Supported BINOLate salts as base catalysts. Preparation and use in the Michael reaction

Article abstract of DOI:10.1039/a700261k

Two silica supported BINOL complexes are prepared, both of which are active catalysts in the Michael reaction.

Full text of DOI:10.1039/a700261k

View Article Online / Journal Homepage / Table of Contents for this issue
Supported BINOLate salts as base catalysts. Preparation and use in the
Michael reaction
Duncan J. Macquarrie†
Department of Chemistry, University of York, Heslington, York, UK YO1 5DD
Two silica supported BINOL complexes are prepared, both
of which are active catalysts in the Michael reaction.
The second catalyst, 2, was prepared, also from 3, in four
3
steps (amide formation with 4-aminobenzoic acid, diazotisa-
6
tion and coupling with BINOL at pH 7, followed by
The use of derivatives of 1,1A-naphthalene-2,2A-diol (BINOL) as
base catalysts has recently been disclosed.1 In particular, the
use of podand-substituted salts of BINOL has led to efficient
deprotonation). The deprotonation procedure was the same for
both catalysts. Elemental analysis was consistent with the
,2
2
1
proposed catalyst structure at a loading of 0.3 mmol g , in
accordance with weight loss as measured by TGA. Diffuse
reflectance UV–VIS (DRUV) spectroscopy indicated a peak at
520 nm, which corresponds well to the diazo coupling product
from 4-aminobenzoic acid and BINOL after physisorption on
silica. Diffuse reflectance IR spectroscopy showed peaks at
2
base catalysts for the Michael reaction. As part of a programme
aimed at the development of heterogenised base catalysts for
use in the synthesis of fine chemicals, we have prepared two
BINOL derivatives supported on silica 1 and 2, and evaluated
them in the Michael reaction.
3
21
21
The first catalyst, 1, was prepared from g-aminopropylsilica
1653 and 1565 cm (amide) and 1600, 1585 and 1506 cm
4,5
(
AMPS, 3) and 4, prepared by known methods from the
(aromatics). The position of attachment of the NNN unit to the
BINOL ring system is, as yet, unknown.
commercially available poly(ethylene glycol) methacrylate
Aldrich, M = 306) by tosylation, displacement with the Na
(
n
The products were evaluated as catalysts in the reaction of
but-3-en-2-one 5 with the b-diketones ethyl cyclohexanone
2-carboxylate 6 and ethyl 2-oxocyclopentane-2-carboxylate 7
(Scheme 1). The catalyst (0.25 g), was added to a solution of
diketone (10 mmol) and but-3-en-2-one (14 mmol) in solvent
(15 ml). The reaction was stirred at room temp. and monitored
salt of BINOL, hydrolysis of the methacrylate, retosylation and
displacement of the tosylate with 3 under reflux in toluene
overnight (Fig. 1). Deprotonation of the BINOL unit was
achieved by stirring with 1.5 equiv. NaOH in methanol (0.05
2
3
mol dm ) for 4 h, followed by exhaustive washing with
methanol to remove excess base and physisorbed material.
Thermogravimetric analysis (TGA) and elemental analysis
2
1
O
were in agreement and indicated a loading of 0.2 mmol g . The
product was further analysed by 13C–CPMAS (300 MHz,
d 8.67, 18.00, 42.93 aminopropyl; 69.57 PEG, 127.04 + weak
signal at 137–140 BINOL unit). The Diffuse Reflectance FTIR
O
O
O
O
Catalyst
OEt
+
CO2Et
(
DRIFT) spectrum of the product displayed the expected peaks
2
1
for the BINOL unit at 1603, 1579 and 1510 cm
.
(
)n
( )n
6
7
n = 1
n = 0
5
Scheme 1 Michael addition between b-diketones and but-3-en-2-one,
catalysed by supported BINOL catalysts
O
O
NaO
100
Si
Si
N
H
n
1
O
75
50
25
N
H
N
N
OH
ONa
2
0
Si
NH2
O
O
nTs
3
OH
Solvent
Fig. 2 Solvent dependence in the reaction between 5 and 6 at room temp. (15
ml solvent, 0.2 g dodecane as internal standard, 14 mmol 5 and 10 mmol
6). The reaction in isopropanol was complete after 24 h.
4
Fig. 1 structures of the silica supported catalysts and precursor
Chem. Commun., 1997
601

View Article Online
by GC using dodecane as internal standard. The reaction of 6
using 1 as catalyst was evaluated first and displayed a
remarkable solvent dependence (Fig. 2). No reaction takes place
in dichloromethane or in acetonitrile, with no product being
detected even after 72 h. This is remarkable since the
homogeneous version of this catalyst is used in dichloro-
h with 6. Similar slowing of reaction was observed upon
reuse.‡
Compound 2 was also evaluated in the above reactions, and
was found to be slightly less active [reactions took 28 h (6) and
15 h (7) respectively]. Reuse was also successful, with less
reduction in performance being observed than with 1. While a
thorough solvent study has not yet been carried out on this
catalyst, reaction in dichloromethane does proceed, albeit
slowly, to the extent of 8% in 24 h.§
Future work will concentrate on further optimisation of the
systems and extension of the methodology to other phenols,
Michael reactions and to enantioselective chiral BINOL
systems.
2
methane. Toluene is poor, and ethers (diethyl ether and THF)
are slightly better. The best solvents by far are alcohols, with
ethanol being particularly effective, complete reaction being
achieved after 15 h (Fig. 2). It is possible that the oxygen-
containing solvents are better able to interact with the PEG
chain, particularly the hydrogen-bonding alcohols. The rela-
tively good performance of hexane is puzzling (we have
recently observed such anomalous behaviour in different
D. J. M. thanks the Royal Society for the award of a
University Research Fellowship.
7
systems ), and it may be due to partitioning effects. Isopropanol
effected complete transesterification of the product under the
reaction conditions. The product can be isolated by filtration of
the catalyst and normal work-up, in yields typically 5–8% lower
than the GC yield. Reuse of catalyst (after decantation of the
supernatant) has been studied (Table 1), and there is a gradual
slowing of reaction with repeated reuse (up to four uses).
Conversions are close to complete and fall off only slightly with
reuse. Compound 7 reacts more rapidly under the same
conditions, reaction being complete after 8 h, as opposed to 15
Footnote
†
‡
8
E-mail address: djm13@york.ac.uk
Turnover numbers for 1 and 2 can thus be estimated to be of the order of
00 and 530, respectively.
§ Leaching does not appear to be a problem with either catalyst, since
reactions which were interrupted before completion by removal of catalyst
did not procceed after removal of the catalyst.
Table 1 Catalyst reuse in the reaction of 5 with 6 and 7a
References
1
H. Sasai, T. Arai, Y. Satow, K. N. Houk and M. Shibasaki, J. Am. Chem.
Soc., 1995, 117, 6194; K. Iseki, S. Oishi, H. Sasai and M. Shibasaki,
Tetrahedron Lett., 1996, 37, 9081.
Catalyst
Substrate
Conversion (%) t/h
1
6
6
6
6
7
6
6
6
6
7
97
98
99
96
99
98
98
96
97
99
15
16
19
23
8
28
29
28
30
15
2
Y. Tamai, A. Kamifuku, E. Koshiishi and S. Miyano, Chem. Lett., 1995,
1
1
1
1
2
2
2
2
2
(1 reuse)
(2 reuses)
(3 reuses)
9
57.
3
4
V. A. Basiuk and A. A. Chuiko, J. Chromatogr., 1990, 521, 29.
Y. Tamai, S. Koike, A. Ogura and S. Miyano, J. Chem. Soc., Chem.
Commun., 1991, 799.
Y. Tamai, M. Akiyama, A. Okamura and S. Miyano, J. Chem. Soc.,
Chem. Commun., 1992, 687.
5
6
7
(1 reuse)
(2 reuses)
(3 reuses)
V. A. Basiuk and E. G. Khil’chevskaya, Anal. Chem. Acta, 1991, 255,
1
97.
D. J. Macquarrie, A. Lambert, A. Priest and J. Mdoe, unpublished results;
J. H. Clark, D. J. Macquarrie and J. C. Ross, unpublished results.
a
Reactions were carried out in 15 ml ethanol at room temp. with dodecane
as internal standard, using 0.25 g catalyst, 10 mmol keto ester and 14 mmol
enone. Isolated yields were 3–8% lower than the reported GC yields.
Received, 10th January 1997; Com. 7/00261K
602
Chem. Commun., 1997