Mendeleev
Communications
Mendeleev Commun., 2007, 17, 230–231
Imidazolium salts as specific catalysts for Michael addition
to nitroalkanes: a structure–catalyst efficiency relationship
Electron A. Mistryukov
N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow,
Russian Federation. Fax: +7 495 135 5328
DOI: 10.1016/j.mencom.2007.06.016
Imidazolium catalyst structures were optimised in relation to the Michael addition of acrylonitrile to secondary nitroalkanes.
Traditionally, strong bases or their combinations with common
phase-transfer reagents are employed as catalysts for the Michael
an explosion-like process. If at the exothermal stage the tem-
perature is controlled by cooling at 40–50 °C, the reaction is
complete after additionally stirring for 1 h from the moment
when the reaction subsides.†
1
addition to nitroalkanes. For example, potassium carbonate,
2
3
4
5
potassium fluoride, cesium fluoride, alkoxides and amines
have been used in catalytic, stoichiometric or excess amounts.
In connection with the synthesis of γ-aminobutyric acid, our
interest was attracted to the specific reaction of acrylonitrile (or
acrylates) with secondary nitroalkanes (2-nitropropane 1 and
nitrocyclohexane 2). Here, the efficiency of a particular catalyst
can be easily monitored by measuring a temperature jump after
the addition of a catalyst to the mixture of reagents (reaction is
strongly exothermic). Evaluation of the traditional catalysts
such as DBU, potassium isopropylate, potassium hydroxide–
polypropylene glycol has shown some general features of these
bases performance.
A comparison of the thermal effects produced by the ‘classical’
n
phase-transfer reagent Bu NBr and imidazolium salt 3a indicates
4
that the efficiency of catalyst 3a is incomparably higher. More-
n
over, besides a modest temperature rise, Bu NBr stops func-
4
tioning short time after addition when most of the reagents are
still unchanged. Dealkylation–neutralisation may be responsible
for catalyst deactivation. Therefore, the assumption that imida-
zolium salts act only as phase-transfer reagents for potassium
carbonate may be discarded.
The properties of other lipophilic imidazolium derivatives 3b
‡
and 3c support this argument. Despite the similarities in the
Really, common to all of these bases is the initiation of side
reactions destroying the catalyst. In our practice, the addition of
a catalyst (e.g. ~1 mol% DBU) to a mixture of 1a or 1b and
acrylonitrile in acetonitrile as a solvent initiated an exothermal
reaction (temperature rise ca. 20 °C above room temperature),
which subsides after a short time (10–15 min). The addition
of a fresh portion of the same catalyst resulted in a new
temperature jump. Several repetitions of the procedure produced
the same (but diminishing) effect. In a search for more adequate
catalysts, we were pleased to find these to be the sterically
crowded imidazolium salts–potassium carbonate combination.
steric bulk with salt 3a, diester 3b and diketone 3c are absolutely
inactive as catalysts for the Michael addition of acrylonitrile to
§
nitroalkanes 1a,b. Moreover, 3b,c block the activity of catalyst
3a, i.e., they act as effective inhibitors of catalyst 3a when used
together.
The reasons of this inhibitory activity of compounds 3b,c
may be the ionisation at the α-carbon atoms to the positively
charged imidazolium ring with the formation of a neutral
†
Syntheses of addition products 2a and 2b. Salt 3a (0.5 g, 2.3 mmol)
was added to the mixture of 1a (13.5 ml, 0.15 mol), acrylonitrile (10 ml,
0.15 mol) and potassium carbonate (1 g, 7.2 mmol) in acetonitrile (10 ml),
and the mixture was stirred with temperature control. After the comple-
tion of heat evolution and stirring for 1 h, filtration and rotary evaporation
R
R
H
R
R
CH CH CN
CH2=CHCN
a,d / K CO
2
2
1
3
gave product 2a as an oil, bp 101–102 °C (1 Torr), yield 85%. H NMR
2
3
NO2
NO2
2
(
250 MHz, [ H ]DMSO) d: 1.6 (s, 6H), 2.26 and 2.60 (2t, 2×2H, J 7 Hz).
6
1
1
a R = Me
b R + R = (CH2)
2a R = Me
2b R + R = (CH2)5
The same procedure was used for the synthesis of 2b as a low melting
5
1
2
solid, mp 42 °C, yield 90%. H NMR (250 MHz, [ H ]DMSO) d: 1.5–2.5
6
X–
(m, 14H).
1
2
3
a: H NMR (250 MHz, [ H ]DMSO) d: 1.63 (s, 18H), 8.15 (s, 2H),
6
N
N
9
.45 (s, 1H).
R
R
‡
Imidazolium salts 3b and 3c. The mixture of equimolar quantities of
t
3
3
3
3
a R = Bu , X = Cl
t
imidazole, tert-butyl chloroacetate and potassium carbonate was stirred
in acetonitrile at 60 °C for 1 h. Fresh equivalents of the base and chloro
ester were added and after heating at 90 °C for 2 h filtration and rotary
evaporation left the product (glassy residue), which after crystallisation
from acetonitrile–ethyl acetate gave imidazolium salt 3b, yield 60%,
b R = CH COOBu , X = Cl
2
t
c R = CH COBu , X = Br
2
t
d R = CH CH(OH)Bu , X = Br
2
Scheme 1
1
2
mp 125–127°C. H NMR (250 MHz, [ H ]DMSO) d: 1.45 (s, 18H), 5.30
The first experiments have shown that stirring the mixture
of reagents containing acrylonitrile, nitroalkane (1a or 1b),
potassium carbonate and the solvent (acetonitrile) for a long
time (1–2 h) resulted in no temperature change and no product
formation. However, the addition of a small quantity of sterically
crowded imidazolium salt 3a produced an almost immediate
temperature rise (up to the boiling point of the solvent). It is
essential to control the temperature at the early stage to prevent
6
(s, 4H), 7.84 (s, 2H), 9.4 (s, 1H).
For the synthesis of salt 3c, a one-stage procedure was used with one
equivalent of imidazole and two equivalents of both bromopinacolone
and potassium carbonate in acetonitrile (cooling at the exothermal stage and
heating at 50 °C for 2 h), yield 60%. H NMR (250 MHz, [ H ]DMSO)
d: 1.23 (s, 18H), 5.76 (s, 4H), 7.72 (s, 2H), 9.20 (s, 1H).
§ Michael additions. All experiments were conducted in a round bottom
1
2
6
flask with an overhead stirrer (1500 rpm).
–
230 –