Selective formation of α-methylene-β-amino acid derivatives through the aza version of the Baylis-Hillman reaction

Full text of DOI:10.1021/jo0158635

6498
J . Org. Chem. 2001, 66, 6498-6501
Sch em e 1
Sch em e 2
Selective F or m a tion of
r-Meth ylen e-â-a m in o a cid Der iva tives
th r ou gh th e Aza Ver sion of th e
Ba ylis-Hillm a n Rea ction
Daniela Balan and Hans Adolfsson*
Department of Organic Chemistry, Stockholm University,
SE-106 91 Stockholm, Sweden
hansa@organ.su.se
Received J une 25, 2001
In tr od u ction
The selective formation of carbon-carbon bonds re-
mains as an important challenge in organic synthesis.
The Baylis-Hillman reaction allows the direct formation
of R-methylene-â-hydroxycarbonyl compounds in a base-
catalyzed tandem reaction (Michael and enolate addition
followed by elimination) of R,â-unsaturated carbonyls
with aldehydes (Scheme 1).1,2 There are a number of
different bases employed to catalyze this reaction, but
the most frequently used catalysts are nucleophilic
nonsterically hindered tertiary amines like 1,4-diazabi-
cyclo[2.2.2]nonane (DABCO). Recently, it was found that
employing a combination of Lewis acids and nucleophilic
bases further improved the reaction.3
The aza version of the reaction, i.e., exchanging the
aldehyde reactant for an aldimine and thus forming
R-methylene-â-aminocarbonyl compounds, has previously
been reported,4 although no general protocol for the
reaction has been established. R-Methylene-â-aminocar-
bonyl compounds can be obtained by a simple substitu-
tion reaction on the adducts formed in the regular
Baylis-Hillman reaction, displacing the alcohol func-
tionality by an amine.5 This reaction, though, normally
leads to a loss in selectivity since competing SN2′ reac-
tions or Michael additions on the allylic substrates result
in the formation of regioisomers.5a,b,6 Herein, we like to
report on a selective three-component one-pot procedure
forming unsaturated â-amino acid derivatives from al-
dehydes, sulfonamides, and R,â-unsaturated carbonyls
based on the aza version of the Baylis-Hillman reaction.
Resu lts a n d Discu ssion
The standard Baylis-Hillman reaction is very sensi-
tive toward the reaction conditions employed, and long
reaction times are typically required to obtain syntheti-
cally useful yields of the desired adducts. Furthermore,
the reaction is very substrate selective and a variety of
reaction conditions are available for specific substrates.1a
The latter is true also for the formation of R-methylene-
â-aminocarbonyl compounds employing the Baylis-Hill-
man protocol.4 Therefore, we decided to investigate the
formation of R-methylene-â-aminocarbonyls using a three-
component system and to optimize the reaction conditions
required for a general protocol for the aza version of the
Baylis-Hillman reaction. As a model system we chose
to study the reaction between p-toluenesulfonamide
(tosylamide), benzaldehyde, and methyl acrylate, in the
presence of a base catalyst (Scheme 2).
Previous studies showed that triphenylphosphine4b and
DABCO4a,e were, at elevated temperatures, efficient
catalysts for the reaction, although in these cases the
aldimine component was sometimes preformed. In an
initial experiment, mixing all three components in sto-
ichiometric amounts and using a catalytic amount of
DABCO (15 mol %) as base, we obtained after 48 h a
moderate yield of the tosylamido adduct 1 together with
a small amount of the alcohol adduct 2 (Table 1, entry
1). Recent developments of the classic Baylis-Hillman
reaction involve the introduction of Lewis acids in
combination with nucleophilic bases to increase the rate
of the adduct formation.3 Several lanthanide triflates,
e.g., La(OTf)3, used in catalytic amounts were shown to
accelerate the reaction.3a Adding 2 mol % of La(OTf)3
combined with 15 mol % DABCO to the three-component
reaction resulted in yields in the same range as obtained
without having the Lewis acid present (entry 2).
* Corresponding author.
(1) (a) Basaviah, D.; Rao, P. D.; Hyma, R. S. Tetrahedron 1996, 52,
8001. (b) Drewes, S. E.; Roos, G. H. P. Tetrahedron 1988, 44, 4653.
(2) (a) Morita, K.; Suzuki, Z.; Hirose, H. Bull. Chem. Soc. J pn. 1968,
41, 2815. (b) Baylis, A. B.; Hillman, M. E. D. German Patent 2155113,
1972 (Chem. Abstr. 1972, 77, 34174q).
(3) (a) Aggarwal, V. K.; Mereu, A.; Tarver, G. J .; McCague, R. J .
Org. Chem. 1998, 63, 7183. (b) Kataoka, T.; Iwama, T.; Tsujiyama, S.;
Iwamura, T.; Watanabe, S. Tetrahedron 1998, 54, 11813. (c) Shi, M.;
J iang, J .-K.; Feng, Y.-S. Org. Lett. 2000, 2, 2397.
The small, but still significant, amount of alcohol
adduct formed in the reaction indicated that the rate of
the aldimine formation was on the same level as or
possibly even slower than the Michael addition forming
the enolate. To increase the rate of aldimine formation,
and thus push the equilibrium toward the aldimine,
molecular sieves (4 Å) were added to trap the water
formed in the reaction. This resulted in a significant
overall rate and selectivity enhancement in the three-
component reaction. After 24 h we obtained 72% yield of
(4) (a) Perlmutter, P.; Teo, C. C. Tetrahedron Lett. 1984, 25, 5951.
(b) Bertenshaw, S.; Kahn, M. Tetrahedron Lett. 1989, 30, 2731. (c)
Cyrener, J .; Burger, K. Monatsh. Chem. 1994, 125, 1279. (d) Campi,
E. M.; Holmes, A.; Perlmutter, P.; Teo, C. C. Aust. J . Chem. 1995, 48,
1535. (e) Richter, H.; J ung, G. Tetrahedron Lett. 1998, 39, 2729.
(5) (a) Drewes, S. E.; Horn, M. M.; Ramesar, N. Synth. Commun.
2000, 30, 1045. (b) Takagi, M.; Yamamoto, K. Chem. Lett. 1989, 2123.
(c) Brown, J . M.; J ames, A. P.; Prior, L. M. Tetrahedron Lett. 1987,
28, 2179.
(6) (a) Kulkarni, B. A.; Ganesan, A. J . Comb. Chem. 1999, 1, 373.
(b) Kundu, M. K.; Bhat, S. V. Synth. Commun. 1999, 29, 93. (c) Nayak,
S. K.; Thijs, L.; Zwanenburg, B. Tetrahedron Lett. 1999, 40, 981. (d)
Foucaud, A.; El Guemmout, F. Bull. Soc. Chim. Fr. 1989, 403.
10.1021/jo0158635 CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/09/2001

Notes
J . Org. Chem., Vol. 66, No. 19, 2001 6499
Ta ble 1. Con d ition s for th e Aza Ver sion of th e
Ta ble 2. F or m a tion of r-Meth ylen e-â-tosyla m id o
Ba ylis-Hillm a n Rea ction a ,b
Ca r bon yl Com p ou n d sa
La(OTf)3
(mol %) MS 4 Å
time
(h)
1
2
entry
base (mol %)
(%) (%)
1
2
DABCO (15)
DABCO (15)
DABCO (15)
DABCO (15)
DABCO (15)
DABCO (5)
3-HQD (15)
cinchonidine (15)
DBU (15)
-
48
48
24
24
24
19
23
25
24
24
62
63
72
76
22
13
87
7
10
9
2
2
2
2
2
2
2
2
2
-
+
+
+
+
+
+
+
+
3c
4d
5e
6
DABCOb
3-HQDb
6
4
t
Ac
B
t
Ac
B
7
entry
Ar
(h)
(%)
(%) (h)
(%)
(%)
8
9
15
50
1
2
3
4
5
6
7
3-chlorophenyl
3-nitrophenyl
4-nitrophenyl
24 83 (58)
24 81 (59)
5
9
23 80
23 69
11
23
10
PPh3 (15)
10
72 43 (14) 14 72 73(56) 23
a
Reaction conditions: benzaldehyde, p-toluensulfonamide, and
methyl acrylate (1 equiv of each), base, La(OTf)3, and molecular
sieves (4 Å, 200 mg) as above in 2-propanol at ambient tempera-
4-methoxyphenyl 72 27 (12)
72 50
72 90(77)
2-naphthyl
2-furanyl
2-pyridyl
72 42 (36)
24 75 (57) 10 23 80
24 54 (51) 13 24 88(83)
12
7
ture. Yields determined by 1H NMR with internal standard.
b
c 86% yield of 1 after 48 h. Temperature 40 °C. e Temperature 0
d
a
Reaction conditions: arylaldehyde, p-toluensulfonamide, and
methyl acrylate (1 equiv of each), base (0.15 equiv), La(OTf)3 (0.02
equiv), and molecular sieves (4 Å, 200 mg) in 2-propanol at
°C.
ambient temperature. Yields determined by 1H NMR with
b
the tosylamido adduct and, most importantly, no alcohol
adduct was formed (entry 3). Prolonging the reaction time
to 48 h resulted in an 86% yield of 1, and the alcohol
adduct was not observed. Increasing the temperature to
40 °C resulted in a slightly higher yield of 1 (entry 4).
Further increase of temperature to refluxing 2-propanol
resulted, however, in a lower yield (50%, 20 h). Reducing
the temperature to 0 °C, which recently was reported to
increase the rate of the classic Baylis-Hillman reaction,7
resulted in low yield and poor selectivity (entry 5). The
ratio between the base and the Lewis acid turned out to
be crucial; decreasing the amount of DABCO to 5 mol %
without changing the amount of La(OTf)3 resulted in a
substantially lower yield (entry 6). This can be explained
by coordination of the base to the Lewis acidic metal
center, thereby decreasing the concentration of the active
nucleophilic catalyst. In fact, running the reaction under
the same conditions (5 mol % DABCO) without any Lewis
acid present resulted in higher yield (1, 58%; 2, 4%; 19 h
at 40 °C). In a recent attempt to classify Lewis acids on
the basis of activity and selectivity in enolate additions
to aldehydes and aldimines, respectively, Kobayashi et
al. demonstrated that lanthanide chlorides are very imine
selective.8 Other Lewis acids showing high selectivity for
imine activation included copper(I/II) chlorides, and
exchanging the chloride for other weakly basic noncoor-
dinating counteranions, e.g., the triflate anion, further
increased the aldimine selectivity. Thus, replacing La-
(OTf)3 with Cu(OTf)2 (2 mol %) in the three-component
reaction resulted in a highly selective, albeit slightly
lower yielding, formation of 1 (60%, 24 h). Previously,
3-hydroxyquinuclidine (3-HQD) has successfully been
used as base in the Baylis-Hillman reaction.9 The rate
enhancement of the reaction observed using 3-HQD was
attributed to a stabilization of the zwitterionic enolate
via hydrogen bonding from the alcohol. Replacing
DABCO for 3-HQD resulted in a higher yield of 1
although with slightly reduced selectivity (entry 7). When
the sterically more demanding cinchonidine was em-
ployed as base, low conversion to the adduct was observed
(entry 8).10 1,8-Diazabicyclo[5.4.0]undec-7-ene (DBU),
internal standard. c Isolated yields within parentheses.
recently reported as the optimum catalyst for the Baylis-
Hillman reaction,11 performed poorly in the aza version
of the reaction, resulting only in low yield of the tosyl-
amido adduct (entry 9). Using triphenylphosphine as base
under these conditions resulted in moderate yield and
selectivity (entry 10). To conclude, the best conditions
found were employing DABCO or 3-HQD (15 mol %)
together with La(OTf)3 (2 mol %) and molecular sieves.
With the optimized conditions in hand we then focused
on the scope of the reaction. In Table 2 we have sum-
marized the results from using a number of different
arylaldehydes. Arylaldehydes with electron-withdrawing
substituents performed well under the conditions em-
ployed, although their greater electrophilicity resulted
in the formation of alcohol adducts to various degrees.
The use of DABCO as base gave somewhat better
selectivities than using 3-HQD. When 4-methoxybenz-
aldehyde and 2-naphthaldehyde were used as electro-
philes, higher yields and excellent selectivities were
obtained using 3-HQD as base, whereas DABCO cata-
lyzed the formation of the adducts in a substantially
lower yield. Hetereocyclic aldehydes showed high reactiv-
ity, although the selectivities for the aza adduct were
moderate. Using the standard conditions, the intermedi-
ate N-sulfonyl aldimines were formed in yields ranging
from 50% to 80%.12 Thus, this procedure has the potential
of being used in N-sulfonyl aldimine formation, a sub-
stance class widely used in organic synthesis.13
Aliphatic aldehydes, however, did not yield any aza
adducts under these conditions, due to very slow forma-
tion of the intermediate imines.
(10) Employing cinchonidine as nucleophile resulted in the formation
of 1 in 40% enantiomeric excess. The low conversion, however, disfavors
the use of this base. The asymmetric Baylis-Hillman reaction, using
modified cinchona alkaloids, was recently reported; see: Iwabuchi, Y.;
Nakatani, M.; Yokohama, N.; Hatakeyama, S. J . Am. Chem. Soc. 1999,
121, 10219.
(11) Aggarwal, V. K.; Mereu, A. Chem. Commun. 1999, 2311.
(12) N-Sulfonyl aldimines were formed in moderate to high yields
using the following conditions: Sulfonamide (1 equiv) and aldehyde
(1 equiv) were dissolved in 2-PrOH (0.5 mL) followed by the addition
of La(OTf)3 (0.02 equiv), MS (4 Å, 200 mg/mmol), and DABCO (0.15
(7) Rafel, S.; Leahy, J . W. J . Org. Chem. 1997, 62, 1521.
(8) Kobayashi, S.; Busujima, T.; Nagayama, S. Chem. Eur. J . 2000,
6, 3491.
(9) (a) Drewes, S. E.; Freese, S. D.; Emslie, N. D.; Roos, G. H. P.
Synth. Commun. 1988, 18, 1565. (b) Bailey, M.; Marko´, I. E.; Ollis, W.
D.; Rasmussen, P. R. Tetrahedron Lett. 1990, 31, 4509. (c) Bode, M.
L.; Kaye, P. T. Tetrahedron Lett. 1991, 32, 5611.
equiv). Stirring for 24-72
h at ambient temperature. For other
methods for the preparation of N-sulfonyl aldimines, see: Chemla, F.;
Hebbe, V.; Normant, J .-F. Synthesis 2000, 75 and references cited
therein.
(13) (a) Adams, J . P. J . Chem. Soc., Perkin Trans. 1 2000, 125. (b)
Bloch, R. Chem. Rev. 1998, 98, 1407.

6500 J . Org. Chem., Vol. 66, No. 19, 2001
Notes
cleavage protocol,14 reacted however poorly and in the
case of using 3-HQD as base with no chemoselectivity.15
Perfoming the reaction with methanesulfonamide as
amine source resulted in the formation of the aza adduct
in high yield and good selectivity within reasonable time.
As stated above, this protocol efficiently produces N-
sulfonyl aldimines in high yields. However, primary
aliphatic and aromatic amines did not form imines under
these reaction conditions and thus no formation of aza
adducts were observed. In fact, performing the reaction
with preformed N-benzylideneaniline did not result in
the formation of the aza adduct, an observation which
strongly suggests that activated aldimines are required
for the reaction to occur. Benzyl and tert-butyl carbamate,
previously reported to yield aza adducts,4b did not result
in the formation of the desired products, either using our
optimized protocol or when we employed PPh3 as catalyst.
Ta ble 3. F or m a tion of r-Meth ylen e-â-tosyla m id o
Ca r bon yl Com p ou n d sa
DABCOb
3-HQDb
Michael
acceptor
t
Cc
(%)
D
t
Cc
(%)
D
(%)
entry
(h)
(%) (h)
1
2
3
4
methyl acrylate
tert-butyl acrylate
acrylonitrile
48 86 (80)
72 45 (27)
20 63 (53)
23 87
72 78(68)
23 65
4
9
20
13
phenyl vinyl sulfone 72 32 (16)
72 21
a
Reaction conditions: benzaldehyde, p-toluensulfonamide, and
Michael acceptor (1 equiv of each), base (0.15 equiv), La(OTf)3 (0.02
equiv), and molecular sieves (4 Å, 200 mg) in 2-propanol at
b
ambient temperature. Yields determined by 1H NMR with
internal standard. c Isolated yields within parentheses.
Con clu sion s
Ta ble 4. F or m a tion of r-Meth ylen e-â-su lfon yla m id o
Ca r bon yl Com p ou n d sa
We have demonstrated that employing a three-com-
ponent reaction mixture of arylaldehydes, sulfonamides,
and a Michael acceptor results in moderate to high yields
of R-methylene-â-sulfonylamido carbonyl derivatives. The
reactions were catalyzed by base (DABCO or 3-HQD) and
Lewis acid (La(OTf)3), in the presence of molecular sieves
(4 Å), and the products were easily isolated by simple
extraction. Regarding the choice of base, we generally
would recommend DABCO, since better chemoselectivi-
ties were obtained with this catalyst. For less electro-
philic aldehydes, however, the use of 3-HQD resulted in
higher yields of the aza adducts.
DABCOb
3-HQDb
entry
R
t (h) E (%) 2 (%) t (h) E (%) 2 (%)
1
2
3
4
phenyl
23
74
74
45
87
4
8
23
22
22
22
82
90
33
80
4
4
26
5
4-methoxyphenyl 23
2-nitrophenyl
methyl
24
24
a
Reaction conditions: benzaldehyde, sulfonamide, and methyl
acrylate (1 equiv of each), base (0.15 equiv), La(OTf)3 (0.02 equiv),
and molecular sieves (4 Å, 200 mg) in 2-propanol at ambient
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r e16 for th e Aza Ver sion
of th e Ba ylis-Hillm a n Rea ction , Exem p lified for th e
F or m a tion of Meth yl r-Meth ylen e-â-[(p-tolu en esu lfon yl)-
a m in o]-3-p h en ylp r op ion a te (1).4d In a dried flask, tosylamide
(855 mg, 5 mmol), DABCO (84 mg, 0.75 mmol), and La(OTf)3‚
H2O (58.5 mg, 0.1 mmol) were measured together with molecular
sieves (4 Å, 900 mg). 2-PrOH (2.5 mL),17 benzaldehyde (505 µL,
5 mmol), and methyl acrylate (450 µL, 5 mmol) were added, and
the reaction mixture was stirred for 48 h at ambient tempera-
ture. The mixture was filtered through a thin layer of Celite,
which was rinsed three times with 2-PrOH (10 mL). The solvent
was evaporated, and to the crude mixture were added methanol
(25 mL) and aqueous sulfuric acid (10 mL, 1M).18 The solution
was stirred for 1 h, and then methanol was evaporated. The
remaining acidic soulution was diluted with water and extracted
with dichloromethane (3 × 30 mL). The organic phase was then
successively washed with NaHCO3 (saturated), NaOH (1 M),
water, and NaCl (saturated) solutions and dried over Na2SO4.19
Evaporation of the solvent gave 1.38 g (80%) of the pure product
b
1
temperature. Yields determined by H NMR with internal stan-
dard.
We then investigated the range of possible Michael
acceptors working in the three-component reaction (Table
3). When comparing acrylic esters, the steric bulkiness
of the ester turned out to be of high importance for the
reactivity. Methyl acrylate reacted readily as described
above, whereas tert-butyl acrylate reacted slower and
gave much lower yield in the case of using DABCO as
base and slightly lower yield using 3-HQD. Acrylonitrile
reacted rather fast under these conditions although the
chemoselectivity was poor. Phenyl vinyl sulfone reacted
poorly, and a surprisingly big difference in selectivity was
observed depending on the choice of base. The low yield
can be explained by an interesting side reaction, resulting
from a competing double Michael addition of tosylamide
on the vinyl sulfone. Other Michael acceptors known to
give Baylis-Hillman adducts, e.g., vinyl phosphonates,
gave using these conditions no detectable amounts of
adducts. Michael acceptors substituted in the â-position,
e.g., ethyl crotonate and cyclohex-2-enone, gave no ad-
ducts under these conditions.
(14) Fukuyama, T.; J ow, C.-K.; Cheung, M. Tetrahedron Lett. 1995,
36, 6373.
(15) Prolonging the reaction time to 72 h did not improve the yield
of the aza adduct.
(16) This procedure was applied for all reactions, except in the case
of 2-pyridinecarboxaldehyde, where the Michael acceptor was added
24 h later to suppress the formation of the alcohol adduct.
(17) For solid aldehydes the amount of solvent was increased to 4
mL.
Finally we investigated the scope of using different
amines (sulfonamides) in the reaction (Table 4). As
observed above for tosylamide, benzenesulfonamide and
its 4-methoxy analogue performed well under the opti-
mized conditions and gave aza adducts in high yields and
good selectivities. The electron-deficient 2-nitrosulfona-
mide, a useful protecting group for amines due to its mild
(18) The acidic workup facilitates cleavage of remaining sulfo-
nylimine and efficiently removes the aldehyde. For reactions with low
conversion, an additional stirring with sulfuric acid (1 M) was repeated
after the basic workup.
(19) Alternative workup procedure: The crude reaction mixture was
separated on silica gel (eluent: chloroform:methanol 20:1) followed by
recrystallization in diethyl ether (in some cases an ethyl acetate diethyl
ether mixture was employed). Using this procedure typically resulted
in lower isolated yields.

Notes
J . Org. Chem., Vol. 66, No. 19, 2001 6501
as a white crystalline material: mp 76-77 °C; 1H NMR (400
MHz, CDCl3) δ 2.41 (s, 3H), 3.61 (s, 3H), 5.31 (d, J ) 8.9 Hz,
1H), 5.61 (d, J ) 8.9 Hz, 1H), 5.83 (s, 1H), 6.22 (d, J ) 0.7 Hz,
1H), 7.13-7.25 (m, 7H), 7.68 (d, J ) 8.4 Hz, 2H); 13C NMR (100
MHz, CDCl3) δ 21.72, 52.20, 59.31, 126.64, 127.45, 127.98,
128.10, 128.80, 129.70, 137.84, 138.74, 138.81, 143.61, 165.98;
MS (MALDI-TOF) (m/z) 384.071 (MK+), 368.094 (MNa+), 346.103
(MH+).
Ack n ow led gm en t. The Swedish Natural Science
Research Council is gratefully acknowledged for finan-
cial support.
Su p p or tin g In for m a tion Ava ila ble: Characterization
data for all compounds in Tables 2-4. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O0158635