Direct preparation of heteroaromatic compounds from alkenes

Article abstract of DOI:10.1055/s-0030-1259034

A series of aromatic heterocycles, thiazoles, imidazoles, and dimethoxyindoles, can be synthesised directly from alkenes via a ketoiodination-cyclisation protocol. The alkene starting materials are themselves easily accessible by many different and well-established approaches, and allow access to various aromatic heterocycles with excellent yields and regioselectivity. Georg Thieme Verlag Stuttgart - New York.

Full text of DOI:10.1055/s-0030-1259034

2956
LETTER
Direct Preparation of Heteroaromatic Compounds from Alkenes
D
irec
t
Preparati
i
on of
H
m
e
teroaromatic Com
o
pounds from
t
A
lken
h
es y J. Donohoe,*a Mikhail A. Kabeshov,a,b Akshat H. Rathi,a Ian E. D. Smithb
a
Department of Chemistry, University of Oxford, Chemistry Reasearch Laboratory, Mansfield Road, Oxford, OX1 3TA, UK
Fax +44(1865)275674; E-mail: timothy.donohoe@chem.ox.ac.uk
GlaxoSmithKline Research and Development Limited, Medicines Research Centre, Gunnels Wood Road, Stevenage, SG1 2NY, UK
Fax +44(1438)764502; E-mail: ian.e.smith@gsk.com
b
Received 11 October 2010
via iodo ketones. Scheme 1 shows the general plan, which
involves ketoiodination of an alkene, followed by dis-
placement and condensation to form a heterocycle. In ad-
dition to testing the general viability of this sequence, we
Abstract: A series of aromatic heterocycles, thiazoles, imidazoles,
and dimethoxyindoles, can be synthesised directly from alkenes via
a ketoiodination–cyclisation protocol. The alkene starting materials
are themselves easily accessible by many different and well-estab-
lished approaches, and allow access to various aromatic hetero- also wanted to explore the range of functionality that
cycles with excellent yields and regioselectivity.
could be incorporated into the final heterocycle, as well as
addressing issues of regiochemistry that arise from the ox-
Key words: heterocycles, alkenes, iodine, iodo ketones, oxidation
idation of unsymmetrical alkenes.
X
R3
A large proportion of modern pharmaceutical chemistry
centres on aromatic heterocyclic compounds and this
makes quick, general, and efficient methods for their syn-
thesis extremely valuable. Amongst precursors to het-
eroaromatic compounds, a-halo- and tosyloxy ketones are
important building blocks. For example, the Hantzsch thi-
azole synthesis from an a-halo ketone and a thioamide has
been well known since the late 19th century,1 and it is also
possible to convert alcohols2 and ketones,3 through a-
tosyloxy ketones, into a series of thiazoles, imidazoles,
and imidazopyridines. Similarly, imidazoles can be ob-
tained by employing a condensation between a-halo ke-
tones and amidines,4 and dimethoxyindoles can be
accessed from a-halo ketones using Bischler and modi-
fied Bischler methododology.5
R3
O
2
H2N
(3 equiv)
4
I
2 (1.1 equiv)
X
R2
N
R2
IBX (2 equiv)
R1
R1
R2
DMSO, 25 °C
10 min to 3.5 h
25 °C, 2–20 h
R1
I
1
3
Scheme 1 General plan for the synthesis of aromatic heterocycles 3
from alkenes 1 via iodo ketones 2
In our initial experiments we carried out the ketoiodina-
tion of 2-methylstyrene using a procedure developed by
Moorthy et al.10b (1.1 equiv NIS, 2 equiv IBX, r.t.,
DMSO). The reaction proceeded smoothly with t1/2 = 10
minutes.12 After the excess of NIS and IBX were washed
out with aqueous NaHCO3–Na2S2O3, the resulting 2-iodo-
1-phenyl-1-propanone13 was reacted with thiourea (3
equiv, 25 °C, DMSO–DMF) to produce 5-methyl-4-phe-
nyl-1,3-thiazol-2-amine (3a) as the only detectable prod-
uct and as a single regioisomer in 71% isolated yield.
However, we found that replacing NIS with I2 (1.1 equiv)
led to a substantial increase of the reaction rate: t1/2 = 2
min and full consumption of the starting material within
10 minutes. After condensation of the iodo ketone with
thiourea the aminothiazole 3a was isolated with a higher
yield of 74% (Figure 1). Generally, we found the use of
iodine/IBX superior and so used this protocol throughout.
As expected, iodo ketones can be prepared from ketones6
or their enolates7 by iodination. However, a direct route to
iodo ketones from alkenes (ketoiodination) is potentially
a more potent and powerful method. Indeed, such a reac-
tion was discovered by Cardillo et al.8 who used the sil-
ver(I) chromate/iodine pair, although this proved to be an
expensive method. Later, bis(symcollidine)iodine(I) tet-
rafluoroborate in DMSO was used as a reagent,9 and re-
cently IBX has been introduced as an oxidant for the
10a
ketoiodination of alkenes when paired with I2 and
NIS.10b
These conditions were successfully applied to the synthe-
ses of various thiazoles (3a–i, Figure 1) utilising different
mono- and disubstituted alkenes as starting materials. The
presence of a thiazole, pyridine, dimethylaminopyridine,
or an a-methoxy group in the structure of the starting alk-
ene 1 does not affect the efficiency of the method (3d–f),
which can be applied to aryl-alkyl disubstituted (3a,b,d–
f), aryl monosubstituted (3c,h,i) and alkyl-alkyl disubsti-
tuted alkenes (3g). Although regioselective heterocycle
formation was not possible with unsymmetrical dialkyl
substituted alkenes, it quickly emerged that the presence
of an aryl group on the starting alkene led to the formation
We observed that, compared to their chloro and bromo an-
alogues, iodo ketones have been rarely employed as elec-
trophilic partners in aromatic heterocycle synthesis.11
Although it is well known that iodide is easily displaced
in nucleophilic substitution reactions, iodo ketones have
not been used because of their limited availability. Herein,
we report an efficient and succinct method for synthesis-
ing various heterocycles from readily available alkenes
SYNLETT 2010, No. 19, pp 2956–2958
x
x
.x
x
.2
0
1
0
Advanced online publication: 10.11.2010
DOI: 10.1055/s-0030-1259034; Art ID: D27010ST
© Georg Thieme Verlag Stuttgart · New York

LETTER
Direct Preparation of Heteroaromatic Compounds from Alkenes
2957
of a single regioisomer of the heterocyclic product, as
shown. Moreover, in addition to thiourea, we showed that
alkyl thioamides (3h) or aryl thioamides (3i) can be used
as nucleophiles, in so doing extending the variation that is
possible at R3 (see 3, 4, Scheme 1). Note that iodo ketones
2 usually react with nucleophiles 4 faster than their chloro
and bromo analogues, and thus it is possible to synthesise
the majority of heterocycles at room temperature whereas
most of the other methods described in the literature em-
ploy chloro or bromo ketones at elevated tempera-
tures.14,15
N
Pr
SEt
N
HN
HN
HN
N
N
N
N
3j
3k
67%
3l
71%
3m
72%
72%
OMe
Ph
N
HN
MeO
MeO
NH
3o
63%a
3n
70%
Figure 2 Synthesis of imidazoles and indoles 3 from alkenes 1 via
iodo ketones 2. Yields are based on the corresponding alkene 1; a con-
densation with dimethoxyaniline was performed at 80 °C.
were grown and subjected to X-ray diffraction (Figure 1).
Compounds 3d,e,i were assigned by analogy and 3l by
HMBC experiments.
Our own studies on the mechanism of the reaction suggest
that the initial ketoiodination produces the regioisomer in
which the carbonyl is conjugated with the unsaturated
substituent, in accordance with literature precedent. Ensu-
ing substitution of the iodide by the most nucleophilic
atom of the corresponding nucleophile, followed by con-
densation would explain the regiochemical outcome of
the sequence.
In summary, we have demonstrated that a series of hetero-
cycles, here thiazoles, imidazoles, and indoles, can be
synthesised directly from alkenes using an efficient and
succinct method. This methodology enables the prepara-
tion of heterocycles having multiple points of diversity
whilst controlling the regiochemistry of unsymmetrical
heterocyclic products when one substituent on the alkene
is aromatic.
Figure 1 2-Aminothiazoles made via ketoiodination and condensa-
tion with thiourea and thioamides. X-ray structures of 3b and 3f are
shown. Yields are based on the starting alkene.
By employing amidines or their salts as nucleophiles in
this sequence, one can access the corresponding imida-
zoles 3j–n (Figure 2). Thus, the method is applicable with
alkyl amidines (3j,l), aryl amidines (3k), mono-N-substi-
tuted amidines (3l), S-alkylated pseudothioureas (3m), in
conjunction with mono- (3j–m) or disubstituted (3n) al-
kenes. As a final example to show the versatility of this
method, the dimethoxyindole (3o) was synthesised by
using 3,5-dimethoxyaniline as a nucleophile (4) by a
Bishler-type method.5
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synlett.
Acknowledgment
We would like to thank the EPSRC and GlaxoSmithKline for finan-
cial support.
For compounds 3j,k,m,n the issue of regioselectivity does
not arise because of imidazole tautomerisation. However,
compounds 3l and 3o were formed as single regioisomers,
as shown.
References and Notes
(1) (a) Alajarin, M.; Cabrera, J.; Pastor, A.; Sanchez-Anrada, P.;
Bautista, D. J. Org. Chem. 2006, 71, 5328. (b) Schwarz, G.
Org. Synth., Coll. Vol. III 1955, 332. (c) Hantzsch, A. Ann.
Chem. 1889, 250, 257.
The structures of the aromatic products 3 were proven by
correlation with literature data for 3a,c,h,o (see Support-
ing Information). To establish unambigously the regio-
chemistry of the sequence,16 single crystals of 3b and 3f
(2) Ueno, M.; Nabana, T.; Togo, H. J. Org. Chem. 2003, 68,
6424.
Synlett 2010, No. 19, 2956–2958 © Thieme Stuttgart · New York

2958
T. J. Donohoe et al.
LETTER
(12) 0.25 M solution, analysed by LCMS calibrated with
(3) John, O. R. S.; Killeen, N. M.; Knowles, D. A.; Yau, S. C.;
Bagley, M. C.; Tomkinson, N. C. O. Org. Lett. 2007, 9,
4009.
(4) (a) Li, B.; Chiu, C. K. F.; Hank, R. F.; Murry, J.; Roth, J.;
Tobiassen, H. Org. Process Res. Dev. 2002, 6, 682.
(b) Kempter, G.; Spindler, J.; Fiebig, H. J.; Sarodnick, G.
J. Prakt. Chem. 1971, 313, 977.
(5) (a) Bischler, A. Ber. Dtsch. Chem. Ges. 1892, 25, 2860.
(b) Pchalek, K.; Jones, A. W.; Wekking, M. M. T.; Black,
D. S. C. Tetrahedron 2005, 61, 77.
nitrobenzene; see Supporting Information for details.
(13) Analytical Data for Crude 2-Iodo-1-phenyl-1-propanone
1H NMR (400 MHz, CDCl3): d = 2.04 (d, 3JCH3,CH = 7.0 Hz,
3 H, CH3), 5.47 (q, 3JCH3,CH = 7.0 Hz, 1 H, CHI), 7.40–7.46
(m, 2 H, Ph), 7.50–7.56 (m, 1 H), 7.92–7.98 (m, 2 H, Ph) are
in agreement with the literature: Cambie, R. C.; Hayward,
R. C.; Lindsay, B. G.; Phan, A. L. T.; Rutledge, P. S.;
Woodgate, P. D. J. Chem. Soc., Perkin Trans. 1 1976, 1961.
(14) (a) Cai, L.; Brouwer, C.; Sinclair, K.; Cuevas, J.; Pike, V. W.
Synthesis 2006, 133. (b) Hirano, K.; Urban, S.; Wang, C.;
Glorius, F. Org. Lett. 2009, 11, 1019. (c) Bailey, N.;
Bamford, M. J.; Brissy, D.; Brookfield, J.; Demont, E.;
Elliott, R.; Garton, N.; Farre-Gutierrez, I.; Heyhow, T.;
Hutley, G.; Neylor, A.; Panchal, T. A.; Seow, H.-X.;
Spalding, D.; Takle, A. K. Bioorg. Med. Chem. Lett. 2009,
19, 3602.
(15) In the reaction of 2-chloro-1-phenylpropan-1-one with
benzamidine in the presence of K2CO3 at r.t. in DMF, only
traces of the corresponding imidazole were detected after
2 h.
(16) To distinguish between two isomeric aminothiazoles: 5-
propyl-4-(3-pyridinyl)-1,3-thiazol-2-amine and 4-propyl-5-
(3-pyridinyl)-1,3-thiazol-2-amine.
(6) Jereb, M.; Stavber, S.; Zupan, M. Tetrahedron 2003, 59,
5935.
(7) (a) Cort, A. D. J. Org. Chem. 1991, 56, 6708.
(b) De Dobbeleer, C.; Pospiil, J.; Marko, I. E.;
De Vleeschouwer, F.; De Proft, F. Chem. Commun. 2009,
2142.
(8) (a) Cardillo, G.; Shimizu, M. J. Org. Chem. 1977, 42, 4268.
(b) Shamsuzzaman, S.; Anwar, A.; Suhail, S. Synth.
Commun. 1997, 27, 3997.
(9) Evans, R. D.; Herman, J. Synthesis 1986, 727.
(10) (a) Yadav, J. S.; Reddy, B. V. S.; Singh, A. P.; Basak, A. K.
Tetrahedron Lett. 2008, 49, 5880. (b) Moorthy, J. N.;
Senapati, K.; Singhal, N. Tetrahedron Lett. 2009, 50, 2493.
(11) See: Basarab, G.; Hill, P.; Zhou, F. WO 2008152418, A1
20081218, 2008.
Synlett 2010, No. 19, 2956–2958 © Thieme Stuttgart · New York