An efficient one-pot synthesis of 3-substituted-5-amino-1,2,4-thiadiazoles from isothiocyanates and amidines

Article abstract of DOI:10.1016/j.tetlet.2008.03.030

An efficient one-pot synthesis of 3-substituted-5-amino-1,2,4-thiadiazoles from isothiocyanates and amidines is described.

Full text of DOI:10.1016/j.tetlet.2008.03.030

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 2869–2871
An efficient one-pot synthesis of 3-substituted-5-amino-
1,2,4-thiadiazoles from isothiocyanates and amidines
Yong-Jin Wu *, Yunhui Zhang
Department of Neuroscience Chemistry, Bristol-Myers Squibb Research and Development, 5 Research Parkway, Wallingford, CT 06492, USA
Received 19 February 2008; accepted 5 March 2008
Available online 10 March 2008
This Letter is dedicated to Professor E. J. Corey on the occasion of his 80th birthday
Abstract
An efficient one-pot synthesis of 3-substituted-5-amino-1,2,4-thiadiazoles from isothiocyanates and amidines is described.
Ó 2008 Elsevier Ltd. All rights reserved.
In a recent medicinal chemistry program, we desired
easy access to a series of 3-substituted-5-amino-1,2,4-thia-
diazoles 1. These compounds have been reported to have
a broad spectrum of biological activity including antibacte-
rial,1 anti-inflammatory,2 antiulcerative,3 and antidiabetic.4
Even though many methodologies exist for the preparation
of thiadiazoles 1,5 only two of them meet our basic criteria
for high speed analoging (Scheme 1). The first approach
involves the amination of N0-carbamothioyl-N,N-dimethyl-
(MSH) or hydroxyamine-O-sulfonic acid (HSA) followed
by cyclization.6 Amidines 2 are made from thioureas and
N,N-dimethylalkanamide dimethyl acetals 3 by condensa-
tion. As acetals 3 are not easily accessible with the excep-
tion of a few simple ones (R1 = H, Me), this method has
limited application for our purpose. The other route is
based on the oxidative heterocyclization of thioacylguan-
idines 4.5 Typical oxidizing agents are bromine, NCS,
NBS, iodine, nitric acid, acidic hydrogen peroxide, and
arylsulfonyl halides in the presence of pyridine. Unfortu-
nately, these reagents suffer from undesirable side reactions
such as halogenation and oxidative degradation. More
recently, Furukawa and co-workers carried out S–N
heterocyclization using diethyl azodicarboxylate (DEAD)
under mild and neutral conditions.7 However, there are
only three relatively simple 3-phenyl-N-aryl-1,2,4-thia-
diazol-5-amines described in the report, and the yields
are moderate (40–63%). To our knowledge, there have been
no other reports applying this methodology to the synthesis
of thiadiazoles. Therefore, in the outset of our medicinal
chemistry program, we sought to investigate the scope of
this DEAD-induced heterocyclization and also improve
the yield of this reaction. This report describes the develop-
ment of an efficient one-pot synthesis of 1,2,4-thiadiazoles
1 from isothiocyanates and amidines.
amidines
2 with O-(mesitylenesulfonyl)hydroxylamine
S
NMe2
R1
S
MeO OMe
R2
R2
+
R1
NMe2
N
N
N
NH2
H
H
2
3
MSH or HSA
N
S
R1
S
NH
R1 or DIAD
R2NCS
NH
NH2
DEAD
R2
N
R1
•HCl
N
N
H
R2
N
Hünig's
base
H
H
1
4
Scheme 1. Two representative approaches to 3-substituted-5-amino-1,2,4-
thiadiazoles.
*
The orginal procedure for the thiadiazole formation
from thioacylguanidines 4 requires 2 equiv of DEAD and
Corresponding author. Tel.: +1 203 677 7485; fax: +1 203 677 7702.
E-mail address: yong-jin.wu@bms.com (Y.-J. Wu).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.03.030

2870
Y.-J. Wu, Y. Zhang / Tetrahedron Letters 49 (2008) 2869–2871
ethanol as a solvent.7 We have recently re-examined this
reaction by using both diisopropyl azodicarboxylate
(DIAD) and DEAD, and by varying solvent, concentra-
tion, and the ratio of the reagent versus the substrate.
Our studies showed that DMF is a more suitable solvent
than ethanol because the solubility of some substrates in
ethanol is limited. In addition, we obtained slightly
improved yields with DIAD over DEAD (vide infra), and
for this reason, we applied DIAD throughout this method-
ology. We also found that 1.1 equiv of DIAD is sufficient
to bring about complete cyclization. In several instances
where DIAD was added in large excess (e.g., 2 equiv),
removal of the excessive reagent from the cyclized products
proved to be difficult and tedious by silica gel chromato-
graphy. Thus, our optimized conditions include 1.1 equiv
of DIAD, 0.80 M concentration, and DMF as a solvent.
In all cases studied, the cyclization was complete within
5–30 min as shown by TLC or LC/MS analysis. As a rule
of thumb, the disappearance of the orange color of DIAD
indicates the completion of the cyclization.
amate (entry 13) is caused by the incomplete conversion
in the thioacylguanidine formation step. Another feature
of this reaction is that the nature of the substituents on
the phenyl isothiocyanates has little influence on the overall
yield. For example, good yields of thiadiazoles were
obtained from phenyl isothiocyanates bearing both
Table 1
3-Substituted-5-amino-1,2,4-thiadiazoles from amidines and isothio-
cyanates8
1.R2NCS (1 equiv),
N
R1
NH
S
Hünig base(1.1 equiv)
2. DIAD (1.1 equiv)
R2
•HCl
NH2
N
R1
R2
N
1
H
Entry
R1
Yielda,b (%)
1
2
3
4
5
6
Ph
Ph
Ph
Ph
Ph
Ph
Ph
82
79
93
88
92
97
2-Bromophenyl
4-Bromophenyl
2,6-Dichlorophenyl
4-Nitrophenyl
2-Naphthyl
With the conditions for the cyclization established, we
turned our attention to the synthesis of the cyclization pre-
cursors, thioacylguanidines 4. In most cases, the addition
of isothiocyanates with amidine hydrochlorides in the pres-
7
Ph
86
CO2Et
ence of Hunig’s base in DMF proceeded cleanly as shown
¨
O
by LC/MS and TLC analyses. However, the isolated yields
of thioacylguanidines were moderate due to loss during
work-up and subsequent purification as they are quite
insoluble in many solvent systems. To this end, it was
logical to determine if 1,2,4-thiadiazoles 1 can be prepared
from isothiocyanates and amidines without the isolation of
the intermediate thioacylguanidines 4. Indeed, the treat-
ment of isothiocyanatobenzene (1.0 equiv) with benz-
8
Ph
92
O
9
Ph
Ph
68
96
NMe2
SMe
10
imidamide hydrochloride (1.0 equiv) and Hunig’s base
¨
(1.1 equiv) in DMF at room temperature for 12 h followed
by the addition of DEAD (1.1 equiv) generated N,3-diphe-
nyl-1,2,4-thiadiazol-5-amine in 75% yield. When the same
reaction was carried out using DIAD instead of DEAD,
the yield was increased to 82% (entry 1). This is a signifi-
cant improvement over the 63% yield obtained from N-
(phenylcarbamothioyl)benzimidamide (4, R1 = R2 = Ph)
using 2 equiv of DEAD in ethanol as disclosed in the origi-
nal report. The one-pot conditions worked well for both
aryl and alkyl isothiocyanates (entries 1–13). The steric
hindrance of aryl isothiocyanates does not exert significant
impact on the yield of this reaction. For example, 2-bromo-
phenyl and 2,6-dichlorophenyl isothiocyanates gave good
yields of thiadiazoles (entries 2 and 4). However, steric
effects may be important in the case of alkyl isothiocya-
nates. Thus, cyclohexyl isothiocyanate gave an 86% yield
of the cyclized product (entry 12), but in contrast, no thia-
diazole formation was observed in the case of 1-adamantyl
isothiocyanate (entry 14). The failure with the latter is due
to the thioacylguanidine formation step as 1-adamantyl
isothiocyanate is inert to the addition of benzimidamide
hydrochloride under current conditions. The moderate
yield obtained with tert-butyl 6-isothiocyanatohexylcarb-
O
Me
11
Ph
74
12
13
14
Ph
Ph
Ph
Cyclohexyl
BocNH(CH2)6
1-Adamantyl
86
52
0
15
16
17
2-Bromophenyl
2-Bromophenyl
2-Bromophenyl
95
82
78
N
N
Ph
18
19
20
21
22
a
Cyclopropyl
t-Butyl
1-Adamantyl
Methyl
2-Bromophenyl
2-Bromophenyl
2-Bromophenyl
2-Bromophenyl
2-Bromophenyl
56
77
42
0
CH2C(O)NH2
0
Isolated yield.
Yields not optimized.
b

Y.-J. Wu, Y. Zhang / Tetrahedron Letters 49 (2008) 2869–2871
2871
electron-donating (entry 8 and 9) and electron-withdrawing
groups (entries 7 and 11). Perhaps the most important
feature is that ester (entry 7), ketone (entry 11), tertiary
amine (entry 9), sulfide (entry 10), and even the acid-sensi-
tive acetal (entry 8) and NHBoc (entry 13) functional
groups remain intact during oxidative heterocyclization
(entries 7 and 8) (Table 1).
idines 4 presumably underwent spontaneous oxygen-
induced S–N heterocyclization. Future efforts will be direc-
ted toward understanding the mechanism of this novel
reaction and exploring the feasibility of oxidative cycli-
zation of thioacylguanidines 4 using oxygen.
In summary, we have developed an efficient one-pot
synthesis of 3-substituted-5-amino-1,2,4-thiadiazoles from
isothiocyanates and amidines. This method has been
successfully applied to the synthesis of biologically active
thiadiazoles bearing electron-rich phenyl and heterocycles,
and these compounds will be reported in due course. Tak-
ing into account the wide variety of isothiocyanates and
amidines either commercially available or easily accessible
plus the compatibility of various functional groups with
the current methodology, we anticipate widespread appli-
cation of this operationally simple procedure to the synthe-
sis of 3-substituted-5-amino-1,2,4-thiadiazoles.
In addition to benzimidamide, we also evaluated hetero-
aryl and alkyl amidines in the reaction with 1-bromo-2-
isothiocyanatobenzene (entries 15–22). Both 2- and 4-pyr-
idyl amidines gave excellent yields of thiadiazoles (entries
15–16). The steric impact from amidines appears to be less
pronounced than that from the alkyl isothiocyanates. For
example, both 1-phenylcyclopropanecarboximidamide
(entry 17) and pivalimidamide (entry 19) furnished good
yields of thiadiazoles. Of particular interest is that the reac-
tion of 1-adamantane-carboximidamide with 1-bromo-2-
isothiocyanatobenzene proceeded in 42% yield (entry 20),
while 1-admantyl isothiocyanate failed to react with iso-
thiocyanatobenzene (entry 14). The moderate yield
obtained with 1-adamantane-carboximidamide results
from the incomplete conversion in the thioacylguanidine
step (ca. 50%). Also, an unidentified side product was
observed during the addition of cyclopropylamidine with
1-bromo-2-isothiocyanatobenzene, and as a result, the
yield was moderate (entry 18). Finally, acetimidamide
(entry 21) and 3-amino-3-iminopropanamide (entry 22)
did not work as both failed to produce the desired thioacyl-
guanidines for cyclization.
Acknowledgments
We thank Drs. Lorin Thompson and John Macor for
their support and Dr. Richard Hartz for critical reading
of this report.
References and notes
1. Yamanaka, T.; Ohki, H.; Ohgaki, M.; Okuda, S.; Toda, A.; Kawabata,
K.; Inoue, S.; Misumi, K.; Itoh, K.; Satoh, K. U.S. Pat. Appl. Publ.
US2005004094A1.
2. Boschelli, D. H.; Connor, D.T. U.S. Pat. Appl. Publ. US005114958A.
3. Kharimian, K.; Tam, T. F.; Leung-Toung, R. C.; Li, W. PCT Int.
Appl. WO9951584A1.
In general, isothiocyanate reacts with amidine hydro-
chloride in the presence of Hunig’s base to give thioacyl-
¨
4. Johnstone, C.; Mckerrecher, D.; Pike, K. G.; Waring, M.J. PCT Int.
Appl. WO2005121110A1.
5. Franz, J. E.; Dhingra, O. P. In Comprehensive Heterocyclic Chemistry;
Potts, T. P., Ed.; 1,2,4-Thiadiazoles; Pergamon: New York, 1984; Vol.
6, Chapter 4.25, pp 463–511.
6. Lin, L.; Lang, S. A.; Petty, S. R. J. Org. Chem. 1980, 45, 3750.
7. Kihara, Y.; Kabashima, S.; Uno, K.; Okawara, T.; Yamasaki, T.;
Furukawa, M. Synthesis 1990, 1020.
guanidine 4 as the sole product. However, in three
instances, we have observed the formation of the cyclized
thiadiazoles in appreciable amounts even before the addi-
tion of DIAD. The ratio of the cyclized product 1 to
adduct 4 ranges from 1:5 to 1:2 (Table 2). Since our reac-
tions were carried out in the presence of air, thioacylguan-
8. Representative procedure: To a solution of benzamidine hydrochloride
(50 mg, 0.32 mmol) and 1-bromo-2-isothiocyanatobenzene (68 mg,
0.32 mmol) in DMF (0.40 mL) in an open vial at room temperature
Table 2
3-Substituted-5-amino-1,2,4-thiadiazoles from amidines and isothiocya-
nates without DIAD
was added Hunig’s base (0.35 mmol, 61 lL), the vial was capped, and
¨
the reaction mixture was stirred at room temperature for 12 h. DIAD
(0.35 mmol, 69 lL) was added, and the reaction mixture was stirred at
room temperature for 1 h. DMF was removed in vacuo, and the
residue was purified by preparative TLC eluting with 50% ethyl
acetate/50% hexanes to give N-(2-bromophenyl)-3-phenyl-1,2,4-thi-
adiazol-5-amine as a white solid (84 mg, 79%). 1H NMR (CDCl3,
400 MHz) d 7.02 (1H, dt, J = 1.2, 7.6 Hz), 7.41–7.48 (4H, m), 7.63 (1H,
dd, J = 1.6, 8.4 Hz), 7.78 (1H, dd, J = 1.6, 8.4 Hz), 8.16 (1H, br s),
8.20–8.25 (2H, m). 13C NMR (CDCl3, 100 MHz) d (attached H’s)
179.62 (0), 169.74 (0), 137.04 (0), 133.28 (1), 132.84 (0), 130.27 (1),
128.99 (1), 128.60 (2C, 1), 128.10 (2C, 1), 124.92 (1), 118.20 (1), and
113.41 (0). Exact mass calcd for C14H11N379Br32S (M+H): 331.9852;
found: 331.9849; C14H11N381Br32S (M+H): 333.9831; found: 333.9825.
N
R1
S
NH
R1
R2NCS
NH
NH2
S
R2
+
N
R1
•HCl
N
N
R2
N
Hünig's
base
H
H
H
4
1
Entry
R1
R2
Ph
4 : 1a
1
2
3
a
Ph
t-Bu
Ph
5 : 1
2 : 1
4 : 1
2-Bromophenyl
2-Naphthyl
LC/MS ratio of the crude reaction mixture prior to the addition of
DIAD.