Efficient Baylis-Hillman reactions promoted by mild cooperative catalysts and their application to catalytic asymmetric synthesis

Article abstract of DOI:10.1016/S0040-4039(00)00125-8

Baylis-Hillman reactions were promoted by mild cooperative catalysts of tributylphosphine with phenols such as (±)-1,1'-bi-2-naphthol (BINOL) in THF to give α-methylene-β-hydroxyalkanones in high yield. The reactions proceeded much faster in the presence

Full text of DOI:10.1016/S0040-4039(00)00125-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 2165–2169
Efficient Baylis–Hillman reactions promoted by mild cooperative
catalysts and their application to catalytic asymmetric synthesis
∗
Yoichi M. A. Yamada and Shiro Ikegami
School of Pharmaceutical Sciences, Teikyo University, Sagamiko, Kanagawa 199-0195, Japan
Received 6 December 1999; accepted 14 January 2000
Abstract
Baylis–Hillman reactions were promoted by mild cooperative catalysts of tributylphosphine with phenols such
0
as (±)-1,1 -bi-2-naphthol (BINOL) in THF to give α-methylene-β-hydroxyalkanones in high yield. The reactions
0
1
proceeded much faster in the presence of 1,1 -bi-2-naphthol than in its absence. The H NMR studies suggested
that 1,1 -bi-2-naphthol functions as a Brønsted acid to activate the carbonyl group of an aldehyde and a polarized
0
alkene. Application of the reactions to catalytic asymmetric synthesis was examined by using cooperative catalysts
of tributylphosphine with the calcium chiral catalyst to give the desired product with fairly good % ee in fairly
good yield. © 2000 Elsevier Science Ltd. All rights reserved.
The Baylis–Hillman reaction, the coupling of a polarized alkene with an aldehyde catalyzed by
a rather strong base, 1,4-diazabicyclo[2,2,2]octane (DABCO), is an important carbon–carbon bond-
forming reaction, which serves as the key step for the synthesis of several biologically active substances.1
However, since long-sustained reaction time, at least 1 week (sometimes 1 month), for completing
1
the Baylis–Hillman reaction is common due to the low reaction rate, the development of an efficient
Baylis–Hillman reaction is among the most challenging themes in organic synthesis.
A mechanism of this reaction is as follows (Scheme 1): A conjugated addition of a Lewis base (I)
to a polarized alkene (III) affords an enolate (IV), which reacts with an aldehyde (V) to give an aldol-
type intermediate (VI) followed by β-elimination with a Lewis base (I) to yield an α-methylene-β-
hydroxyalkanone (VII). It is well-known that the rate-determining step is the addition stage of IV to
2
V. Therefore, we envisioned that a mild Brønsted acid as a co-catalyst (II) would activate the carbonyl
groups of an enolate (IV) and an aldehyde (V) to accelerate the reaction rate, affording a Baylis–Hillman
3
product in high yield as shown in Scheme 1. Recently, Hatakeyama et al. reported excellent asymmetric
Baylis–Hillman reactions catalyzed by a amine possessing a phenolic hydroxy group derived from a
4
cinchona alkaloid which stabilized VI and facilitated the enantioselective reactions. Unfortunately, the
phenolic proton of the catalyst did not accelerate the reactions in this system. Herein, we would like
∗
Corresponding author.
0
040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00125-8
tetl 16404

2
166
to report remarkable acceleration of Baylis–Hillman reactions catalyzed by mild cooperative catalysts
consisting of tributylphosphine as a mild Lewis base and phenols or naphthols as a mild Brønsted acid.
Scheme 1. Mechanism of the Baylis–Hillman reaction
Predictably, when the reaction of 2-cyclopenten-1-one (1a) with 3-phenyl-1-propanal (2a) was carried
out in the presence of 20 mol% of DABCO in THF for 1 h, no products were obtained (Table 1, entry
3
a,5
1). The reaction catalyzed by 20 mol% of tributylphosphine (4)
proceeded to give the corresponding
Baylis–Hillman product, albeit in only 23% yield (entry 2). However, we were pleased to find that the
0
reaction catalyzed by 20 mol% of 4 in the presence of 10 mol% of a mild Brønsted acid, 1,1 -bi-2-
naphthol (BINOL) (5), as a co-catalyst in THF for 1 h proceeded smoothly to give the desired product 3a
in quantitative yield (entry 6), while the reaction catalyzed by methanol, benzoic acid or p-toluenesulfonic
acid as the Brønsted acid did not proceed efficiently (entries 3–5). The reaction catalyzed by 20 mol%
0
0
0
of 2-naphthol, 10 mol% of 2-hydroxy-2 -methoxy-1,1 -binaphthyl (6) and 10 mol% of 2,2 -dimethoxy-
1
0
,1 -binaphthyl (7) gave 3a in quantitative yield, 80% and 24%, respectively (entries 7–9), indicating
that both of the two protons of 5 work as Brønsted acids to accelerate the Baylis–Hillman reaction. The
6
use of DABCO, triphenylphosphine or dibutyl sulfide as Lewis bases was not effective and afforded
no products (entries 10–12). Phenol and p-toluenesulfonamide, which has the same Brønsted acidity as
7
phenol, were effective (entries 13 and 14).
Following on from these results, we turned our attention to broadening the range of substrates (Table
2
). The reaction of 1a with octanal (2b) in THF gave the desired product 3b in quantitative yield (entry 1).
An aromatic aldehyde (benzaldehyde (2c)) also reacted with 1a and 1b to give 3c and 3g in 92% and 57%
7
yields, respectively (entries 2 and 6). Moreover, the (2-methoxyethoxy)methyl (MEM) group, which is
8
often deprotected by Lewis acids such as ZnBr and TiCl , was intact under these mild conditions, so
2
4
that the reaction of 2d with 1a afforded 3d in 98% yield (entry 3). The reaction of methyl and ethyl
acrylate proceeded smoothly (entries 8–10).
The typical procedure is as follows: Under an argon atmosphere, to a stirred solution of (±)-5 (0.05
mmol; 14.3 mg) in THF (0.5 ml) was added 1a (0.5 mmol; 41.9 µl), 2a (0.75 mmol; 98.8 µl) and 4 (0.10
mmol; 24.9 µl) successively, which was stirred for 1 h at room temperature. The mixture was diluted with
0
.5 ml of hexane, and was purified by flash column chromatography (SiO , Et O/hexane=1/1) to give 3a
2 2
in quantitative yield. After completion of the reaction, the mixture was subjected to chromatographic
purification without the usual work-up procedure.
To determine the possibility of coordination between an aldehyde and an acidic proton of 5 and/or
1
between a polarized alkene and an acidic proton of 5, we examined H NMR studies on 1a, 2a and 5

2
167
Table 1
The Baylis–Hillman reaction of 1a with 2a catalyzed by a Lewis base and a Brønsted acid as a
co-catalyst
in THF-d . When 5 was added to the THF-d solution of 1a or 2a, the upfield shifts (∼0.1 ppm) of the
8
8
hydrogen signals of 1a and 2a were observed in a concentration-dependent manner of 5. By contrast,
when 1a or 2a was added to the THF-d solution of 5, the downfield shifts (∼0.1 ppm) of the signals of
8
the hydroxy protons of 5 were observed in a concentration-dependent manner of 1a or 2a. These results
might indicate that 5 functions as a Brønsted acid to coordinate with the carbonyl groups of an enone and
an aldehyde.
On the basis of the reaction system mentioned above, we focused on developing a catalytic asymmetric
Baylis–Hillman reaction. Recently, several examples of catalytic asymmetric Baylis–Hillman reactions
4
,5d,9
have been reported.
However, moderate % ees and yields were achieved when substituted-aryl
4
aldehydes were used as substrates with a chiral Lewis base. Therefore, the development of this reaction
constitutes a formidable challenge for organic chemists. We report herein a preliminary example of a
catalytic asymmetric Baylis–Hillman reaction of an aliphatic aldehyde and an α,β-unsaturated ketone
promoted by cooperative catalysts of the first optically active calcium catalyst (R)-6 as a chiral Lewis
acid with tributylphosphine (4) as an achiral Lewis base.
1
0
0
Unfortunately, the reaction of 1a with 2a catalyzed by 4 and (R)-1,1 -bi-2-naphthol ((R)-5) gave 3a
with low ee (<10% ee). After screening, we were surprised to find that an optically active calcium catalyst
(
R)-6 which was prepared from a strong base Ca(O-i-Pr) and (R)-5 in THF was an effective chiral Lewis
2
11
acid for a catalytic asymmetric Baylis–Hillman reaction (Scheme 2). To our knowledge, this is the
first example of an optically active calcium catalyst. The reaction of 1a and 1.5 mol equiv. of 2a in the
presence of 16 mol% of the chiral calcium catalyst (R)-6 and 10 mol% of 4 afforded (S)-3a with 56%
12
1
3
ee in 62% yield.
0
In conclusion, we have found that the cooperative catalysts of tributylphosphine with (±)-1,1 -bi-
2
-naphthol are effective for the Baylis–Hillman reactions to give the desired products in high yield.
Moreover, it was found that the cooperative catalysts of the first calcium chiral catalyst with tributylphos-

2
168
Table 2
The Baylis–Hillman reaction of a polarized alkene with an aldehyde catalyzed by tributylphosphine
0
(
4) with 1,1 -bi-2-naphthol (5)
Scheme 2. A catalytic asymmetric Baylis–Hillman reaction promoted by the first chiral calcium catalyst with tributylphosphine
4)
(
phine were effective for a catalytic asymmetric Baylis–Hillman reaction. Since there is a great potential
for the combination catalysts of the chiral calcium catalyst and tributylphosphine, further investigations
for the improvement of the catalysts are currently in progress.
Acknowledgements
We thank Ms. J. Shimode, Ms. J. Nonobe and Ms. M. Kitsukawa for spectroscopic measurement,
and are grateful to Mr. M. Ichinohe, Ms. M. Mizuno, Ms. Y. Ohshige, Ms. M. Honma and Mr. D.

2
169
Monguchi for their technical assistance. We thank Dainippon Ink and Chemicals, Inc. for the 1998 Award
in Synthetic Organic Chemistry, Japan to Y.M.A.Y.
References
1
. For reviews, see: (a) Drewes, S. E.; Roos, G. H. P. Tetrahedron 1988, 44, 4653–4670. (b) Basavaiah, D.; Rao, P. D.; Hyma,
R. S. Tetrahedron 1996, 52, 8001–8062. (c) Ciganek, E. Org. React. 1997, 51, 201–350.
2
3
. Bode, M. L.; Kaye, P. T. Tetrahedron Lett. 1991, 32, 5611–5614.
. Recently, Kawanisi et al., Aggarwal et al., Kataoka et al., and Kobayashi et al. reported the Lewis acid-accelerated
Baylis–Hillman reaction. (a) (Et
984, 14, 1267–1273. (b) (La(OTf)
Tarver, G. J.; McCague, R. Chem. Commun. 1996, 2713–2714. (c) (TiCl
T.; Iwama, T.; Tsujiyama, S. Chem. Commun. 1998, 197–198. (d) (LiClO
Kobayashi, S. Tetrahedron Lett. 1999, 40, 1539–1542.
3
Al+Bu
+(R)-BINOL (or other alcoholic ligands)+DABCO, MeCN, 25°C) Aggarwal, V. K.;
+Me S (or other chalcogens), CH Cl , rt) Kataoka,
+DABCO, Et
O, −25°C–rt) Kawamura, M.;
3 2 2
P, CH Cl , 80°C) Imagawa, T.; Uemura, K.; Nagai, Z.; Kawanisi, M. Syn. Commun.
1
3
4
2
2
2
4
2
4
. Recently, Hatakeyama et al. reported catalytic asymmetric Baylis–Hillman reactions of 1,1,1,3,3,3-hexafluoro-isopropyl
acrylate. Iwabuchi, Y.; Nakatani, M.; Yokoyama, N.; Hatakeyama, S. J. Am. Chem. Soc. 1999, 121, 10219–10220.
. For use of trialkylphosphine in the Baylis–Hillman reaction, see: (a) Morita, K. Japan Patent 1968, 68 03,364. (b) Miyakoshi,
T.; Ohmichi, H.; Saito, S. Nippon Kagaku Kaishi 1979, 748. (c) Rafel, S.; Leahy, J. W. J. Org. Chem. 1997, 62, 1521–1522.
5
(d) Hayase, T.; Shibata, T.; Soai, K.; Wakatsuki, Y. Chem. Commun. 1998, 1271–1272.
6
. Drewes et al. and Ollis et al. independently reported that the rate of DABCO-catalyzed Baylis–Hillman reactions was
enhanced by the use of methanol as solvent or using 3-hydroxyquinuclidine as a catalyst. (a) Drewes, S. E.; Freese, S. D.;
Emslie, N. D.; Roos, G. H. P. Syn. Commun. 1988, 18, 1565–1572. (b) Bailey, M. B.; Markó, I. E.; Ollis, W. D.; Rasmussen,
P. R. Tetrahedron Lett. 1990, 31, 4509–4512. Also, see Ref. 4.
7
8
9
. Phenols such as phenol, 4-methoxyphenol and catechol were also effective co-catalysts.
. Kocienski, P. J. Protecting Groups; Thieme: New York, 1994; pp. 74–77.
. (a) Roth, F.; Gygax, P.; Fráter, G. Tetrahedron Lett. 1992, 33, 1045–1048. (b) Roos, G. H. P.; Haines, R. J.; Raab, C. E.
Syn. Commun. 1993, 23, 1251–1259. (c) Oishi, T.; Oguri, H.; Hirama, M. Tetrahedron: Asymmetry 1995, 6, 1241–1244. (d)
Markó, I. E.; Giles, P. R.; Hindley, N. J. Tetrahedron 1997, 53, 1015–1024. (e) Barrett, A. G. M.; Cook, A. S.; Kamimura,
A. Chem. Commun. 1998, 2533–2534.
1
1
0. For examples of asymmetric Lewis acid–Lewis base bifunctional catalyst, see: (a) Ito, Y.; Sawamura, M.; Hayashi, T. J.
Am. Chem. Soc. 1986, 108, 6405–6406. (b) Noyori, R.; Kitamura, M. Angew. Chem., Int. Ed. Engl. 1991, 30, 49–69. (c)
Kobayashi, S.; Tsuchiya, Y.; Mukaiyama, T. Chem. Lett. 1991, 541–544. (d) Corey, E. J.; Helal, C. J. Angew. Chem., Int.
Ed. Engl. 1998, 37, 1986–2012. (e) Hamashima, Y.; Sawada, D.; Kanai, M.; Shibasaki, M, J. Am. Chem. Soc. 1999, 121,
2
641–2642.
1. Preparation of the calcium catalyst: Under an argon atmosphere, to a solution of Ca(O-i-Pr)
2
(120.6 mg; 0.762 mmol)
0
(purchased from High Purity Chemicals Lab. Co., Ltd) in THF (7.65 ml) was added a solution of (R)-1,1 -bi-2-naphthol
(327.3 mg; 1.14 mmol) in THF (11.4 ml) at room temperature, and the mixture was stirred overnight at the same temperature.
THF was evaporated under reduced pressure, the residue was dried in vacuo for 4 h, THF (19.5 ml) was again added under
an argon atmosphere, and the suspension was allowed to stand for several hours to give a solution of the catalyst.
2. The enantiomeric excess of 3a was determined by HPLC analysis (Daicel Chiralpak AS; detection at 254 nm; eluent:
1
1
2
(
1
-propanol/hexane (1/9)), and the absolute configuration was determined by Mosher’s method after conversion to (2R*)-2-
(1S*)-1-hydroxy-3-phenylpropyl)cyclopentanone. For Mosher’s method, see: Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc.
973, 95, 512–519.
3. This result might indicate that the calcium catalyst has been self-assembled with 4. For self-assembly of heterobimetallic
catalyst and reactive nucleophiles, see: Shimizu, S.; Ohori, K.; Arai, T.; Sasai, H.; Shibasaki, M. J. Org. Chem. 1998, 63,
7
547–7551, and references cited therein.