Highly regioselective direct halogenation: A simple and efficient method for preparing 4-halomethyl-5-methyl-2-aryl-1,3-thiazoles

Article abstract of DOI:10.1016/j.tetlet.2003.10.113

An unprecedented C4-methyl regioselective halogenation of 4,5-dimethyl-2-aryl-1,3-thiazoles (1) has been accomplished. The reaction of compound 1 with N-chlorosuccinimide and N-bromosuccinimide under mild conditions provides an efficient and operationally simple method for obtaining 4-chloromethyl-5-methyl-2-aryl-1,3-thiazoles (2) and 4-bromomethyl-5-methyl-2- aryl-1,3-thiazoles (3), respectively, in good yields without the formation of 4-methyl-5-halomethyl regioisomers.

Full text of DOI:10.1016/j.tetlet.2003.10.113

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 69–73
Highly regioselective direct halogenation: a simple and
efficient method for preparing 4-halomethyl-5-methyl-2-
aryl-1,3-thiazoles
Taihei Yamane,* Hiroyuki Mitsudera and Takatsugu Shundoh
Chemical Development Laboratories, Pharmaceutical Production Division, Takeda Chemical Industries Ltd, Osaka 532-8686, Japan
Received 8 September 2003;revised 16 October 2003;accepted 22 October 2003
Abstract—An unprecedented C4-methyl regioselective halogenation of 4,5-dimethyl-2-aryl-1,3-thiazoles (1) has been accomplished.
The reaction of compound 1 with N-chlorosuccinimide and N-bromosuccinimide under mild conditions provides an efficient and
operationally simple method for obtaining 4-chloromethyl-5-methyl-2-aryl-1,3-thiazoles (2) and 4-bromomethyl-5-methyl-2-aryl-
1,3-thiazoles (3), respectively, in good yields without the formation of 4-methyl-5-halomethyl regioisomers.
ꢀ 2003 Elsevier Ltd. All rights reserved.
S
Substituted 1,3-thiazoles (also known simply as thia-
zoles) are ubiquitous structural motifs found in com-
pounds of biological interest. As a result, a number of
methods have been developed to synthesize these com-
pounds.1;2 Despite the accessibility of these methods, the
ability to approach certain specific synthetic paths is
limited and requires the design of novel synthetic strat-
egies. Particular examples of these are 4-chloromethyl-
and 4-bromomethyl-5-methyl-2-arylthiazoles, which are
able to serve as key intermediates in the synthesis of
biologically interesting and pharmaceutically useful
molecules as dipeptidyl peptidase IV (DPP-IV) inhibi-
tors, agonists of peroxisome proliferation-activated
receptors (PPARs), subtype selective N-methyl-
NH2
O
O
O
CO2Et
CO2H
CO2Et
N
Br
CO2Et
OH
X
N
S
N
S
S
2 (X = Cl)
3 (X = Br)
Scheme 1. Conventional method for preparing 2 and 3.
D
-aspartate antagonists, glucose and lipid lowering
agents and antifungal agents.3
rearrangement.6 However, the former method failed
only to give an undesired isomer exclusively and the
latter method gave only a minor amount of desired
compound in three reaction sequences.
Previously, these compounds have been prepared via
ethyl 5-methyl-2-phenylthiazole-4-carboxylates,4 which
were synthesized by coupling reactions between thio-
benzamides and ethyl 3-bromo-4-methyl-2-oxobutano-
ate (Scheme 1).5 As a preliminary study, we have tried
some synthetic methods including a direct synthesis in-
volving a coupling reaction between thiobenzamide and
1,3-dibromo-2-butanone or a chlorination of 4,5-di-
methyl-2-phenylthiazole N-oxide via a Pummerer-type
In our efforts aimed at the development of new methods
of synthesizing 4-chloromethyl- and 4-bromomethyl-5-
methyl-2-phenylthiazole, we were interested in a direct
halogenation of 4,5-dimethyl-2-phenylthiazole, which
can be readily prepared from inexpensive thiobenzamide
and 2-chlorobutanone,7 reacted with NCS or NBS as an
efficient and useful alternative to a known method
(Scheme 2).
Keywords: thiazole;halogenation;regioselectivity.
* Corresponding author. Fax: +81-663006251;e-mail: yamane_taihei@
takeda.co.jp
Prior to our study, Mohanazadeh reported that a radical
halogenation of 4,5-dimethyl-2-phenylthiazole with
0040-4039/$ - see front matter ꢀ 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.10.113

70
T. Yamane et al. / Tetrahedron Letters 45 (2004) 69–73
and 4-methyl-5-chloromethyl-2-phenylthiazole prepared
X
according to a known procedure10 as a comparison
sample by 1H NMR, 13C NMR11 and HPLC analyses.12
The reaction also proceeded without AIBN in the ab-
sence of light. The result suggests that the chlorination
does not proceed via a radical process. It is of note that
the regioselectivity of the reaction mixture and isolated
compound was considerably high as ascertained by the
HPLC analysis (>99% regioselectivity). The results show
that the chlorination exclusively proceeds at the C4-
methyl site and no 4-methyl-5-chloromethyl regioiso-
mers are obtained even in the reaction mixture.
N
N
S
S
1
2 (X = Cl)
3 (X = Br)
Scheme 2.
NCS using a 200 W light bulb proceeded without
selectivity and with NBS occurred C5-methyl site regio-
selectively.8 On the other hand, it has been reported that
a bromination of 2,3-dimethylbenzofuran derivative
with NBS proceeds regioselectively depending on con-
ditions, which allow a selective formation of either
2-chloromethyl derivative (reflux in CCl4) or 3-chloro-
methyl derivative (an ambient temperature in CH2Cl2).9
This prompted us to investigate a direct halogenation of
4,5-dimethyl-2-phenylthiazoles (1) under mild condi-
tions with NCS or NBS. In general, highly regioselective
reactions are required especially in pharmaceutical
fields, because even a small amount of regioisomers
cannot be removed easily from objective intermediates
or drug candidates by any means of purification quite
often due to similar physical properties. To the best of
our knowledge, there are no reported reactions pro-
ceeding at the C4-methyl site with high regioselectivity.
Herein we report a highly C4-methyl regioselective
synthesis of 4-chloromethyl-5-methyl-2-arylthiazoles (2)
and 4-bromomethyl-5-methyl-2-arylthiazoles (3) from 1.
Conveniently, 1ÆHCl could be isolated as a crystal13
directly from a reaction mixture of thiobenzamide and
2-chlorobutanone in 2-propanol (90% yield) (Scheme 3).
O
HCl
NH2
S
N
Cl
S
2-Propanol
reflux, 10h
1
90% Yield
Scheme 3. Preparation of 1.
With 1ÆHCl, the reaction was carried out via the one-pot
neutralization (of HCl)––chlorination sequence using
1 equiv of NCS,14 furnished 215 in a good yield (83%),
directly isolated as a crystal by adding water to the re-
action mixture (Table 1, entry 1) (>99% purity detected
by HPLC analysis as an isolated compound without
further purification).16
Initial attempts were carried out using 1, NCS
(1.2 equiv) and 2,20-azobis(isobutyronitrile) (AIBN;
0.1 equiv) in acetonitrile at 60 ꢁC. The reaction furnished
mono-chlorinated compound, that is 4-chloromethyl-5-
methyl-2-phenylthiazole (2) or 5-chloromethyl-4-methyl-
2-phenylthiazole (the regioisomer of 2). Surprisingly, the
chloride was confirmed as the desired 2 utilizing
the conventionally prepared 2 as an authentic sample
On the basis of this finding, we extended the scope of the
chlorination with a variety of different substituted phenyl
groups. The reaction worked well with satisfactory
yields that ranged from 57% to 89% for each derivative
with phenyl groups bearing electron-donating groups
Table 1. Chlorination of 4,5-dimethyl-2-arylthiazoles
1) Et3N (1equiv.)
HCl
N
Cl
N
Acetonitrile
60 oC, 2h
S
2) NCS (1 equiv.)
S
o
, 2h
60 C
2
1
R
R
Entry
Substratea (R ¼ )
Yield (%)b
Regioselectivity (%)c;d
1
2
3
4
5
6
7
8
H
83
78
89
57
85
75
73
87
>99 (>99)
>99
p-tBu
p-Me
>99 (>99)
>99 (>99)
>99 (>99)
>99
p-MeO
3,4-Methylenedioxy
p-F
p-Cl
o-Cl
>99
>99 (>99)
a Substrates were used as hydrochlorates.
b Yields refer to single runs and are given for isolated products.
c Regioselectivity was determined for isolated products by HPLC analysis (YMC Pack-SIL A-002 column with 40:1 n-hexane/THF as mobile phase).
d Regioselectivity in reaction mixtures determined by HPLC analysis (the same conditions for isolated products) are shown in parentheses.

T. Yamane et al. / Tetrahedron Letters 45 (2004) 69–73
Table 2. Bromination of 4,5-dimethyl-2-arylthiazoles
71
Br
N
N
NBS (1 equiv.)
Acetonitrile
25 oC,2h
S
S
3
1
R
R
Entry
Substratea (R ¼ )
Yield (%)b
Regioselectivity (%)c;d
1
2
3
4
5
6
7
8e
H
70
65
78
82
69
57
57
53
>99 (>99)
>99
p-tBu
p-Me
>99 (>99)
>99
p-MeO
3,4-Methylenedioxy
p-F
>99 (>99)
>99 (>99)
>99
p-Cl
o-Cl
––
a Substrates were used as free form.
b Yields refer to single runs and are given for isolated products.
c Regioselectivity was determined for isolated products by HPLC analysis (YMC Pack-SIL A-002 column with 40:1 n-hexane/THF as mobile phase).
d Regioselectivity in reaction mixtures determined by HPLC analysis (the same conditions for isolated products) are shown in parentheses.
e Yield was given for the product purified by column chromatography due to a difficulty of crystallization from the reaction mixture. The structure
was determined by 1H NMR, 13C NMR and HPLC analyses.
(entries 2–5), as well as electron-withdrawing groups
(entries 6–8).
Following the successful demonstration of the regio-
selective chlorination, we then proceeded to extend the
generality of the reaction to bromination with NBS,
with the hope that the reaction could proceed at the
C4-methyl site regioselectively. The first bromination
we carried out between 4,5-dimethyl-2-phenylthiazole
(1) and NBS (1 equiv) in acetonitrile at 25 ꢁC, and gave
a 70% yield of 3 successfully, isolated directly as a
crystal by adding water to the reaction mixture (Table
2, entry 1).17 The regioselectivity of the reaction mix-
ture and the isolated product were determined by
HPLC analysis (methods and conditions used were the
same as for the chlorination;a >99% regioselectivity
were ascertained) and 1H and 13C NMR by an
authentic sample of 3 and a comparison sample of the
regioisomer, as well.18
To examine the generality of this reaction further, we
proceeded to study a variety of different substituted
phenyl groups. The results are shown in Table 2.
Though the reactions proceeded at a lower temperature
(25 ꢁC) than the chlorination, almost the same reactivity
trends as the chlorination case were observed. The re-
Figure 1. X-ray structures of 2 (Table 1, entry 1) and 3 (Table 2,
entry 1).
actions afforded the desired 3 with yields from 53% to
82%. In addition, bromine can also be used instead of
NBS, however the yield (30%) is much lower than with
NBS, as determined by H NMR due to the low con-
version and formation of various unknown by-products.
ing on the reaction conditions. One possibility is that the
reaction might proceed with a mechanism via a Pum-
merer-type rearrangement20;21 under the mild reaction
condition (Scheme 4).
1
In summary, we have developed a highly regioselective
and efficient method that allows for synthesis of 4-
chloromethyl-5-methyl-2-arylthiazoles (2) and 4-bromo-
methyl-5-methyl-2-arylthiazoles (3) utilizing the
readily prepared 4,5-dimethyl-2-arylthiazoles (1) from
thiobenzamides and 2-chlorobutanone. This facile
process involves no complicated operation. Further
The structures of 2 (Table 1, entry 1) and 3 (Table 2,
entry 1) were unambiguously confirmed by X-ray crys-
tallography (Fig. 1).19
We have not determined the source of this C4-methyl
selectivity and the difference of regioselectivity depend-

72
T. Yamane et al. / Tetrahedron Letters 45 (2004) 69–73
Yotsutsuji, A.;Sakai, H. Jpn. Kokai Tokkyo Koho
O
O
JP62178590, 1987; Chem. Abstr. 1988, 108, 112450.
4. Cornwall, P.;Dell, C. P.;Knight, D. W. J. Chem. Soc.,
Perkin Trans. 1 1991, 2417–2428, and references cited
therein.
5. Okonya, J. F.;Hoffmann, R. V.;Johnson, M. S. J. Org.
Chem. 2002, 67, 1102–1108, and references cited therein.
6. (a) Goto, Y.;Yamazaki, M.;Hamana, M. Chem. Pharm.
Bull. 1971, 19, 2050–2057;(b) Dalley, N. K.;Petrie, C. R.,
III, Revankar, G. R.;Robins, R. K. Nucleos. Nucleot.
1985, 4, 651–659.
H
N
N
N
N
X
S
S
O
R
R
O
X
X
N
N
X
S
S
7. Vernin, G.;Aune, J. P.;Dou, H. J. M.;Metzger, J. Bull.
Soc. Chim. Fr. 1967, 4523–4533.
R
R
(X = Cl, Br)
8. Mohanazadeh, F. Int. J. Chem. 1995, 6, 73–77; Chem.
Abstr. 1996, 125, 221682.
Scheme 4. Proposed reaction mechanism.
9. Wollowitz, S.;Isaacs, S. T.;Rapoport, H.;Spielmann, H.
P.;Nerio, A. U.S. Patent 5,625,079, 1997.
10. (a) Chao, E. Y.;Haffner, C. D.;Lambert, M. H.;
Maloney, P. R.;Sierra, M. L.;Sternbach, D. D.;
Sznaidman, M. L.;Willson, T. M.;Xu, H. E.;Gellibert,
F. J. Patent WO 01/00603 A1, 2001; Chem. Abstr.
2001, 134, 86235;(b) Hubacher, K. Liebigs Ann. Chem.
application of these methodologies to other substrates
such as oxazoles is under investigation in the laboratory.
1890, 259, 228–253;(c) Simiti, I.;Muresan, A.
rev.
Roum. Chim. 1976, 21, 1073–1081; Chem. Abstr. 1997,
86, 29697.
11. The desired compound and its regioisomer are discrimi-
nated apparently by their differences of chemical shifts on
1H and 13C NMR.
12. HPLC conditions: YMC Pack-SIL A-002 column with
40:1 n-hexane/THF as mobile phase. To investigate
regioselectivity, we used HPLC analysis with the normal
phase. The 4-methyl-5-halomethyl regioisomers easily
decomposed during HPLC analysis with the reverse
phase.
Acknowledgements
We thank Ms. K. Higashikawa and Mr. A. Fujishima
for single-crystal X-ray structural analyses, and Dr.
Kiminori Tomimatsu and Mr. Mitsuhisa Yamano for
helpful discussions.
13. Compound 1 in a free form is liquid at an ambient
temperature.
References and Notes
14. General procedure for the preparation of compound 2. To a
solution of 4,5-dimethyl-2-phenyl-1,3-thiazole hydrochlo-
rate (1ÆHCl, 13.3 mmol) in acetonitrile (·10 v/w of 1ÆHCl),
triethylamine (13.3 mmol) was added dropwise at rt, and
the reaction mixture was stirred at 60 ꢁC for 2 h, then
allowed to cool to rt. To the mixture was added NCS
(13.3 mmol) and the reaction mixture was stirred at 60 ꢁC
for 2 h, then allowed to cool to rt. Water (·5–10 v/w of
1ÆHCl) was added to the reaction mixture dropwise at rt
to give colourless to pale yellow precipitation. The
mixture was stirred at 0–5 ꢁC for 1 h. The precipitation
was filtered and washed with ice-cooled acetonitrile–water
(1:1, v/v) (·2 v/w of 1ÆHCl) to afford the corresponding
chloride 2.
15. Data for 4-(chloromethyl)-5-methyl-2-phenyl-1,3-thiazole
(Table 1, entry 1). Colourless crystal: mp 86–88 ꢁC. 1H
NMR (CDCl3): d 2.53 (s, 3H), 4.72 (s, 2H), 7.39–7.44 (m,
3H), 7.87–7.90 (m, 2H). 13C NMR (CDCl3): d 11.4, 38.9,
126.3, 128.9, 129.9, 132.5, 133.4, 148.6, 164.7. IR (KBr,
cmÀ1): 1468, 1259, 764, 687, 654, 623. LRMS (EI): m=z
223 (Mþ). HRMS (EI): Calcd for C11H10ClNS (M):
223.0222. Found: 223.0257. Anal. Calcd for C11H10NSCl:
C, 59.05;H, 4.51;N, 6.26;S, 14.33;Cl, 15.85. Found: C,
58.95;H, 4.35;N, 6.26;S, 14.39;Cl, 15.81.
1. For reviews, see: (a) Dondoni, A.;Merino, P. In
Comprehensive Heterocyclic Chemistry II;Shinkai, I.,
Ed.;Katritzky, A. R., Rees, C. W., Scriven, E. F. V., Eds.
in Chief;Elsevier: Tarrytown, 1996;Vol. 3, pp 373–474;
(b) Metzger, J. V. In Comprehensive Heterocyclic Chem-
istry;Potts, K. T., Ed.;Katritzky, A. R., Rees, C. W.,
Eds. in Chief;Pergamon: Elmsford, 1984;Vol. 6, pp 235–
331;(c) Vernin, G. In The Chemistry of Heterocyclic
Compounds;Metzger, J. V., Ed.;Weissberger, A., Taylor,
E. C., Eds. in Chief;John Wiley & Sons: New York, 1979;
Vol. 34, Part 1.
2. (a) Al Hariri, M.;Jouve, K.;Pautet, F.;Domard, M.;
Fenet, B.;Fillion, H. J. Org. Chem. 1997, 62, 405–410;(b)
Al Hariri, M.;Galley, O.;Pautet, F.;Fillion, H.
Eur. J.
Org. Chem. 1998, 593–594;(c) Jouve, K.;Pautet, F.;
Domard, M.;Fillion, H. Chem. Pharm. Bull. 1999, 47,
1064–1067.
3. (a) Boehringer, M.;Hunziker, D.;Kuehne, H.;Loeffler,
B. M.;Sarabu, R.;Wessel, H. P. Patent WO 03/037327
A1, 2003; Chem. Abstr. 2003, 138, 368754;(b) Bach, A. T.;
Kapa, P. K.;Lee, G. T.;Loeser, E. M.;Sabio, M. L.;
Stanton, J. L., Vedananda, T. R. Patent WO 03/043985
A1, 2003; Chem. Abstr. 2003, 139, 6767;(c) Momose, Y.;
Maekawa, T.;Yamano, T.;Kawada, M.;Odaka, H.;
Ikeda, H.;Sohda, T. J. Med. Chem. 2002, 45, 1518–1534;
(d) Deorazio, R. J.;Nikam, S. S.;Scott, I. L.;Sherer, B.
A. European Patent EP 1 251 128 A1, 2002; Chem. Abstr.
2002, 137, 310908;(e) Momose, Y.;Maekawa, T.;Odaka,
H.;Ikeda, H.;Sohda, T. Chem. Pharm. Bull. 2002, 50,
100–111;(f) Takano, S.;Imaizumi, H.;Kajita, T.;
Takasima, K.;Takezawa, K.;Yotsutsuji, M.;Hoda, T.;
16. HPLC conditions: YMC Pack-ODS A-302 column with
1:1 0.05 M KH2PO4/acetonitrile as mobile phase.
17. General procedure for the preparation of compound 3. To a
solution
of
4,5-dimethyl-2-phenyl-1,3-thiazole
(1,
22.1 mmol) in acetonitrile (·10 v/w of 1) was added
NBS (22.1 mmol) and the reaction mixture was stirred at
25 ꢁC for 2 h. To the reaction mixture, water (·5–10 v/w of
1) was added dropwise at rt to give colourless to pale
yellow precipitation. The mixture was stirred at 0–5 ꢁC

T. Yamane et al. / Tetrahedron Letters 45 (2004) 69–73
73
for 1 h. The precipitation was filtered and washed
with water (·3 v/w of 1) to afford the corresponding
bromide 3.
555. LRMS (EI): m=z 267 (Mþ). HRMS (EI): Calcd for
C11H10BrNS (M): 266.9717. Found: 266.9739.
19. The deposition numbers at the Cambridge Crystallogra-
phy Data Centre, CCDC, are CCDC 216411 (for
compound 3) and 216412 (for compound 2).
20. Tuleen, D. L.;Stephens, T. B. J. Org. Chem. 1969, 34, 31–
35.
18. Data for 4-(bromomethyl)-5-methyl-2-phenyl-1,3-thiazole
1
(Table 2, entry 1). Pale yellow crystals: mp 90–92 ꢁC. H
NMR (CDCl3): d 2.49 (s, 3H), 4.62 (s, 2H), 7.40–7.44 (m,
3H), 7.86–7.89 (m, 2H). 13C NMR (CDCl3): d 11.6, 25.5,
126.4, 128.9, 130.1, 132.5, 133.1, 148.4, 164.7. IR (KBr,
cmÀ1): 1614, 1485, 1458, 1288, 1252, 1016, 766, 683, 606,
21. We do not find evidence of the electrophilic attack on S
atom.