Mendeleev Commun., 2006, 16(4), 202–204
reaction, and the latter initiated tandem transformations leading
Table 1 Interaction of ketone 1 with malononitrile.
to polyfunctional compound 3, which is formed as a mixture of
two diastereomers.
The mechanism suggested for the reaction forming com-
pound 3 was confirmed by the fact that starting dinitrile 2 was
not converted to compound 3 under the conditions of its reaction
with malononitrile.
Yielda (%)
Conversion Reaction
Catalyst
[a]D20 eeb
for 3 (%)
2/3
(%)
time (days)
2
3
Csβ8
73
100
98
2
2
40
30
7
0
5.7
—
0
0
8
CsxOy
—
—
8%Met-Csβ
8%Met-Csβ
11%Met-Csβ
14%Met-Csβ 100
Met-Cs 100
14
18
28
18
18
28 10
2.8 +11 3.3
The formation of asymmetric centres in product 3 in the
course of the above transformations prompted us to synthesise
from Csβ a heterogeneous chiral catalyst suitable for asymmetric
catalysis of reactions of α,β-unsaturated carbonyl compounds
with malononitrile. For test reaction, we chose the transforma-
tion of ketone 1 into polyfunctional compound 3. Currently, we
know of only two catalytic systems suitable for Michael reactions
between malononitrile and α,β-unsaturated carbonyl compounds;
the systems use homogeneous complexes of Mg2+ and Ni2+ ions
with hardly accessible chiral ligands based on substituted phenyl-
oxazoles.10
Heterogeneous chiral catalysts described in the literature were
created by modifying zeolites with optically active compounds.
Thus, zeolites NaX and NaY modified with (–)- or (+)-ephedrine
were employed for asymmetric catalysis of the Norrish–Young
photoreaction.11 Chiral 1,3-dithiane-1-oxide applied to the zeolite
HY support was used for investigating dehydration of a mixture
of S- and R-isomers of butan-2-ol, which led to preferable con-
version of one of the enantiomers.12 No examples of chiral
compounds applied to zeolites and designed for the preparation
of asymmetric basic catalysts have been found in the literature.
The structure of zeolite Csβ involves caesium oxide clusters,
the caesium content being ~34 wt.%.8 For the chiral modifier,
we chose L-methionine. Its interaction with caesium oxides led
to caesium methionate, which is insoluble in nonpolar solvents
and is tightly held by the zeolite surface. We prepared catalyst
samples with different mass contents of L-methionine: 8%, 11%,
and 14%.† For zeolite with 8% L-methionine (8%Met-Csβ), the
area of the specific free surface decreased considerably: 23 m2 g–1
(as determined by argon adsorption at 77 K) versus 51 m2 g–1
for the starting Csβ. Table 1 presents the results of the catalytic
activity test for the catalysts in the reaction of ketone 1 with
malononitrile.
c
88c
67
—
9
—
4
0.2
0.67
0
+20
0
0
6
0
0
44 11
25
10 15
5
0
aYield given based on the converted terpenoid 1 or 2. Determined using
1H NMR spectral data and a Eu(hfc)3 addition as a chiral shift reagent.
cCompound 2 was employed as a starting compound.
b
active compound; the diastereomeric excess (ee, 3.3%) was deter-
mined from H NMR spectral data using Eu(hfc)3 as a chiral
1
shift reagent. Using 8%Met-Csβ resulted in a considerable
retardation of the reaction and decreased total yield of the pro-
ducts compared to the case with Csβ used as a catalyst. The
ratio between the yields of dinitrile 2 and polyfunctional com-
pound 3 changed in favour of the latter. This change may be
explained, at least partially, by a reaction between dinitrile 2
and malononitrile leading to compound 3 in the presence of
the 8%Met-Csβ catalyst,§ which does not occur in the presence
of Csβ. The product did not possess optical activity. Thus, in
the presence of modified zeolite 8%Met-Csβ as a catalyst, the
reaction follows one of the routes possible in this case: primary
formation of a Michael or Knoevenagel reaction product, the
latter leading to a racemic compound.
When the content of L-methionine in the catalyst was raised
to 11%, the total yield of the reaction products increased to
55%, while conversion decreased to 67%, and the optical purity
of 3 increased to 6%.
To check if the optical purity of the product has been changed,
e.g., by racemization or kinetic separation, we kept compound 3
on the 11%Met-Csβ catalyst (mass ratio of compound 3 to
catalyst was 1:2) for 14 days under reaction conditions. This
gave tar product 3 (only 10% of compound 3 was recovered)
whose optical purity did not change.
When 8%Met-Csβ was used as a catalyst, the yields of
reaction products 2 and 3 were 28 and 10%, respectively (con-
version was nearly 100%) (Table 1).‡ Product 3 is an optically
With 14% L-methionine in the catalyst, the reaction of
ketone 1 with malononitrile led to racemic compound 3 as the
major reaction product. The absence of optically active products
in this case may be explained by the fact that due to excess
L-methionine applied to zeolite, the reaction is catalysed by
caesium methionate lying on the surface of the zeolite and is
not affected by the crystal structure of the zeolite. Indeed, when
this reaction was carried out using amorphous caesium methionate
as a catalyst, compound 3 isolated as a product did not possess
optical activity. Thus, the crystal structure of zeolite proved to
be essential to enantioselective catalysis.
The resulting heterogeneous chiral catalyst was tested in a
reaction of another α,β-unsaturated ketone, benzalacetone 4, with
malononitrile (Scheme 2). When conducted in the presence of
8%Met-Csβ, this reaction gave compound 5 in 26% yield based
on the converted ketone (conversion 43%).¶ The isolated pro-
duct was optically active (ee, 3%). The interaction of benzal-
acetone 4 with malononitrile on the starting zeolite Csβ led to
the formation of racemic compounds 5 and 6 in 3 and 6% yields,
respectively.
†
A methanol solution of the calculated amount of L-methionine was
poured into zeolite Csβ (100 mg) preliminarily calcined for 2 h at 500 °C.
The solvent was distilled off on a rotary vacuum evaporator. The
resulting white powder was dried in a vacuum (0.07 Torr) at 50 °C for
2 h immediately before use.
‡
In the presence of 8%Met-Csb. A solution of malononitrile (0.151 g,
2.3 mmol) in diethyl ether (2 ml) was added to Met-Csβ (0.05 g). The
ether was distilled off, and 5,5,8-trimethylnona-3,7-dien-2-one 1 (0.106 g,
0.6 mmol) was added. The mixture was allowed to stand for 14 days at
20 °C and then treated with ethyl acetate. The reaction products were
separated on a silica gel column using hexane as an eluent with a diethyl
ether gradient from 0 to 100%. This gave starting ketone 1 (0.002 g,
conversion 98%), 2-(1,4,4,7-tetramethylocta-2,6-dienylidene)malononitrile
2 (0.038 g, yield 28% based on converted ketone 1) and 1-hydroxy-
2-imino-4-methyl-6-(1,1,4-trimethylpent-3-enyl)cyclohex-3-ene-1,3-di-
nitrile 3 (0.018 g, yield 10% based on converted ketone 1). For com-
pound 3 [a]52080 = +11 (c 0.6, CHCl3).
In the presence of 11%Met-Csb. Similarly, when malononitrile (0.472 g,
7.2 mmol) reacted with ketone 1 (0.333 g, 1.8 mmol) in the presence of
11%Met-Csβ, (0.158 g) for 28 days, the products were starting ketone 1
(0.109 g, conversion 67%), compound 2 (0.124 g, yield 44% based
on converted ketone 1), and compound 3 (0.036 g, yield 11% based on
converted ketone 1). For compound 3 [a]52800 = +20 (c 1.2, CHCl3).
In the presence of 14%Met-Csb. Similarly, the reaction of malono-
nitrile (0.158 g, 2.4 mmol) with ketone 1 (0.104 g, 0.6 mmol) in the
presence of 14%Met-Csβ (0.108 g) for 18 days gave compound 2
(0.007 g, yield 5%) and racemic compound 3 (0.042 g, yield 25%).
In the presence of caesium L-methionate. Similarly, in the reaction of
malononitrile (0.339 g, 5.2 mmol) and ketone 1 (0.336 g, 1.9 mmol) in
the presence of caesium methionate (0.173 g) for 18 days, the product
was compound 2 (0.045 g, yield 10%) and racemic compound 3 (0.080 g,
yield 15%).
When the reaction of benzalacetone 4 with malononitrile was
carried out in the presence of caesium L-methionate, we isolated
racemic compounds 5 and 6, in 7% yield each. A similar result
was obtained using caesium oxides modified with L-methionine
as catalysts (yields of racemic products 5 and 6 were 9% for
§
A solution of malononitrile (0.032 g, 0.5 mmol) in diethyl ether (1 ml)
was added to 8%Met-Csβ (0.01 g). The ether was distilled off, and
compound 2 (0.016 g, 0.07 mmol) was added. The mixture was kept at
20 °C for 18 days, and treated with ethyl acetate as an extractant. The
reaction products were separated on a silica gel column using hexane as
an eluent with a gradient of diethyl ether from 0 to 100%. The products
were starting dinitrile 2 (0.002 g, conversion 88%) and racemic com-
pound 3 (0.002 g, yield 9% based on converted dinitrile 2).
Mendeleev Commun. 2006 203
Suslov, Evgeniy V.
