Kinetic study of Michael addition catalyzed by N-methylimidazole in ionic liquids: Residual N-methylimidazole in ionic liquids as a strong base

Article abstract of DOI:10.1002/ejoc.200800621

Michael addition of malonodinitrile to chalcone was studied in classical organic solvent as well as in some common ionic liquids (ILs). Kinetic studies proved that the reaction proceeded faster in the studied ILs than in molecular solvents. N-methylimidaz

Full text of DOI:10.1002/ejoc.200800621

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200800621
Kinetic Study of Michael Addition Catalyzed by N-Methylimidazole in Ionic
Liquids: Residual N-Methylimidazole in Ionic Liquids as a Strong Base
[a]
[b]
[a]
[b]
ˇ
ˇ
ˇ
Mária Meciarová,* Marek Cigán,* Stefan Toma, and Anton Gáplovský
Keywords: Ionic liquids / Michael addition / Basicity / Kinetics
Michael addition of malonodinitrile to chalcone was studied
in classical organic solvent as well as in some common ionic
liquids (ILs). Kinetic studies proved that the reaction pro-
ceeded faster in the studied ILs than in molecular solvents.
N-methylimidazole was found to be a better basic catalyst in
some ILs than in conventional solvents. Residual N-methyl-
imidazole in imidazolium-based ILs can act as a basic cata-
lyst.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
condensation[13] between carboxylic esters and acid chlo-
rides or acids, and other[14] reactions. Small amounts (1–
5 mol-%) of N-methylimidazole, as a basic impurity signifi-
cantly affected H/D exchange in the synthesis of perdeuter-
ated ILs.[15]
ILs have been described as solvent media able to increase
amine basicity with respect to conventional solvents.[16]
This is in accord with our observation[1] that -proline and
piperidine were much better catalysts for Michael additions
of methylene-active compounds to chalcone in ILs than in
dichloromethane. Imidazolium ILs have been described as
polymeric hydrogen-bonded supramolecules, and mixed
with other molecules they can be regarded as nanostruc-
tured materials.[17] As a consequence, generally occurring
solvation processes could be drastically reduced or absent
in this media,[18] and the basic strength of the amines could
be much higher in ILs than in conventional organic sol-
vents.[16]
The aim of the present paper was to examine the activity
of N-methylimidazole both in some common ILs as well as
in classical organic solvents. It has been proved that ad-
dition of molecular solvent into IL partially destroyed its
polymeric structure and enhanced its solvation ability,[16]
but the solvation ability of a mixture of ILs has not been
described before. So, the further aim of our work was to
examine if the catalytic effect of N-methylimidazole would
changed in the mixtures of some imidazolium ILs.
Introduction
In our previous paper,[1] we demonstrated that Michael
additions of malonodinitrile as well as several other methyl-
ene-active compounds to chalcone proceeded well in some
pure ionic liquids (ILs), without the addition of any cata-
lyst. In that paper we speculated that the the acidity of
methylene-active compounds could be influenced by ILs. In
contrast, no product was formed in THF in the Pd-cata-
lyzed α-allylation of 2-phenylpropanal with allyl ethyl car-
bonate run in IL without any additional basic catalyst.[2]
Nucleophilic aromatic substitutions[3] and synthesis of
quinazoline derivatives[4] are other examples of reactions
that proceeded well in ILs without additional catalysts, but
Michael addition reactions of carbonyl compounds to (E)-
β-nitrostyrene catalyzed by N-toluensulfonyl--proline
amide in acidic ILs did not proceed.[5]
The key step in the synthesis of most ILs is alkylation of
N-methylimidazole with the corresponding 1-haloalkane. It
was proved that unreacted N-methylimidazole in ILs can
act as a coordinating base, which is difficult to remove from
the IL.[6] Giernoth[7] described that several ILs can contain
up to 1% of N-methylimidazole, which could explain our
observation that Michael addition of C nucleophiles in ILs
proceed without additional basic catalyst.[1]
N-Methylimidazole is known to be an effective catalyst
for the aza-Michael reaction of N-heterocycles,[8] the enzy-
matic acylation of ribavirin,[9] esterification and amide for-
mation reactions,[10] aldol reactions of silyl enol ether with
aldehydes,[11] the Baylis–Hillman reaction,[12] the Claisen
Results and Discussion
[a] Department of Organic Chemistry, Faculty of Natural Sciences,
Comenius University
We have studied Michael addition of malonodinitrile (2)
to chalcone (1) in various ILs as well as conventional or-
ganic solvents (Scheme 1). The rate constants of the reac-
tions were determined spectrophotometrically and their val-
ues are given in the Table 1.
Mlynská dolina CH2, 842 15 Bratislava, Slovakia
E-mail: mmeciarova@fns.uniba.sk
[b] Institute of Chemistry, Faculty of Natural Sciences, Comenius
University
Mlynská dolina CH2, 842 15 Bratislava, Slovakia
E-mail: cigan@fns.uniba.sk
4408
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 4408–4411

Kinetic Study of Michael Addition Catalyzed in Ionic Liquids
tries 9–11, 14, 15). To confirm the hypothesis about residual
N-methylimidazole in imidazolium-based ILs, we per-
formed the addition of 2 to 1 in [bmim]PF6, which was
extracted several times with HCl (10 vol.-% in water) before
the reaction. No reaction (k = 0 s–1) was observed during
4 h at room temperature in such a solvent (Table 1, En-
try 13), thus confirming our assumption.
Scheme 1. Michael addition of malonodinitrile to chalcone.
The Michael addition proceeded very well in [bmmim]-
PF6 (1-butyl-2,3-dimethylimidazolium hexafluorophos-
phate), which is an IL with a methyl group instead of a
hydrogen atom at C(2). Reaction without any additional
catalyst proceeded with k = 4.78ϫ10–4 s–1 and with
5 mol-% of N-methylimidazole with k = 5.50ϫ10–4 s–1
(Table 1, Entries 17 and 18). It is well known that the C(2)
proton of the 1,3-dialkylimidazolium cation is acidic; the
pKa of the simple imidazolium cation in DMSO and H2O
was found to be in the range of 21–24.[17] The acidic C(2)
proton in [emim]SO4Et and [bmim]PF6 is probably able to
interact with a basic catalyst through hydrogen bonding
and decreases its activity.
Table 1. Solvent effect on the reaction rate of the Michael addition
of malonodinitrile (2) to chalcone (1) catalyzed by N-methylimid-
azole.
Entry Solvent
Catalyst
[mol-%]
k (ϫ10–4)
t1/2
[s]
[s–1]
[a]
1
2
3
4
5
6
7
8
CH2Cl2
CH2Cl2
CH3CN
CH3CN
CH3CN
THF
THF
THF
[emim]SO4Et
[emim]SO4Et
[emim]SO4Et
[bmim]PF6[c]
[bmim]PF6
[bmim]PF6
[bmim]PF6
[bmim]PF6
[bmmim]PF6
[bmmim]PF6
[empyr]SO4Et
[empyr]SO4Et
[empyr]SO4Et
10
1[b]
5
10
1[b]
0
5
10
0
5
10
0
0
5
–
–
–
[a]
–
0.30
0.43
6.72
1.00
1.26
1.52
0.16
1.80
2.12
0.17
23105
16120
1031
6931
5501
4560
43475
3851
3270
40066
–
4077
2567
366
1450
1260
41259
11438
27182
9
10
11
12
13
14
15
16
17
18
19
20
21
Reaction in the IL [empyr]SO4Et (1-ethyl-3-methylpyridi-
nium ethylsulfate) without any catalyst proceeded similarly
to those in [emim]SO4Et and [bmim]PF6, k = 0.17ϫ10–4 s–1
(Table 1, Entry 19). Addition of N-methylimidazole
(5 mol-%) had a smaller influence on the reaction rate (k =
0.61ϫ10–4 s–1) than addition of the same amount of this
catalyst to [emim]SO4Et or [bmim]PF6. It is probably due
to a less organized structure of pyridinium ILs in compari-
son to imidazolium ILs.[16,17] Michael addition of 2 to 1 in
[empyr]SO4Et with 5 mol-% of pyridine as catalyst pro-
[a]
–
1.70
2.70
18.90
4.78
5.50
0.17
0.61
0.26
10
1[b]
0
5
0
5
5[d]
[a] No reaction was observed during 4 h. [b] Piperidine was used as ceeded with a rate constant of 0.26ϫ10–4 s–1 (Table 1, En-
a catalyst. [c] IL was extracted with aqueous HCl before reaction.
[d] Pyridine was used as a catalyst.
try 21).
Comparison of the reaction rate for the Michael addition
of malonodinitrile (2) to chalcone (1) in various solvents
No reaction was observed in CH2Cl2 neither with the catalyzed by 5 mol-% of N-methylimidazole is depicted in
addition of N-methylimidazole (10 mol-%) nor with the ad- the Figure 1.
dition of piperidine (1 mol-%). The rate constant of the re-
It was of interest also to check the basicity of piperidine
action in CH3CN was k = 0.30ϫ10–4 s–1, when 5 mol-% of in ILs. In accordance with previous studies[1,19] the basic
N-methylimidazole was used, and it increased (k = strength of piperidine was found to be much higher in some
0.43ϫ10–4 s–1) when 10 mol-% of the catalyst was used.
The most common solvent for Michael addition of malono-
dinitrile to chalcone is THF. The reaction proceeded in
THF without any catalyst with a rate constant of
1.00ϫ10–4 s–1. Reactions with 5 and 10 mol-% of N-
methylimidazole proceeded more quickly with
k =
1.26ϫ10–4 s–1 and k = 1.52ϫ10–4 s–1, respectively.
The test reaction of 1 with 2 proceeded in [emim]SO4Et
(1-ethyl-3-methylimidazolium ethylsulfate) and [bmim]PF6
(1-butyl-3-methylimidazolium hexafluorophosphate) with-
out any additional catalyst with rate constants of
0.16ϫ10–4 s–1 and 0.17ϫ10–4 s–1, respectively. We assumed
that some residual N-methylimidazole catalyzed these reac-
tions. In contrast, the influence of a basic catalyst was more
impressive in ILs than in THF. Addition of N-methylimid-
azole to the reactions mixtures in ILs increased the reaction
rate to 1.80ϫ10–4 s–1 (5 mol-%) and 2.12ϫ10–4 s–1 (10 mol-
%) for [emim]SO4Et and to 1.70ϫ10–4 s–1 (5 mol-%) and
Figure 1. Comparison of the reaction rate for the Michael addition
of malonodinitrile (2) to chalcone (1) in various solvents catalyzed
2.70ϫ10–4 s–1 (10 mol-%) for [bmim]PF6 (Table 1, En- by 5 mol-% of N-methylimidazole.
Eur. J. Org. Chem. 2008, 4408–4411
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4409

ˇ
M. Mecˇiarová, M. Cigánˇ, S. Toma, A. Gáplovský
SHORT COMMUNICATION
common ILs than in conventional organic solvent. No reac-
tion of 1 with 2 under piperidine (1 mol-%) catalysis was
observed during 4 h in CH2Cl2. The rate constant of the
reaction with 1 mol-% piperidine in CH3CN was k =
6.72ϫ10–4 s–1, whereas the reaction (with the same amount
of catalyst) proceeded in [bmim]PF6 with the rate constant
k = 18.90ϫ10–4 s–1 (Table 1, Entries 2, 5, and 16).
Recently, it was described that ILs are nanostructurally
organized with ionic networks and nonpolar regions. There
are “blocks” along the network of the mixture that are or-
dered in the same way as in the neat liquids, with clustering
of the alkyl groups into nonpolar domains.[20,21] D’Anna[16]
described that the addition of dioxane into [bmim]BF4 par-
tially destroys its polymeric structure and enhance its sol-
vation ability. A decrease in amine basicity in [bmim]BF4/
dioxane binary mixture was detected. Therefore, we won-
dered what would be the reaction rate for the addition of 2
to 1 in some binary mixtures of imidazolium-based ILs. The
results are collected in Table 2. Rate constants of the
Conclusions
We proved that residual N-methylimidazole in several ILs
can act as a basic catalyst. Its basicity, as well as the basicity
of piperidine is higher in ILs than in conventional solvents.
For reactions studied in ILs, residual amounts of N-methyl-
imidazole or the acidity of the IL itself can have a profound
effect on the outcome of the reaction.
Experimental Section
General: The ILs [bmim]PF6, [bmmim]PF6, and [bmim]N(CN)2
were purchased from Merck (high purity), [emim]SO4Et was pur-
chased from Io-li-tec (99+%), and [empyr]SO4Et was purchased
from Solvent Innovation (99%). ILs were used without any purifi-
cation.
General Procedure: Chalcone (1; 0.208 g, 1.0 mmol) was added to
the IL (2 mL), and the mixture was irradiated in an ultrasonic
cleaning bath for 15 min to dissolve chalcone. Malonodinitrile (2)
reaction with 5 mol-% of N-methylimidazole in [bmim]PF6 (0.660 g, 10.0 mmol) and the catalyst (0, 5, and 10 mol-%, respec-
and [emim]SO4Et were k = 1.70ϫ10–4 s–1 and k =
tively) were then added, and the reaction mixture was stirred for 2
or 4 h at room temperature. Aliquots (0.1 mL) of the reaction mix-
ture were removed at 20 min intervals and diluted with CH2Cl2 to
100 mL.
1.80ϫ10–4 s–1, respectively, but the reaction rate signifi-
cantly dropped (k = 1.03ϫ10–4 s–1) in a mixture of these
solvents (Table 2, Entries 1, 2, and 5). Similar results were
The kinetic measurements were performed in a quartz spectro-
photometric cell (l = 0.1 cm) by using an Agilent 8453 UV/Vis spec-
trophotometer. The rate constants k were determined by monitor-
ing the decrease of chalcone in dilute reaction mixture
(0.5ϫ10–3 mmol/mL) by detecting the decrease in its absorption
band at 308 nm (wavelength where chalcone has its maximum ab-
sorption in CH2Cl2). At very low concentration of chalcone, on the
basis of the Beer rule and its linear relation between the concentra-
tion of chalcone and its absorbance, a first-order disappearance of
chalcone is observed. The slope of the plot is the pseudo-first order
constant k and t1/2 is the half-life time of the Michael addition.
obtained when the reaction was performed in a mixture of
[emim]SO4Et and [bmim]N(CN)2. The rate constant was k
= 1.11ϫ10–4 s–1, whereas for neat ILs it was k =
1.80ϫ10–4 s–1 ([emim]SO4Et) or
k
=
3.69ϫ10–4 s–1
{[bmim]N(CN)2}. In contrast, the reaction rate in the mix-
ture of [bmim]N(CN)2 and [bmim]PF6 decreased just
slightly (k = 3.39ϫ10–4 s–1) in comparison with neat
[bmim]N(CN)2 (k = 3.69ϫ10–4 s–1). Similarly, the reaction
rate in a mixture of [bmim]PF6 and [bmmim]PF6 dropped
to 5.31ϫ10–4 s–1, whereas in neat [bmmim]PF6 it was
5.50ϫ10–4 s–1. These results can be explained either by dis-
tortion of solvent structure of individual ILs or by the fact
that ILs with the SO4Et anion can, in the presence of traces
amount of water, hydrolyze to the HSO4 anion, which can
protonate the residual N-methylimidazole. Our observa-
tions are in accordance with the results of D’Anna[22] that
the effect of binary mixtures [bmim]X/cosolvent on the re-
action course cannot be ascribable to simple polarity, vis-
cosity, or conductivity effects, but a whole set of parameters
has to be considered.
Acknowledgments
Our thanks for the financial support are due to the Slovak Grant
Agency VEGA (grant No. 1/3569/06) and the CATAFLUOR
7FPEU project.
ˇ
[1] M. Mecˇiarová, S. Toma, Chem. Eur. J. 2007, 13, 1268–1272.
ˇ
ˇ
[2] M. Hutka, S. Toma, Monats. Chem. 2007, 138, 1175–1179.
[3] I. Newington, J. M. Perez-Arlandis, T. Welton, Org. Lett. 2007,
9, 5247–5250.
[4] J. Chen, W. Su, H. Wu, M. Liu, C. Jin, Green Chem. 2007, 9,
972–975.
Table 2. Study of the reaction rate for the Michael addition of ma-
lonodinitrile (2) to chalcone (1) in various IL and their mixtures
catalyzed by 5 mol-% of N-methylimidazole.
ˇ
[5] M. Mecˇiarová, K. Hubinská, S. Toma, A. Berkessel, Monats.
Chem. 2007, 138, 1181–1186.
[6] J. D. Holbrey, K. R. Seddon, R. Wareing, Green Chem. 2001,
Entry Solvent
k (ϫ10–4) [s–1] t1/2 [s]
3, 33–36.
[7] R. Giernoth, D. Bankmann, Tetrahedron Lett. 2006, 47, 4293–
4296.
[8] B. K. Liu, Q. Wu, X. Q. Qian, D. S. Lv, X. F. Lin, Synthesis
2007, 2653–2659.
[9] B. K. Liu, Q. Wu, J. M. Xu, X. F. Lin, Chem. Commun. 2007,
295–297.
[10] H. Nakatsuji, J. Morita, T. Misaki, Y. Tanabe, Adv. Synt. Catal.
2006, 348, 2057–2062.
1
2
3
4
5
6
7
8
[emim]SO4Et
[bmim]PF6
[bmim]N(CN)2
[bmmim]PF6
[emim]SO4Et + [bmim]PF6
[emim]SO4Et + [bmim]N(CN)2
[bmim]N(CN)2 + [bmim]PF6
[bmim]PF6 + [bmmim]PF6
1.80
1.70
3.69
5.50
1.03
1.11
3.39
5.31
3851
4077
1879
1786
6730
6245
2045
1305
[a]
[a]
[a]
[a]
[11] H. Hagiwara, H. Inoguchi, M. Fukushima, T. Hoshi, T. Su-
zuki, Tetrahedron Lett. 2006, 47, 5371–5373.
[a] Mixture 1:1 (vol./vol.).
4410
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 4408–4411

Kinetic Study of Michael Addition Catalyzed in Ionic Liquids
[12]
ongia Lopes, A. A. H. Pádua, J. Phys. Chem. B 2006, 110,
3330–3335.
a) P. Narender, B. Gangadasu, M. Ravinder, U. Srinivas,
G. Y. S. K. Swamy, K. Ravikumar, V. Jayathirtha Rao, Tetrahe-
dron 2006, 62, 954–959; b) Y.-S. Lin, Ch.-W. Liu, T. Y. R. Tsai,
Tetrahedron Lett. 2005, 46, 1859–1861.
T. Misaki, R. Nagase, K. Matsumoto, Y. Tanabe, J. Am. Chem.
Soc. 2005, 127, 2854–2855.
a) A. Iida, S. Nakazawa, T. Okabayashi, A. Horii, T. Misaki,
Y. Tanabe, Org. Lett. 2006, 8, 5215–5218; b) H. Ding, Ch. Ma,
Y. Yang, Y. Wang, Org. Lett. 2005, 7, 2125–2127.
[18] a) F. D’Anna, V. Frenna, R. Noto, V. Pace, D. Spinelli, J. Org.
Chem. 2005, 70, 2828–2831; b) F. D’Anna, V. Frenna, R. Noto,
V. Pace, D. Spinelli, J. Org. Chem. 2006, 71, 9637–9642; c) C.
Thomazeau, H. Olivier-Bourbigou, L. Magna, S. Luts, B. Gil-
bert, J. Am. Chem. Soc. 2003, 125, 5264–5265.
[19] S. Chowdhury, R. S. Mohan, J. L. Scott, Tetrahedron 2007, 63,
2363–2389.
[20] D. Xiao, J. R. Rajian, S. Li, R. A. Bartsch, E. L. Quitevis, J.
Phys. Chem. B 2006, 110, 16174–16178.
[21] C. Chiappe, Monats. Chem. 2007, 138, 1035–1043.
[22] F. D’Anna, V. Frenna, S. La Marca, R. Noto, V. Pace, D. Spi-
nelli, Tetrahedron 2008, 64, 672–680.
[13]
[14]
[15] R. Giernoth, D. Bankmann, Eur. J. Org. Chem. 2008, 2881–
2886.
[16] F. D’Anna, R. Noto, Tetrahedron 2007, 63, 11681–11685.
[17] a) J. Dupont, P. A. Z. Suarez, R. F. De Souza, R. A. Burrow,
J.-P. Kintzinger, Chem. Eur. J. 2000, 6, 2377–2382; b) P. Migow-
ski, J. Dupont, Chem. Eur. J. 2007, 13, 32–39; c) J. N. A. Can-
Received: June 24, 2008
Published Online: August 1, 2008
Eur. J. Org. Chem. 2008, 4408–4411
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4411