 185 (49), 178 (9), 158 (10), 153 (7), 141 (30), 128 (13), 121
3,3ꢀ-Bi(5-phenylisothiazole-4-carbonitrile) (38; using Cu(0)
catalysed Ullmann conditions)
A stirred mixture of 3-iodo-5-phenylisothiazole-4-carbonitrile (28;
50 mg, 0.16 mmol) and Cu(0) powder (20.3 mg, 0.32 mmol) in
DMF (2 ml) under an argon atmosphere, was heated to ca. 110 ◦C,
until no starting material remained (TLC). The mixture was
allowed to cool to ca. 20 ◦C, diluted with DCM (15 ml) and washed
with H2O (4 × 10 ml). The organic layer was separated, dried and
absorbed on silica. Chromatography (hexane–DCM, 1 : 4) gave
the title compound 38 (9.2 mg, 31%) as colourless needles, mp
◦
151–152 C (from cyclohexane–PhH) identical to that described
(14), 114 (9), 91 (56), 77 (C6H5 , 29), 51 (12) (found: M+–HN2,
+
above. Further elucidation (hexane–DCM, 1 : 4) gave 2-[(4-cyano-
5-phenylisothiazol-3-ylthio)(phenyl)methylene]malononitrile (49;
10 mg, 34%) as colourless needles, mp 118–119 ◦C (from cyclohex-
ane); (found: C, 64.7; H, 2.6; N, 14.9. C20H10N4S2 requires C, 64.8;
H, 2.7; N, 15.1%); kmax (DCM)/nm 228 (log e 3.92), 302 (4.10);
mmax/cm−1 3046w (Ph CH), 2228m (C≡N), 1593w, 1531m, 1510w,
1481m, 1444m, 1379m, 1339w, 1331w, 1288w, 1252w, 1238w,
1190w, 1105w, 1080w, 1049w, 1026w, 1001w, 949w, 926w, 864w,
384.0381, C20H10N5S2 requires M–HN2, 384.0378).
3,5-Diphenylisothiazole-4-carbonitrile (5; typical Suzuki
conditions for coupling at C-3: see Table 5)
A stirred mixture of 3-iodo-5-phenylisothiazole-4-carbonitrile (28;
50 mg, 0.16 mmol), phenylboronic acid (58.5 mg, 0.48 mmol,
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3681–3693 | 3691
©

View Article Online
826m, 804w, 762s, 700m; dH (300 MHz; CDCl3) 7.61–7.38 (8H,
m, Ph H); dC (75 MHz; CDCl3) 176.8, 174.2, 156.2, 133.0 (Ph
CH), 132.5 (Ph CH), 132.1 (Ph C), 129.9 (Ph CH), 129.6 (Ph
CH), 129.0 (Ph CH), 127.3 (Ph CH), 126.8 (Ph C), 112.0 (C≡N),
112.0 (C≡N), 111.4 (C≡N), 109.1 [C(C≡N)], 83.7 [C(CN)2]; dC
(75 MHz; DEPT 90, CDCl3) 133.0 (Ph CH), 132.5 (Ph CH), 129.9
(Ph CH), 129.6 (Ph CH), 129.0 (Ph CH), 127.3 (Ph CH); m/z (EI)
370 (M+, 100%), 344 (M+–CN, 4), 337 (M+–HS, 13), 312 (4), 305
(5), 293 (3), 267 (8), 242 (3), 218 (18), 185 (6), 184 (3), 159 (7), 153
(33), 141 (17), 128 (15), 126 (16), 121 (PhCS+, 50), 114 (9), 100
CuI (2.1 mg, 10 mol%), (PPh3)2PdCl2 (3.9 mg, 5 mol%) and
ethynylbenzene (24.2 ll, 0.22 mmol, 2 equiv.) in dry and degassed
DMF (2 ml) under an argon atmosphere, was heated to ca.
◦
100 C, until no starting material remained (TLC). The mixture
was allowed to cool to ca. 20 ◦C, diluted with DCM (15 ml) and
washed with H2O (4 × 10 ml). The organic layer was separated,
dried and absorbed on silica. Chromatography (hexane–DCM, 7 :
3) gave the title compound 54 (27.4 mg, 77%) as pink crystals, mp
◦
122–123 C (from cyclohexane); (found: C, 75.4; H, 3.4; N, 9.8;
C18H10N2S requires C, 75.5; H, 3.5, N, 9.8%); kmax (DCM)/nm 287
(log e 3.41), 302 (3.31); mmax/cm−1 3064w (Ph CH), 2230m (C≡N),
2216m (C≡C), 1518m, 1495m, 1481m, 1447m, 1418m, 1362m,
1219w, 1090w, 1080w, 1069w, 1026w, 999w, 961w, 918w, 839m,
770m, 758s, 718m; dH (300 MHz; CDCl3) 7.84–7.77 (2H, m, Ph
H), 7.70–7.66 (2H, m, Ph≡H), 7.60–7.52 (3H, m, Ph H), 7.47–
7.37 (3H, m, Ph≡H); dC (75 MHz; CDCl3) 174.5, 152.6, 132.4 (Ph
CH), 131.8 (Ph CH), 130.0 (Ph CH), 129.8 (Ph CH), 128.5 (Ph
CH), 127.5 (Ph C), 127.4 (Ph CH), 120.7 (Ph C), 113.1 (C◦N),
108.1 [C(C≡N)], 94.7 (C≡C), 81.1 (C≡C); dC (75 MHz; DEPT 90,
CDCl3) 132.4 (Ph CH), 131.8 (Ph CH), 130.0 (Ph CH), 129.7 (Ph
CH), 128.5 (Ph CH), 127.4 (Ph CH); m/z (EI) 286 (M+, 100%),
253 (2), 159 (M+–Ph–C≡C–CN, 59), 143 (9), 127 (M+–Ph–C≡C–
CNS, 31), 121 (6), 115 (8), 100 (6), 88 (3), 77 (7), 63 (3), 51 (5)
(found: M+, 286.0571 C18H10N2S requires M, 2860565).
+
(6), 90 (6), 84 (7), 77 (C6H5 , 50), 69 (5), 63 (5), 56 (10), 51 (23)
(found: M+, 370.0348, C20H10N4S2 requires M, 370.0347).
3,3ꢀ-Bi(5-phenylisothiazole-4-carbonitrile) (38; using Pd(0)
catalysed Ullmann conditions: see Table 6)
A stirred mixture of 3-iodo-5-phenylisothiazole-4-carbonitrile (28;
50 mg, 0.16 mmol) and Pd(OAc)2 (35.9 mg, 0.16 mmol) in DMF
(2 ml) under an argon atmosphere, was heated to ca. 140 ◦C, until
no starting material remained (TLC). The mixture was allowed to
cool to ca. 20 ◦C, diluted with DCM (15 ml) and washed with H2O
(4 × 10 ml). The organic layer was separated, dried and absorbed
on silica. Chromatography (hexane–DCM, 1 : 4) gave the title
compound 38 (22.5 mg, 76%) as colourless needles, mp 151–152 ◦C
(from cyclohexane–PhH) identical to that described above.
Acknowledgements
3,5-Diphenylisothiazole-4-carbonitrile (5) via Stille coupling
reaction at C-3 (typical Stille conditions for coupling at C-3: see
Table 7)
The authors wish to thank Cyprus Research Promotion Foun-
dation (grant no. DRASI/TEXNO/0603/18) and 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 and finally Bert
U. W. Maes and Charles W. Rees for helpful discussions.
A stirred mixture of 3-iodo-5-phenylisothiazole-4-carbonitrile (28;
30 mg, 0.096 mmol), tributylphenyltin (37.6 ll, 0.115 mmol,
1.2 equiv.) and Pd(OAc)2 (1 mg, 5 mol%) in dry and degassed DMF
(2 ml) under an argon atmosphere, was heated to ca. 100 ◦C, until
no starting material remained (TLC). The mixture was allowed to
cool to ca. 20 ◦C, diluted with DCM (15 ml) and washed with H2O
(4 × 10 ml). The organic layer was separated, dried and absorbed
on silica. Chromatography (hexane–DCM, 7 : 3) gave the title
compound 5 (23.6 mg, 94%) as white needles, mp 146–147 ◦C
(from cyclohexane) identical to that described above.
References
1 E. R. Larson and M. C. Noe, T. G. Gant , US Pat. 6 235 764, 2001.
2 D. L. Pain, B. J. Peart and K. R. H. Wooldridge, in Comprehensive
Heterocyclic Chemistry, vol. 6, ed. K. T. Potts, A. R. Katritzky and
C. W. Rees, Pergamon, Oxford, 1984, ch. 4.17, p. 131.
3 R. F. Chapman and B. J. Peart, in Comprehensive Heterocyclic
Chemistry II, vol. 3, ed. I. Shinkai, A.R. Katritzky, C. W. Rees and
E. F. V. Scriven, Pergamon, Oxford, 1996, ch. 3.05, p. 319.
4 R. V. Kaberdin and V. I. Potkin, Russ. Chem. Rev., 2002, 71, 673.
5 D. W. Brown and M. Sainsbury, in Science of Synthesis, vol. 11, ed.
E. Schaumann, 2002, p. 567.
3,5-Diphenylisothiazole-4-carbonitrile (5) via Negishi coupling
reaction at C-3
A stirred mixture of 3-iodo-5-phenyl-4-isothiazolecarbonitrile (28;
30 mg, 0.096 mmol), phenylzinc chloride (576 ll, 0.5 M in THF,
3 equiv.) and (PPh3)2PdCl2 (3.4 mg, 5 mol%) in dry and degassed
DMF (2 ml) under an argon atmosphere, was heated to ca.
6 A.-S. S. H. Elgazwy, Tetrahedron, 2003, 59, 7445.
7 W. R. Hatchard, J. Org. Chem., 1964, 29, 660.
◦
8 N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457.
9 J. Hassan, M. Se´vignon, C. Gozzi, E. Schulz and M. Lemaire, Chem.
Rev., 2002, 102, 1359.
10 S. Schro¨ter, C. Stock and T. Bach, Tetrahedron, 2005, 61, 2245.
11 S. G. Zlotin, P. G. Kislitsin and O. A. Luk’yanov, Russ. Chem. Bull.,
1998, 47, 517.
100 C, until no starting material remained (TLC). The mixture
was allowed to cool to ca. 20 ◦C, diluted with DCM (15 ml) and
washed with H2O (4 × 10 ml). The organic layer was separated,
dried and absorbed on silica. Chromatography (hexane–DCM, 7 :
3) gave the title compound 5 (19.6 mg, 78%) as white needles, mp
146–147 ◦C (from cyclohexane) identical to that described above.
12 S. G. Zlotin, P. G. Kislitsin and O. A. Luk’yanov, Russ. Chem. Bull.,
1998, 47, 519.
13 B. H. Kaae, P. Krogsgaard-Larsen and T. N. Johansen, J. Org. Chem.,
2004, 69, 1401.
14 H. Yamanaka, M. Shiraiwa, E. Yamamoto and T. Sakamoto, Chem.
Pharm. Bull., 1981, 29, 3543.
5-Phenyl-3-(phenylethynyl)isothiazole-4-carbonitrile (54; typical
Sonogashira conditions for coupling at C-3: see Table 8)
15 I. C. Christoforou, P. A. Koutentis and C. W. Rees, Org. Biomol. Chem.,
2003, 1, 2900.
16 R. R. Crenshaw and R. A. Partyka, J. Heterocycl. Chem., 1970, 7, 871.
A stirred mixture of 3-bromo-5-phenylisothiazole-4-carbonitrile
(4; 30 mg, 0.11 mmol), triethylamine (30.7 ll, 0.22 mmol, 2 equiv.),
3692 | Org. Biomol. Chem., 2006, 4, 3681–3693
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
17 J. K. Stille, Angew. Chem., Int. Ed. Engl., 1986, 25, 508.
18 E.-I. Negishi, Acc. Chem. Res., 1982, 15, 340.
19 K. Sonogashira, Y. Tohda and N. Hagihara, Tetrahedron Lett., 1975,
4467.
20 F. Ullmann and J. Bielecki, Chem. Ber., 1901, 34, 2174.
21 N. Miyaura, T. Yanagi and A. Suzuki, Synth. Commun., 1981, 11,
513.
22 (a) L. Magos, in Handbook on the Toxicology of Metals, 2nd edn, ed.
L. Friberg, G. F. Nordberg and V. Voulk, Elsevier, New York, 1986,
p. 568; (b) P. J. Smith, Toxicological Data on Organotin Compounds,
International Tin Research Institute, London, 1978.
23 J. M. Berge and S. M. Roberts, Synthesis, 1979, 471.
24 P. Cuadrado, A. M. Gonza´lez and F. J. Pulido, Synth. Commun., 1988,
18, 1847.
Y. Araki, H. Sakuma, T. Kinbara and K. Imanishi, WO Pat. 00/29398,
2000.
32 R. J. A. Walsh and K. R. H. Wooldridge, J. Chem. Soc., Perkin Trans.
1, 1972, 1247.
33 A. W. K. Chan and W. D. Crow, Aust. J. Chem., 1968, 21, 2967.
34 K. Hartke and L. Peshkar, Arch. Pharm., 1968, 611.
35 S. Kobayashi, Chem. Pharm. Bull., 1973, 21, 941.
36 C. Musante, Gazz. Chim. Ital., 1942, 72, 537.
37 C. Musante, Gazz. Chim. Ital., 1943, 73, 355.
38 H. Fu¨rstenwerth, US Pat. 4 892 958, 1990.
39 C. C. C. Cutr`ı, A. Garozzo, M. A. Siracusa, M. C. Sarva`, A. Castro,
E. Geremia, M. R. Pinizzotto and F. Guerrera, Bioorg. Med. Chem.,
1999, 7, 225.
40 A. Kunugi, M. A. Jabbar, K. Mori and H. Uno, Electrochim. Acta,
1999, 44, 4583.
25 S. G. Zlotin, P. G. Kislitsin and O. A. Luk’yanov, Russ. Chem. Bull.,
1997, 46, 1792.
26 A. Adams and R. Slack, J. Chem. Soc., 1959, 3061.
27 H. D. Krebs, Aust. J. Chem., 1989, 42, 1291.
28 R. N. Butler, Chem. Rev., 1975, 75, 241.
41 S. R. Baker, A. F. Parsons and M. Wilson, Tetrahedron Lett., 1998, 39,
331.
42 C. DeMilt and G. V. Zandt, J. Org. Chem., 1936, 58, 2044.
43 L. Friedman and J. F. Chlebowski, J. Org. Chem., 1968, 33, 1636.
44 C. Meyers, B. U. W. Maes, K. T. J. Loones, G. Bal, G. L. F. Lemie`re
and R. A. Dommisse, J. Org. Chem., 2004, 69, 6010.
45 S. Hu¨nig, G. Kießlich and H. Quast, Liebigs Ann. Chem., 1971, 748,
201.
29 J. Goerdeler and M. Roegler, Chem. Ber., 1970, 103, 112.
30 M. P. L. Caton, D. H. Jones, R. Slack and K. R. H. Wooldridge,
J. Chem. Soc., 1964, 446.
31 L. Assmann, Y. Kitagawa, K. Ishikawa, D. Yamazaki, H. Sawada,
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 3681–3693 | 3693
©