An unexpected reaction of arenesulfonyl cyanides with allylic alcohols: Preparation of trisubstituted allyl sulfones

Article abstract of DOI:10.1002/anie.200803836

(Chemical Equation Presented) An efficient and practical protocol for the highly selective preparation of substituted allyl sulfones has been developed. Arenesulfonyl cyanides, Baylis-Hillman adducts, and simple allylic alcohols give an unforeseen outcome

Full text of DOI:10.1002/anie.200803836

Communications
DOI: 10.1002/anie.200803836
Synthetic Methods
An Unexpected Reaction of Arenesulfonyl Cyanides with Allylic
Alcohols: Preparation of Trisubstituted Allyl Sulfones**
Leleti Rajender Reddy,* Bin Hu,* Mahavir Prashad, and Kapa Prasad
For an ongoing program within our research group we needed
to synthesize sulfinates of the type 1. We reasoned that 1
could be easily prepared by a reaction of Baylis–
Hillman adduct 3 with p-toluenesulfonyl cyanide
(Scheme 1).[1] These reaction conditions, how-
ever, led to an unexpected trisubstituted allyl
sulfone 2. To the best of our knowledge, such a
reaction of a Baylis–Hillman adduct[2] with p-
toluenesulfonyl cyanide[1,3] to form substituted
allyl sulfones has not been reported. These types
of substituted allyl sulfones are important inter-
mediates in organic synthesis[4] and have been
recently found to be highly potent against cancer
and abnormal cell proliferation diseases.[5] The
synthesis of these substituted compounds has received scant
Scheme 2. A possible mechanism and conformations.
attention in the literature, with only two methods outlined.
Kabalka et al.[6] reported the nucleophilic addition of sodium
p-toluene sulfinate to an acetate of the Baylis–Hillman
adduct, and found that the reaction only proceeded in ionic
liquids at high temperatures. Later, Chandrasekhar et al.[7]
reported a nucleophilic addition of sodium p-toluene sulfinate
to the Baylis–Hillman adduct in polyethylene glycol as the
solvent at high temperatures. It is therefore significant that
the new method we report herein is unprecedented, quite
general, and proceeds efficiently at ambient temperature.
Treatment of methyl 2-(hydroxyphenylmethyl) acrylate
(3a, 1 equiv) with p-toluenesulfonyl cyanide (4b, 1.2 equiv) in
the presence of diisoproylethylamine (1.3 equiv) in dichloro-
methane at room temperature for 12 hours afforded trisub-
stituted allyl sulfone 2a in high yield (92%) with good
selectivity (E/Z 5:95; Table 1, entry 1). The structure of 2a
1
was assigned based on H,13C NMR spectroscopy and mass
spectrometry as well as by comparison with literature data.[6,7]
The E/Z ratio was determined to be 5:95 by 1H NMR analysis
of the crude product. Similarly, the reaction of 3a with
benzenesulfonyl cyanide (4a) in the presence of iPr2NEt in
CH2Cl2 at room temperature for 12 hours also proceeded to
give the trisubstituted allyl sulfone 2b in 95% yield and with
an E/Z ratio of 6:94 (Table 1, entry 2).
Encouraged by these results, we turned our attention to
other substituted aromatic acrylates. Interestingly, a large
number of these acrylates such as p-methyl, o-bromo,
cinnamyl, and furfuryl derivatives reacted cleanly with
benzenesulfonyl cyanide (4a) or p-toluenesulfonyl cyanide
(4b) in the presence of base leading to the corresponding
trisubstituted allyl sulfones 2c–2h (Table 1, entries 3–8) in
high yields (80–86%) and with good selectivity (E/Z ratio
from 6:94 to 2:98). In the same way, aliphatic Baylis–Hillman
adducts such as methyl 3-hydroxy-2-methylenehexanoate
(3 f) and methyl 3-hydroxy-2-methyleneoctanoate (3g)
reacted smoothly with 4a to afford the corresponding
trisubstituted allyl sulfones 2i (75% yield) and 2j (72%
yield) with an E/Z ratio of 7:93 and 6:94, respectively
(Table 1, entries 9 and 10).
Scheme 1. Reaction of arenesulfonyl cyanides with various allylic
alcohols.
[*] Dr. L. R. Reddy, Dr. B. Hu, Dr. M. Prashad, Dr. K. Prasad
Chemical and Analytical Development
Novartis Pharmaceuticals Corporation
One Health Plaza, East Hanover, NJ 07936 (USA)
Fax: (+1)973-781-4384
Interestingly, the reaction of other Baylis–Hillman
adducts such as 3-(hydroxymethylphenyl)but-3-en-2-one
(3h) with 4a in the presence of iPr2NEt in CH2Cl2 at room
temperature for 12 hours afforded trisubstituted allyl sulfone
2k in high yield (85%) and with high selectivity (E/Z 5:95;
Table 1, entry 11). Likewise, the reaction of 4a with 2-
E-mail: rajender.leleti@novartis.com
Binhu@novartis.com
[**] We gratefully thank Prof. Dr. Dieter Seebach and Dr. Yugang Liu for
helpful suggestions regarding the mechanism.
172
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 172 –174

Angewandte
Chemie
Table 1: Reaction of arenesulfonyl cyanide with Baylis–Hillman adducts.
selective allyl sulfones 2o–2r in
good yields (Table 2, entries 1–4).
The structure of 2r was confirmed
by single-crystal X-ray diffraction
analysis (Figure 1). Finally, the
reactions of 4b with 3m or 3n
resulted in highly regiospecific
additions and yielded allyl sulfones
2s and 2t, respectively (Table 2,
entries 5 and 6). Notably, the new
Entry
Substrate 3
4
Product 2
Yield
[%][a]
E/Z[b]
1
2
3
4
5
6
7
8
9
10
3a
3a
3b
3b
3c
3c
3d
3e
3 f
4b
4a
4a
4b
4a
4b
4a
4a
4a
4a
2a
2b
2c
2d
2e
2 f
2g
2h
2i
92
95
86
84
85
84
80
80
75
72
5:95
6:94
3:97
3:97
2:98
2:98
4:96
6:94
7:93
6:94
carbon–heteroatom
bond
is
formed exclusively from the most
substituted end of the allylic
system, and led to 2s and 2t.
For the sake of understanding
this reaction, the use of other bases
was explored (Table 3). No reac-
tion was observed in the absence of
base, or with pyridine or 2,6-luti-
dine (Table 3, entries 1–3). The
reaction did not proceed to com-
pletion with triethylamine or DBU,
and thereby gave low yields
(Table 3, entries 4 and 5). A possi-
ble mechanism[8,9] for this reaction
is depicted in Scheme 2. In the first
step, allylic alcohol 3 reacts with 4
in the presence of base to form
intermediate 5, then an arene sul-
finate nucleophile and intermedi-
ate 6 are subsequently generated.
Next, the addition of the nucleo-
phile to the allylic double bond
occurs with the elimination of
HOCN to form allyl sulfones 2.
This mechanism also explains the
observed selectivity[2,10] (E and Z).
When R2 is a large group (R2 =
COOMe, COMe), conformation
II is favored and thus predomi-
nantly forms the Z isomer. If R2 is a
small group (R2 = H, CN), then
3g
2j
11
3h
4a
2k
85
5:95
12
13
14
3i
3j
3j
4a
4a
4b
2l
81
85
90
97:3
97:3
97:3
2m
2n
conformation
I is favored and
therefore predominantly forms
the E isomer.
1
[a] Yield of isolated product. [b] The selectivity was determined by H NMR analysis. The E/Z ratio of
2:98 denotes that signals for only one isomer were observed.
In summary, a general method
and a practical protocol for the
preparation of substituted allyl sul-
fones in both high yields and
selectivity was described. The reac-
tion used arenesulfonyl cyanides
(hydroxymethylphenyl)acrylonitrile (3i) or 2-[(4-chloro-
phenyl) hydroxymethyl]acrylonitrile (3j) in the presence of
iPr2NEt in CH2Cl2 at ambient temperature proceeded to give
trisubstituted allyl sulfone 2l and 2m, respectively, in good
yields (81% and 85%) and with high selectivity (E/Z 97:3;
Table 1, entries 12 and 13). In the same way, the reaction of 4b
with 3j gave trisubstituted allyl sulfone 2n in 90% yield and
with an E/Z ratio of 97:3 (Table 1, entry 14).
We also investigated the reaction of simple allylic alcohols
3k and 3l with 4a or 4b in the presence of iPr2NEt in CH2Cl2
at room temperature and obtained the corresponding E-
Figure 1. X-ray crystal structure of 2r.
Angew. Chem. Int. Ed. 2009, 48, 172 –174
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
173

Communications
of charge from The Cambridge Crys-
tallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
Table 2: Reaction of arenesulfonyl cyanide with allylic alcohols.
Entry
Substrate 3
4
Product 2
Yield
[%][a]
E/Z[b]
Received: August 4, 2008
Published online: November 28, 2008
1
2
3
3k
3k
3l
4a
4b
4a
2o
2p
2q
80
82
81
98:2
98:2
98:2
Keywords: allyl sulfones · bases ·
.
Baylis–Hillman adducts ·
synthetic methods
4
3l
4b
4b
2r
80
84
98:2
–
[1] D. H. R. Barton, J. C. Jaszberenyi,
E. A. Theodorakis, Tetrahedron
1991, 47, 9167.
[2] Recent reviews for applications of
Baylis–Hillman adducts: a) V.
Singh, S. Batra, Tetrahedron 2008,
64, 4511; b) P. Langer, Angew.
Chem. 2000, 112, 3177; Angew.
Chem. Int. Ed. 2000, 39, 3049;
c) D. Basavaiah, P. D. Rao, R. S.
Hyma, Tetrahedron 1996, 52, 8001;
d) S. E. Drewes, G. H. P. Roos,
Tetrahedron 1988, 44, 4653.
5[c]
3m
2s
6[c]
3n
4b
2t
80
–
1
[a] Yield of isolated product. [b] The selectivity was determined by H NMR analysis. The E/Z ratio of
98:2 denotes that signals for only one isomer were observed. [c] R=C6H4CH3.
[3] a) B. Gaspar, E. M. Carreira,
Angew. Chem. 2007, 119, 4603;
Angew. Chem. Int. Ed. 2007, 46,
4519; b) A. H. Stoll, P. Knochel,
Org. Lett. 2008, 10, 113.
Table 3: Optimization of reaction conditions with 3a.
[4] a) S. Patai, Z. Rapport, C. Stirling, The Chemistry of Sulfones
and Sulfoxides, Wiley, New York, 1988; b) Z. Wrꢁbel, Tetrahe-
dron 1998, 54, 2607; c) B. Quiclet-Sire, S. Seguin, S. Z. Zard,
Angew. Chem. 1998, 110, 3056; Angew. Chem. Int. Ed. 1998, 37,
2864; d) S. Kim, C. J. Lim, Angew. Chem. 2002, 114, 3399;
Angew. Chem. Int. Ed. 2002, 41, 3265.
[5] N. Neamati, G. W. Kabalka, B. Venkataiah, R. Dayam,
WO2007081966, 2007.
[6] G. W. Kabalka, B. Venkataiah, G. Dong, Tetrahedron Lett. 2003,
44, 4673.
Entry
Base
Yield [%][a]
1
2
3
4
5
6
none
no reaction[b]
pyridine
2,6-lutidine
triethylamine
DBU
no reaction[b]
no reaction[b]
50
78
95
[7] S. Chandrasekhar, B. Saritha, V. Jagadeshwer, C. Narsihmulu, D.
Vijay, G. D. Sarma, B. Jagadeesh, Tetrahedron Lett. 2006, 47,
2981.
iPr2NEt
[a] Yield of isolated product. [b] Starting material was recovered
[8] To understand the reaction mechanism, sulfinate 1a was
prepared and subjected to the present reaction conditions. No
change was observed over a period of 48 hours. This result is
consistent with reported observations where allyl sulfinate–allyl
sulfone rearrangement was observed under harsh reaction
conditions; a) K. Hiroi, R. Kitayama, S. Sato, J. Chem. Soc.
quantitatively. DBU=1,8-diazabicyclo[5.4.0]undec-7-ene.
with Baylis–Hillman adducts and simple allylic alcohols.
Extension of this work to other related systems is currently
underway.
Experimental Section
General Procedure for the synthesis of substituted allyl sulfones 2:
iPr2NEt (6.5 mmol) was slowly (over 5 min) added dropwise to a
stirred mixture of allylic alcohol 3 (5.0 mmol) and arenesulfonyl
cyanide 4 (6.0 mmol) in CH2Cl2 (20 mL) at 238C under nitrogen.
After stirring for 12 hours at 238C, the reaction mixture was
quenched with water (20 mL) then extracted with ethyl acetate (2 ꢀ
20 mL). The combined organic layers were washed with 20% citric
acid (1 ꢀ 20 mL), water (1 ꢀ 10 mL), and the solvent was removed
in vacuo. The crude product was purified by column chromatography
on silica gel (ethyl acetate/hexanes) to afford pure 2.
Chem. Commun. 1983, 1470; b) K. Hiroi, R. Kitayama, S. Sato,
Chem. Pharm. Bull. 1984, 2628.
[9] Based on a control experiment where allylic alcohol 3k was
treated with benzenesulfonyl cyanide 4a in the presence of
sodium p-toluene sulfinate, a mixture of allyl sulfones 2o and 2p
was observed. In view of this scrambling a cyclic mechanism was
ruled out.
[10] H. J. Lee, H. S. Kim, J. N. Kim, Tetrahedron Lett. 1999, 40, 4363.
Crystallography data: CCDC-697287 contains the supplementary
crystallographic data for this paper. These data can be obtained free
174
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 172 –174