Cycloaddition of CO2 and epoxides catalyzed by imidazolium bromides under mild conditions: Influence of the cation on catalyst activity

Article abstract of DOI:10.1039/c3cy01024d

The synthesis and characterization of a library of imidazolium-based compounds of the type [R¹R²R³im]Br (R ¹ = H, CH₃, benzyl (Bz), 1-(2,3,4,5,6-pentafluoro)benzyl (Bz^F5); R² = H

Full text of DOI:10.1039/c3cy01024d

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Cycloaddition of CO2 and epoxides catalyzed by
imidazolium bromides under mild conditions:
influence of the cation on catalyst activity†
Cite this: Catal. Sci. Technol., 2014,
4, 1749
a
a
Michael H. Anthofer,‡ Michael E. Wilhelm,‡ Mirza Cokoja, a Iulius I. E. Markovits,a
*
Alexander Pöthig,a János Mink,bc Wolfgang A. Herrmanna and Fritz E. Kühna
The synthesis and characterization of
a library of imidazolium-based compounds of the type
[R1R2R3im]Br (R1 = H, CH3, benzyl (Bz), 1-(2,3,4,5,6-pentafluoro)benzyl (BzF5); R2 = H, CH3, C2H5; R3
=
n-butyl, n-octyl) are reported. All compounds were characterized by means of 1H and 13C NMR spectroscopy,
elemental analysis and single crystal X-ray diffraction in the case of 1-BzF5-3-n-octylimidazolium bromide
and examined as catalysts for the cycloaddition of carbon dioxide and propylene oxide (PO) to yield propylene
carbonate (PC). The catalyst screening shows the influence of different substituents (steric and ion pairing
effects) on the catalytic activity. It is shown that the most active catalyst, 1-BzF5-3-n-octylimidazolium
bromide (10), gives very good conversions and carbonate yields at 70 °C and below 5 bar CO2 pressure,
which is the lowest temperature for the synthesis of PC from PO and CO2 using metal-free catalysts.
Catalyst 10 can be recycled and reused at least 10 times without loss of yield and activity.
Received 7th December 2013,
Accepted 17th February 2014
DOI: 10.1039/c3cy01024d
www.rsc.org/catalysis
reactant in the chemical industry, it is evident that CCR alone
will most certainly not solve the ‘greenhouse gas’ problem in
Introduction
Carbon dioxide may be considered as one of the key renew-
able carbon sources of the future, which could – provided
sufficient energy or energy conserving processes are available –
in the long term replace the currently exploited fossil carbon
sources as a feedstock for the chemical industry.1 Many con-
cepts for the valorization of CO2 are currently being discussed,
among which CO2 capture and recycling (CCR) is regarded
to be the most promising one.2–4 Nevertheless, CO2 is an
unreactive molecule, and its activation is associated with a
considerable energy barrier. Further, CO2 is one of the major
greenhouse gases, which leads to the tempting misbelief that
by CCR the CO2 content in the atmosphere could (significantly)
be reduced. Since the worldwide annual amount of anthropo-
genic CO2 emission (industry, combustion, etc.) is more than
a hundred times higher than the total amount of CO2 used as
the foreseeable future. Hence, current research rather focuses
on the question of how the CO2 produced by human activities
can be recycled and used as a C1-feedstock without leaving a
further carbon footprint. Several possibilities are being consid-
ered to reduce the industry- and technology-based CO2 emis-
sion and to capture and store CO2.1,5 However, these concepts
are either still in the nascent stage or they consume consider-
able amounts of energy, thus are far from large-scale applica-
bility. For this reason, the development of high-performance
molecular homogeneous catalysts is of particular interest, as
they usually operate under much milder conditions compared
to heterogeneous catalysts.6,7
To date, only few catalytic transformations of CO2 in the
homogeneous phase have the potential to be used on an
industrial scale with positive CO2 balance. One of them is the
reaction of epoxides, above all the industrially important
propylene oxide (PO),8–10 with CO2 to yield cyclic carbonates
and polycarbonates. Particularly, propylene carbonate (PC)
and poly(propylene carbonate) plastics are very important
in industry.8 As catalysts, mostly M–carboxylate, –salen and
–porphyrin/–phthalocyanin complexes are used (M = Zn, Al,
Sn, Cr, Fe, Co).11–13 For the copolymerization of PO and CO2,
a wide variety of molecular transition metal catalysts is known.14–18
Furthermore, the carbonate synthesis can be carried out with-
out a metal catalyst as well, using nucleophilic reagents, such
as halides or N-donor bases.19 The use of halides is well docu-
mented, and meanwhile, ILs containing halides as anions or
a Chair of Inorganic Chemistry/Molecular Catalysis, Catalysis Research Center,
Technische Universität München, Ernst-Otto-Fischer-Straße 1, D-85747 Garching
bei München, Germany. E-mail: mirza.cokoja@ch.tum.de; Fax: +49 89 289 13473;
Tel: +49 89 289 13478
b Hungarian Academy of Sciences, Chemical Research Center, Pusztszeri u. 59-67,
1025 Budapest, Hungary
c Faculty of Information Technology, University of Pannonia, Egyetem u. 10,
8200 Veszprém, Hungary
† Electronic supplementary information (ESI) available: X-ray single crystal
data of compound 10. CCDC 928238. For ESI and crystallographic data in CIF
or other electronic format see DOI: 10.1039/c3cy01024d
‡ These authors have equally contributed to this work.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 1749–1758 | 1749

View Article Online
Paper
Catalysis Science & Technology
functionalized ILs are known to catalyze the cycloaddition,
bearing the advantage of facilitated product separation due to
the low volatility of the ILs.19–23 Although metal-free catalysts
are generally considered to be less active than molecular tran-
sition metal catalysts, they exhibit enormous potential regard-
ing the introduction of functional groups and the tailoring
of electronic properties, which even surpasses the portfolio
of ligands for transition metal centers. Some examples of
organocatalytic systems are covalent organic frameworks,24
ionic liquids19,25 and other supported catalytic matrices.26–31
The most active organocatalysts, to date, are tandem catalysts
containing phenol derivatives and nucleophiles (N-donor
bases or halides), which are efficient even at temperatures
below 45 °C.32–35 The activity relies on the interaction of the
hydroxyl group with the epoxide. In addition, new catalytic
systems involving a cooperative or a carboxylic acid group and
a nucleophilic reagent such as a halide or N-donor bases are
known in the literature.36 Yet, very often turnover frequencies
(TOF) and numbers (TON) are not reported and catalyst
recycling has not been studied. In the organocatalytic cycload-
dition of CO2 and epoxide, the efficiency of the reaction is
mainly governed by the nucleophilicity of the catalyst.37 In
the first step of the reaction, a ring carbon atom of the epox-
ide is attacked by the nucleophile, yielding an alcoholate
species. Subsequently, the alcoholate itself acts as a nucleo-
phile, attacking the CO2 carbon atom and resulting in the for-
mation of an open carbonate intermediate, which undergoes
‘back-biting’, leading to the cyclic carbonate. Protic substitu-
ents at the catalyst, such as –COOH or –OH groups, can prin-
cipally facilitate the ring opening by hydrogen bonding to the
epoxide.25,26,29
Analytical equipment
Microanalyses of the obtained products were performed at the
Mikroanalytisches Labor of the Technische Universität München
in Garching using a HEKATech EURO EA. 1H, 13C, and 19F NMR
spectra were recorded in CDCl3 or [D6]-DMSO using a 400 MHz
Bruker Avance DPX-400 spectrometer (400 MHz for 1H, 101 MHz
for 13C, and 376 MHz for 19F). IR spectra were recorded using
a Varian IR FT670 that was equipped with an ATR cell
(diamond crystal). Catalytic runs were monitored by gas chroma-
tography using a Hewlett-Packard instrument HP 5890 Series II
equipped with a FID, a Supelco column Alphadex 120 and a
Hewlett-Packard integration unit 3396 Series II. The melting
points of the compounds were determined using a melting
point meter MPM-H2. The absorption of CO2 in the ILs was
determined by thermogravimetric analysis using a NETZSCH
STA 409 PC. Fast Atom Bombardment (FAB) mass spectros-
copy was carried out using a Finnigan MAT (type MAT90) with
xenon as the ionization gas in a 3-nitrobenzyl alcohol matrix.
Synthesis of catalysts 1–10
1-Octylimidazole,39 1-methyl-2-ethylimidazole,40 1-methyl-3-butyl
imidazolium bromide (1),41 1,2-dimethyl-3-butylimidazolium
bromide (2),42 1-methyl-3-octylimidazolium bromide (4),43
1,2-dimethyl-3-octylimidazolium bromide (5),44 1-benzyl-3-
butylimidazolium bromide (7),45 and 1-(2,3,4,5,6-pentafluoro)
benzyl-3-butylimidazolium bromide (9) (ref. 45) were synthe-
sized according to literature procedures.
General synthesis of alkyl and benzyl substituted
imidazolium bromides 3, 6 and 8
Very recently, we have shown that in ionic liquid solutions,
anions such as perrhenate [ReO4]−, which is notoriously unreactive
in aqueous or organic solutions, show enhanced reactivity as
nucleophiles, especially when the cation allows only weak ion
pairing.38 Our present study focuses on the effect of different
substituents on imidazolium cations on the ion pairing, and
the resulting effect on the nucleophilicity of the counterion
and hence on the catalytic activity for the epoxide ring open-
ing. Herein, we report the synthesis and characterization of
a series of imidazolium bromides and their catalytic activity
in the reaction of CO2 and various epoxides to yield cyclic car-
bonates under mild conditions.
To a stirred solution of 10.0 mmol of imidazole in THF,
11.0 mmol of the alkyl bromide was added dropwise and the
solution was refluxed for 24 h. The solvent was distilled and
the resulting residue was washed with ether and ethyl ace-
tate. After drying in vacuum at 70 °C for 6 h, the imidazolium
bromides 3, 6 and 8 were obtained.
1-Methyl-2-ethyl-3-butylimidazolium bromide (3): white solid;
yield 90%; 1H NMR (400 MHz, CDCl3): δ(ppm) = 7.80
3
3
(d, J(H,H) = 2.2 Hz, 1H, CH), 7.50 (d, J(H,H) = 2.2 Hz, 1H,
3
CH), 4.19 (t, J(H,H) = 7.6 Hz, 2H, CH2), 4.06 (s, 3H, NCH3),
3.19 (q, 3J(H,H) = 7.8 Hz, 2H, CH2), 1.93–1.78 (m, 2H, CH2),
1.50–1.36 (m, 2H, CH2), 1.32 (t, 3J(H,H) = 7.8 Hz, 3H, CH3),
0.98 (t, 3J(H,H) = 7.3 Hz, 3H, CH3); 13C NMR (101 MHz,
CDCl3): δ(ppm) = 123.72, 121.22, 48.75, 36.17, 32.45, 19.92,
17.93, 13.76, 11.86; m.p. 85 °C; elemental analysis (%): calcd.
C 48.59, H 7.75, N 11.33, Br 32.3; found C 48.28, H 7.82,
N 11.39, Br 31.1; MS (FAB): m/z calcd. for [C10H19N2+] = 167.2;
found 167.7.
Experimental
General remarks
All preparations were carried out using standard tech-
niques. Benzyl bromide, 2,3,4,5,6-pentafluorobenzyl bromide,
1-methylimidazole and 1-bromooctane were purchased from
ABCR. Methyl iodide, imidazole, 1-butylimidazole, propylene
oxide, styrene oxide, glycidol, cyclohexene oxide, epichlorohy-
drin and 1,2-epoxyoctane were purchased from Sigma Aldrich.
All chemicals were used as received without further purifi-
cation. All ionic liquids were dried under vacuum at 70 °C
for 6 h.
1-Methyl-2-ethyl-3-octylimidazolium bromide (6): white solid;
yield 86%; 1H NMR (400 MHz, CDCl3): δ(ppm) = 7.78
(d, 3J(H,H) = 2.1 Hz, 1H, CH), 7.42 (d, 3J(H,H) = 2.1 Hz, 1H, CH),
4.16 (t, 3J(H,H) = 7.6 Hz, 2H, CH2), 4.06 (s, 3H, CH3), 3.19
(q, 3J(H,H) = 7.7 Hz, 2H, CH2), 1.86 (m, 2H, CH2), 1.45–1.18
(m, 13H, CH2, CH3), 0.88 (t, 3J(H,H) = 7.2 Hz, 3H, CH3); 13C NMR
(101 MHz, CDCl3): δ(ppm) = 123.74, 121.06, 49.01, 36.20,
1750 | Catal. Sci. Technol., 2014, 4, 1749–1758
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
31.86, 30.48, 29.22, 26.67, 22.78, 17.99, 14.28, 11.85; m.p. 54 °C;
elemental analysis (%): calcd. C 55.44, H 8.97, N 9.24, Br 26.4;
found C 55.40, H 9.13, N 9.12, Br 26.2; MS (FAB): m/z calcd.
of ethyl acetate and afterwards the catalyst was precipitated
by the addition of ether, whereas PC stayed in the organic
layer and could be separated by evaporation of the solvent.
The catalyst was dried at 70 °C for 6 h in vacuo. By adding the
same amount of epoxide (10.0 mmol), the next catalytic run
was started.
+
for [C14H27N2 ] = 223.2; found 223.8.
1-Benzyl-3-octylimidazolium bromide (8): white solid; yield
1
91%; H NMR (400 MHz, CDCl3): δ(ppm) = 11.01 (s, 1H, CH),
7.48–7.43 (m, 2H, arom.-H), 7.39 (m, 3H, arom.-H), 7.12
(s, 1H, CH), 7.10 (s, 1H, CH), 5.60 (s, 2H, CH2), 4.28 (t, 3J(H,H) =
7.5 Hz, 2H, CH2), 1.90 (m, 2H, CH2), 1.43–0.92 (m, 10H, CH2),
0.85 (t, 3J(H,H) = 6.7 Hz, 3H, CH3); 13C NMR (101 MHz, CDCl3):
δ(ppm) = 132.92, 129.86, 129.74, 129.29, 121.49, 53.75, 50.57,
31.84, 30.43, 29.21, 29.10, 26.48, 22.78, 14.28; m.p. 69 °C;
elemental analysis (%): calcd. C 61.45, H 7.75, N 7.89, Br 22.7;
found C 61.27, H 7.75, N 7.97, Br 22.9; MS (FAB): m/z calcd.
for [C18H27N2+] = 271.2; found 271.7.
Determination of CO2 solubility in ionic liquids 4, 5, 6, 8 and 10
An exact amount of IL was weighed in an aluminium oxide
crucible and closed. The sample was heated in vacuo in the
thermogravimetric analyzer for 30 min until a constant tem-
perature was reached and the atmosphere was switched to CO2
(1 bar). After 90 min, a constant mass was recorded and the
weight increase was determined. Based on these values, the
mole fraction CO2 to IL was calculated as shown in Table 2.
Synthesis of 1-(2,3,4,5,6-pentafluoro)benzyl-3-octylimidazolium
bromide (10)
Results and discussion
Synthesis and characterisation of imidazolium bromides 1–10
To a stirred solution of 1.80 g of 1-octylimidazole (10.0 mmol,
1.0 equiv.) in THF, 2.87 g of 2,3,4,5,6-pentafluorobenzyl bro-
mide (11.0 mmol, 1.1 equiv.) was added dropwise at 0 °C.
After complete addition of the bromide, the solution was
allowed to warm to r.t. and stirred for 24 h. The solvent was
distilled off and the resulting residue was washed with ether
and n-hexane. 10 was obtained as a white solid after drying
in vacuum at 70 °C for 6 h (4.06 g, 9.2 mmol, 92%); since
one signal of the backbone of the imidazolium ring was cov-
ered by the solvent signal, the spectra of 10 were recorded in
DMSO-d6. 1H NMR (400 MHz, [D6]-DMSO): δ(ppm) = 9.30
(s, 1H, CH), 7.85 (s, 1H, CH), 7.81 (s, 1H, CH), 5.64 (s, 2H, CH2),
Imidazolium salts with varying alkyl group lengths, as
well as with fluorinated substituents, were synthesized since
it was reported that imidazolium-based ionic liquids with
long alkyl chain moieties allow a higher solubility of CO2.46
Further, in order to avoid possible in situ formation of
N-heterocyclic carbenes (NHC), the protons at the 2-position
of the imidazolium ring were replaced by alkyl groups (Me, Et).
All imidazolium bromide catalysts were prepared by treatment
of a THF solution of an imidazole with the respective alkyl
bromide (Scheme 1).
1
The purity of compounds 1–10 was determined by H and
3
3
4.16 (t, J(H,H) = 7.2 Hz, 2H, CH2), 1.77 (p, J(H,H) = 7.1 Hz,
13C NMR spectroscopy and elemental analysis (see the Experi-
mental section). Further, the structure of compound 10 was
confirmed by single crystal X-ray crystallography. The struc-
ture of 10 is shown in Fig. 1 and the intermolecular con-
tacts in the solid state are depicted in Fig. 2. The crystal
structure of 10 shows intermolecular hydrogen bonds between
the imidazolium cation and bromide. Therefore, five different
hydrogen–bromide interactions are observed. As expected, the
strongest contact was obtained from the proton at C(1) of the
imidazolium ring with a distance of 2.584 Å. This distance is
comparable to the known crystal structure of 1, where a value
of 2.446 Å was observed.47 Additionally, the backbone proton
3
2H, CH2), 1.33–1.11 (m, 10H, CH2), 0.86 (t, J(H,H) = 6.7 Hz,
3H, CH3); 13C NMR (101 MHz, [D6]-DMSO): δ(ppm) = 136.63,
122.74, 49.03, 31.06, 29.18, 28.44, 28.21, 25.37, 22.01, 13.89;
19F NMR (376 MHz, [D6]-DMSO): δ(ppm) = −141.41 (dd, 3J(F,F) =
22.9, 7.6 Hz), −152.84 (t, 3J(F,F) = 22.9 Hz), −161.70 (td, 3J(F,F) =
22.9, 7.6 Hz); m.p. 56 °C; elemental analysis (%): calcd. C 48.99,
H 5.03, N 6.35, Br 18.1; found C 49.13, H 5.15, N 6.22, Br 18.7;
+
MS (FAB): m/z calcd. for [C18H22F5N2 ] = 361.2; found 361.6.
Experimental procedure for the catalytic conversion of CO2
and epoxides
A Fisher–Porter bottle was charged with catalyst 10 (1.0 mmol)
and equipped with a magnetic stirring bar. The epoxide
(10.0 mmol) was added under Ar and the Fisher–Porter bottle
was then placed under a constant pressure of CO2 for 1 min
and heated to the desired temperature. After reaction comple-
tion, the bottle was cooled down to 0 °C and the excess CO2
was vented. The reaction mixture was collected by adding
5 mL of chloroform and a sample was collected for GC analy-
sis to determine yield and selectivity.
For recycling studies, propylene carbonate was separated
from the catalyst by adding a mixture of ethyl acetate and ether
(1 : 4). The reaction mixture was dissolved in a minimum amount
Scheme 1 General synthesis of the imidazolium bromide catalysts
1–10 (Me = methyl, Bu = n-butyl, Oct = n-octyl, Bz = benzyl, BzF5
=
1-(2,3,4,5,6-pentafluoro)benzyl).
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 1749–1758 | 1751

View Article Online
Paper
Catalysis Science & Technology
between the fluorine atoms in ortho- and para-positions of
the aromatic ring and the protons at C(3), C(13) and C(15)
could be observed in the crystal structure.
Synthesis of PC from CO2 and PO with catalysts 1–10
The screening of the catalytic activity of the synthesized
imidazolium bromides 1–10 bearing aliphatic, benzyl or
fluorinated benzyl wingtips (Scheme 1) was carried out using
propylene oxide (PO) as the substrate for the cycloaddition
of CO2. Imidazolium-based ionic liquids are known to effi-
ciently catalyse the cycloaddition of epoxides and CO2.22,23
The reactions were performed in neat PO without further sol-
vents, since using solvents (chloroform, dimethyl carbonate,
methyl ethyl ketone) has a detrimental effect on the cata-
lytic activity. In a typical experiment, a Fisher–Porter bottle
(pressure reaction glass vessel) was charged with the catalyst
and PO (catalyst : substrate ratio 1 : 10) and then pressurized
with 4 bar CO2 for 1 min. Subsequently, the mixture was
heated at 70 °C for 22 h. A reaction time of 22 h is necessary
to reach 91% conversion of PO using 10 as the catalyst under
the observed reaction conditions (Fig. 4). Based on these
results, the catalyst screening was carried out for a reaction
time of 22 h to compare catalysts 1–10 and therefore, get a
better insight into the role of the cation on the catalytic
activity. Note that at this temperature, all examined catalysts
are liquid and PO is in the gas phase; hence, at the reaction
temperature, the catalyst also acts as a solvent for the par-
tially dissolved substrates. This allows the investigation of
the influence of the substitution pattern of the cation of the
different catalysts 1–10 without any solvent-related effects.
The conversion of PO was determined by gas chromatogra-
phy (GC). The results of the screening are shown in Table 1.
Overall, it can be seen that all organocatalysts are quite active,
leading to good to very good conversions of PO with a high
selectivity of PC. Due to the absence of water, the formation
of 1,2-propanediol as a side product is suppressed, which is
also observed in several other reports using ionic compounds
as catalysts.22,23 Note that the formation of polycarbonates
has not been observed in any of our catalytic experiments.
The mechanism of the cycloaddition of epoxides with CO2
using tetrabutylammonium halides as catalysts was shown in
several reports.37,49 The epoxide ring is opened by the halide,
forming an oxyanion species, which reacts with CO2 to yield
a cyclic carbonate after intramolecular cyclic elimination
(‘back-biting’). The mechanism with imidazolium halide
catalysts follows an analogous reaction pathway and is illus-
trated in Scheme 2.22,23
Fig.
1 ORTEP view of the structure of compound 10. Thermal
ellipsoids are shown at the level of 50% probability. Selected bond
lengths (Å) and angles (°): C1–N1, 1.323(6); C1–N2, 1.325(6); C2–N1,
1.382(6); C3–N2, 1.375(6); C2–C3, 1.344(7); N1–C11, 1.468(6); N2–C4,
1.465(5); N1–C1–N2, 198.8(4); C1–N1–C2, 108.4(4); C1–N2–C3, 108.6(4);
C3–C2–N1, 107.0(5); C2–C3–N2, 107.1(4); C1–N1–C11, 126.1(4); C1–N2–C4,
127.2(4); C1–N2–C4–C5, 36.0; C2–N1–C11–C12, 58.4.
Fig. 2 Illustration of the crystal packing in compound 10 and the
H⋯Br contacts.
at C(3) shows a rather strong interaction (2.685 Å). Weaker
interactions were detected from a methylene proton at C(4) to
Br− (distances, 2.800 Å and 2.939 Å) and even the n-octyl side
chain shows interaction to the bromide from the proton at
C(11) with a distance of 2.827 Å. These contacts were con-
firmed by Hirshfeld surface analysis (see the ESI,† Fig. F2).
Therefore, for a fluorinated bromide like 10, the contact is
quite strong compared to an imidazolium bromide bearing
fluorinated and non-fluorinated aromatic side chains (distance,
2.750 Å).48 In this case, no contact between the methylene
protons of the BzF5 moiety and Br− was observed. However, the
methylene bridge of the Bz group shows hydrogen bonding to
Br− with a distance of 2.884 Å and, hence, is in a similar range
as determined for 10. The backbone proton in the crystal
structure of 1 shows a distance of 2.873 Å and therefore is
longer than the observed distance in 10. The first carbon
atom of the n-butyl side chain showed a hydrogen bond
to the bromide with a distance of 2.845 Å and is comparable
to that in 10.47 Further, hydrogen bonding interactions
In the absence of a catalyst, no PO conversion could be
detected under the observed reaction conditions (70 °C, 4 bar
CO2 pressure, neat, 22 h). Regarding the PO conversions
obtained with catalysts 1–3, it is notable that substitution of
the acidic ring proton at the 2-position by methyl and ethyl
groups leads to a reduction of the conversion. The contact
between the imidazolium ring proton at the C2-position
and the epoxide oxygen atom facilitates the opening of the
1752 | Catal. Sci. Technol., 2014, 4, 1749–1758
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Table 1 Synthesis of PC from CO2 and PO using catalysts 1–10
Cat.
mmol cat.
mmol PO
p(CO2) [bar]
T [°C]
Conv.a [%]
PC sel. [%]
—
1
2
3
4
5
6
7
8
0
1
1
1
1
1
1
1
1
1
1
10
10
10
10
10
10
10
10
10
10
10
4
4
4
4
4
4
4
4
4
4
4
70
70
70
70
70
70
70
70
70
70
70
0
85
64
69
88
80
71
77
73
86
91
—
>99
>99
>99
>99
>99
>99
>99
>99
>99
>99
9
10
a
Conversion is based on GC analysis.
consistent with the fact that, in general, fluorinated cations
increase the solubility of CO2 in the ionic liquid phase.50
Previous studies show that the CO2 solubility in imidazolium-
based ionic liquids largely depends on the nature of the
anion, rather than that of the cation,51 hinting that in our
case the gas solubility does not have a significant impact on
the different catalytic activities, since in all catalysts, the
anion is bromide. Nevertheless, to shed some light on the
actual impact of the CO2 solubility in the reaction, the CO2
solubility in the ILs 4, 5, 6, 8 and 10 was determined at 70 °C
and 1 bar CO2 pressure using thermogravimetric analysis.
The calculated mole fractions CO2 to IL were calculated and
listed in Table 2.
Scheme 2 Proposed mechanism for the cycloaddition of CO2 with
Very recently, Torralba-Calleja et al. presented a com-
prehensive summary of the solubility of CO2 in various
(imidazolium-based) ionic liquids.52 Nonetheless, despite our
best efforts, no data on imidazolium bromide ionic liquids
could be found and therefore a direct comparison of the
obtained results is difficult. For example, analogues of 1 and
PO using an imidazolium halide catalyst.
epoxide ring by the nucleophilic bromide anion. When the
proton is substituted by an alkyl group, this interaction does
not exist anymore and hence, the relative catalytic activity is
weaker. This trend is corroborated by the relative activities of
the catalyst series 4–6. In contrast, a non-fluorinated benzyl
substituent does not exhibit any beneficial effect compared
with the alkyl moieties (cf. catalysts 7, 8 vs. 1, 4). A positive
catalytic effect can be observed upon fluorinating the benzyl
side chain. The best results were achieved with catalyst 10,
which bears one octyl group and one fluorinated benzyl group
on the imidazolium ring N atoms (91%). These results are
−
4 with BF4 as the anion were examined under slightly differ-
ent conditions (10 bar CO2 pressure, 60 °C). These compounds
show mole fractions (χ) of 0.0895 and 0.121 CO2 to IL,52 which
corroborate the observation of better CO2 solubility upon
increasing the chain length and CO2 pressure.46 However,
when more than one parameter is changed (temperature,
pressure, anion), it is difficult to rank the obtained CO2 solu-
bilities in our catalyst in the series of existing data. The
results show an increase of CO2 solubility upon substitution
at the C2-position of the imidazolium moiety in the order
H (4) < Me (5) < Et (6), which does not point to a correlation
between the CO2 solubility and the catalytic activity. Further-
more, an aromatic wingtip (Table 2, entry 4) increases the sol-
ubility of CO2 in the IL compared to a methyl group (Table 2,
entry 1) and the fluorination of the aromatic ring (Table 2,
entry 5) has only a small influence on the solubility of CO2 in
the IL. In summary, the investigated ILs only dissolve a small
amount of CO2 (approx. 1% compared to the used substrate)
Table 2 CO2 absorption of 4, 5, 6, 8 and 10 at 70 °C under 1 bar CO2
pressure using thermogravimetric analysis
Entry
IL
Mole fraction (χ) CO2 to IL
1
2
3
4
5
4
5
6
8
0.058 0.002
0.077 0.003
0.109 0.001
0.096 0.004
0.102 0.011
10
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 1749–1758 | 1753

View Article Online
Paper
Catalysis Science & Technology
and their differences are not significant. Nevertheless, recent
investigations indicate a significant impact of the cation struc-
ture on CO2 capture in imidazolium-based ionic liquids.53,54
Therefore, the impact on catalysis appears to be negligible,
at least under the conditions applied in our experiments
(70 °C, <5 bar CO2). The presence of a proton at the
C2-position of the imidazolium ring appears to have the
largest impact on the PO conversion using catalysts 1, 4 and
7–10, followed by the steric demand of the substituents at
C1- and C3-positions.
outcome of the investigated reactions is generally considered
to derive from their electrostatic influence.63,64
Nevertheless, based on the catalytic results, the cation has
an influence on the cycloaddition of PO and CO2, most prob-
ably due to different hydrogen bonding interactions of the IL
with PO. Despite our best efforts, we have so far been unable
to grow crystals of compound 10 from a PO solution to deter-
mine the existence of a H-contact of the C2-proton and PO in
the solid state. In order to determine the hydrogen bonding
character between 10 and PO, a FTIR study was carried out
(Fig. 3). The IR spectrum of neat compound 10, shown in
Fig. 3a, exhibits a distinct vibration band at 3064 cm−1, which
becomes broader (from ~20 cm−1 to 60 cm−1 half width) and
shifts to 3040 cm−1 upon addition of an excess of PO to 10 at
room temperature. This frequency shift (24 cm−1) can be
ascribed to the intermolecular interaction of 10 and PO
through hydrogen bonding. This band is assigned to the C–H
stretching mode of the C2-position of the imidazolium ring
(C(1) carbon in Fig. 1). It is reasonable to assume that this
vibration is coupled with the stretching mode of the backbone
C–H bonds (C(2) and C(3) carbons in Fig. 1) at 3134 cm−1.
Further, some out-of-plane deformation modes at 650 and
620 cm−1 also shift as a consequence of the interaction with
PO. Several imidazolium-ring vibrations are shifted, which
could all be responsible for the interaction with PO. This
makes it difficult to exactly pinpoint which part or fragment
of the cation interacts with PO. Nevertheless, the strongest
interaction was observed between the acidic imidazolium pro-
ton and the oxygen atom of PO through hydrogen bonding
(see Fig. 3).
In order to exclude the formation of N-heterocyclic carbenes
(NHC) during the reaction (resulting from deprotonation at
the 2-position of the imidazolium ring), catalysts 2, 3, 5 and 6
bearing Me and Et groups were examined as well (see Scheme 1).
It is well known that NHCs efficiently catalyse the cycloaddi-
tion of PO and CO2.55,56 If a free carbene was responsible for
the catalysis, 2, 3, 5 and 6 would be (almost) inactive, which
would result in poor conversion of PO. However, only a slight
decrease of the PO conversion was observed and therefore, the
formation of free NHCs during the reaction can be excluded.
Nevertheless, apparently the acidic imidazolium proton has a
high influence on the catalytic activity, which is manifested in
the conversions (see Table 1). This effect is most presumably
caused by hydrogen interactions between the imidazolium
proton and the oxygen atoms of the epoxide, leading to a
stronger polarization of the C–O bond and hence, to an
increased electrophilicity of the epoxide carbon, facilitating
the ring opening. Theoretical calculations support the assump-
tion of strong interactions of the C2 proton of the imidazolium
ring with the anion.57,58 In addition, weaker interactions
around the backbone protons with the anion are also found;
however, they are less pronounced. Furthermore, substituting
the C2 proton with a methyl group results in a different situa-
tion, where the anion is preferably located above and below
the ring.58 This is also observed using IR, Raman and NMR
spectroscopy, showing changes in the electron density which
are attributed to the position and strength of interionic inter-
actions.59 Additionally, interactions of the backbone with the
anion are found and the probability of interactions of the
carbon atom between the two nitrogen atoms with the anion
is feasible. Interestingly, the simulations show no stronger
ion pairing for unsubstituted ILs, whereas stronger hydrogen
bonds can be formed compared to substituted ILs. According
to the computational simulations, several possible interactions
of PO with an imidazolium cation can be considered. By using
neutron diffraction, it was shown that the interaction from
the anion to the cation is dependent on the charge density
and anion size.60,61 Nonetheless, interactions between the
hydrogen atoms at C2-, C4- and C5-positions and the anion
are also detected in the liquid state. The access to the back-
bone protons can be restricted due to steric effects arising
from long alkyl side chains.62 Further examination of the role
of a cation of an IL in nucleophilic SN2 and SNAr reactions
show no clear effect of the degree of the cation substitution
pattern on the reactivity.63,64 No exact interaction site could
be determined and the positive effect of ionic liquids on the
Fig. 3 FTIR spectra of a) neat compound 10 and b) 10 in the presence
of excess PO between 3200 and 2900 cm−1
.
1754 | Catal. Sci. Technol., 2014, 4, 1749–1758
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
This interaction, leading to the weakening of the C–H
bond, facilitates the nucleophilic attack of the bromide anion
at the ring carbon and the epoxide ring opening. Conse-
quently, increasing the acidity of the proton at the 2-position
of the ring, e.g. by introducing electron-withdrawing substitu-
ents to the N ring atoms, would further strengthen this effect.
It is noteworthy that with fluorination of the benzyl moiety
(see catalysts 9 and 10) the electron density in the imidazolium
ring is reduced, leading to a stronger imidazolium–H⋯O–epoxide
contact and hence, higher PO conversions. For this reason,
further investigations, such as optimization of the reaction
conditions, recycling studies and screening of different sub-
strates, were carried out using the ionic liquid 10 (m.p. 56 °C)
as the catalyst.
lowering the temperature from 70 °C to 50 °C, the conversion
of PO also decreased from 91% to 45% (Fig. 5).
Interestingly, the difference in the conversion of PO at a reac-
tion temperature of 60 °C is only a little lower than at 70 °C,
which may offer the opportunity to further reduce the tem-
perature. A reduction of the catalyst concentration occurs
along with a decrease of the catalytic activity, showing a nearly
linear relationship between the catalyst concentration and the
yield of propylene carbonate (Fig. 6). By lowering the catalyst
loading to 1 mol%, the catalyst is still active, but the PO conver-
sion decreases to 33%. Additionally, for a better understanding
of the activity of the catalyst at different catalyst loadings the
influence on the conversion of PO within the first 4 h of the
reaction was monitored and compared. The dependence of the
catalyst concentration on the activity (= turnover frequency,
TOF) follows the order of 10 mol% > 5 mol% > 1 mol%
(see Fig. 6 and 7).
Kinetic studies
Under the applied reaction conditions, 1 mol% catalyst
results in very poor activity for the conversion of PO. The
For a detailed insight into the reaction of PO and CO2 using
catalyst 10, further investigations were carried out. In a row of
nine identical experiments, each reaction was stopped after a
defined time period and analysed by GC analysis and the con-
version of PO was plotted versus the corresponding reaction
time (Fig. 4). In comparison to the so far most efficient halide
organocatalyst for this reaction, [HDBU]Cl (97% PC after 2 h
at 140 °C, cat. conc. 1 mol%, 10 bar CO2),25 catalyst 10 is
active at lower reaction temperature and CO2 pressure, yet higher
reaction time. To compare the efficiency of catalyst 10 with
that of [HDBU]Cl (DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene),
the reaction was carried out under the same reaction condi-
tions as described in ref. 25. A PO conversion of 91% with a
selectivity of 99% towards PC was obtained, which is only
slightly lower than that using [HDBU]Cl, thus rendering cata-
lyst 10 comparable to the state-of-the-art. Lowering the pres-
sure to 4 bar CO2 reduces the conversion of PO to 64% after a
2 h reaction time.
Fig. 5 Influence of the reaction temperature on the conversion of PO.
Reaction conditions: 10 mmol of PO, 1 mmol of catalyst 10, 4 bar CO2,
22 h.
In addition, the influence of temperature and catalyst
loading on the carbonate yield was investigated. Upon
Fig. 4 Kinetic curve of the reaction of PO and CO2 using catalyst 10.
Reaction conditions: 10 mmol of PO, 1 mmol of catalyst 10, 4 bar CO2,
70 °C.
Fig. 6 Dependence of the concentration of catalyst 10 on the
conversion of PO. Reaction conditions: 10 mmol of PO, 4 bar CO2,
70 °C, 22 h.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 1749–1758 | 1755

View Article Online
Paper
Catalysis Science & Technology
Fig. 8 Studies of the influence of catalyst recycling on the activity and
PC yield. Reaction conditions: 10 mmol of PO, 1 mmol of catalyst 10,
4 bar CO2, 70 °C, 22 h. Conversions are based on GC analysis.
PC selectivity: ≥99%.
Fig. 7 Kinetic curve of the first 4 h with 10 mol%, 5 mol% and 1 mol%
of catalyst 10. Reaction conditions: 10 mmol of PO, 4 bar CO2, 70 °C.
difference in the TOFs for catalyst loadings of 5 and 10 mol%
is somewhat smaller; however, the trend is already clear at an
early stage of the reaction (1 h). Notably, a higher reaction
temperature and CO2 pressure have a more pronounced impact
on the activity. As stated below, at 140 °C, 10 bar CO2 and only
1 mol% catalyst 10, after 2 h, a conversion of 91% is reached
(vide supra), showing that the decisive parameter is not the
catalyst concentration (cf. 5% conversion with 10 mol% 10 at
70 °C and 4 bar CO2). On the other hand, it can be argued
that the more favourable scenario would be shorter reaction
times but harsher reaction conditions (taking into account
the carbon footprint) or milder conditions associated with
longer reaction times.
Table 3 Catalyzed cycloaddition of different epoxides using catalyst
10 under optimized reaction conditionsa
Entry
1
Substrate
Conv. [%]
91b
2
3
4
5
99c
98c
72c
96c
6
78c
a
10 mmol of epoxide, 1 mmol of catalyst 10, 4 bar CO2, 70 °C, 22 h.
Recycling studies
b
Conversion is based on GC analysis; selectivity ≥99% for the
carbonate product. Conversions are based on 1H NMR using
mesitylene as the internal standard. Carbonate selectivity: ≥99%.
c
Stability and recyclability of a catalyst are of great importance
for an application beyond the laboratory scale. In order to
examine the reusability of catalyst 10 after separation of the
product and to exclude possible leaching effects, the catalytic
cycloaddition of PO and CO2 was carried out under the condi-
tions described above. After separation of PC from the cata-
lyst by extraction with ethyl acetate/diethyl ether (for details
see the Experimental section) the catalyst was dried and used
for the next run under the same conditions. The results,
shown in Fig. 8, indicate that the conversion of PO remains
stable after ten consecutive cycles. The slight differences in
the conversion lie in the range of the standard deviation of
the gas chromatograph. From this, it can be stated that cata-
lyst 10 may indeed be a suitable candidate for long-term
catalysis experiments. Furthermore, leaching of the catalyst
was not observed.
encumbering substituents. For each experiment, the carbon-
ate selectivity is ≥99%. The good conversions of epichlorohy-
drin and glycidol (entries 2 and 3; 99 and 98%, respectively)
can be explained by their good electron-withdrawing capabil-
ity, allowing for an easier nucleophilic attack of the epoxide
ring carbon atoms. Even for cyclohexene oxide (entry 6),
which is known to be notoriously more difficult to undergo
cycloaddition than epoxides of open-chained terminal olefins,
the corresponding carbonate is obtained in good yield.22
Conclusion
The new imidazolium-based ionic liquid 10 catalyses the
transformation of CO2 to cyclic carbonates under mild condi-
tions. It is the first imidazolium-based organocatalyst which
shows nearly quantitative conversion of propylene oxide to
propylene carbonate below 80 °C and 5 bar CO2. The absence
of metals and additional solvents as well as the facile recycling
and reusability without loss of activity render this catalyst
Conversion of CO2 with other substrates
In order to show the potential of the new imidazolium cata-
lyst 10, the scope of epoxide substrates for the cycloaddition
with CO2 is extended (Table 3). Compound 10 is an efficient
catalyst for the conversion of a variety of substrates with
either electron withdrawing, electron donating or sterically
1756 | Catal. Sci. Technol., 2014, 4, 1749–1758
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
particularly ‘green’. The catalytic system can be reused at
least ten times without any loss of catalytic activity. These
results render compound 10 as a viable catalyst for CO2 fixa-
tion with respect to a low carbon footprint. With the right
choice of the substitution pattern at the imidazolium ring,
the solubility towards CO2 is enhanced. The effect of ‘wingtip’
substituents at the imidazolium ring on the acidity (and hence
on the epoxide ring activation) on one hand and the ion
pairing effect on the halide nucleophilicity on the other hand
has been investigated. Further research efforts should focus
on experimental and theoretical studies of the interplay
between the hydrogen bonding, the acidity of the imidazolium
salts, and the electronic and steric effects of the substituents,
as well as on studies of the influence of CO2 solubility on
the catalytic activity. These studies aim at the development
of ionic liquids with tailor-made cations, which should
further facilitate the cycloaddition, allowing for a decrease
of the reaction temperature below 70 °C (down to room tem-
perature) and thus helping to minimize the carbon footprint
for this reaction.
14 D. J. Darensbourg, Chem. Rev., 2007, 107, 2388–2410.
15 M. R. Kember, A. Buchard and C. K. Williams, Chem. Commun.,
2011, 47, 141–163.
16 X.-B. Lu and D. J. Darensbourg, Chem. Soc. Rev., 2012, 41,
1462–1484.
17 X.-B. Lu, W.-M. Ren and G.-P. Wu, Acc. Chem. Res., 2012, 45,
1721–1735.
18 C. Chatterjee and M. H. Chisholm, Chem. Rec., 2013, 13,
549–560.
19 Z.-Z. Yang, Y.-N. Zhao and L.-N. He, RSC Adv., 2011, 1,
545–567.
20 Y. J. Kim and R. S. Varma, J. Org. Chem., 2005, 70,
7882–7891.
21 J. Sun, S.-I. Fujita and M. Arai, J. Organomet. Chem., 2005,
690, 3490–3497.
22 M. H. Anthofer, M. E. Wilhelm, M. Cokoja and F. E. Kühn,
in Transformation and Utilization of Carbon Dioxide,
ed. B. M. Bhanage and M. Arai, Springer, in print.
23 Q. He, J. O'Brien, K. Kitselman, L. Tompkins, G. Curtis and
F. M. Kerton, Catal. Sci. Technol., 2014, DOI: 10.1039/C3CY00998J.
24 J. Roeser, K. Kailasam and A. Thomas, ChemSusChem, 2012,
5, 1793–1799.
25 Z.-Z. Yang, L.-N. He, C.-X. Miao and S. Chanfreau, Adv.
Synth. Catal., 2010, 352, 2233–2240.
26 Y. Xiong, H. Wang, R. Wang, Y. Yan, B. Zheng and Y. Wang,
Chem. Commun., 2010, 46, 3399–3401.
Acknowledgements
MHA, MEW and IIEM thank the TUM Graduate School for
financial support.
27 L. Han, H.-J. Choi, D.-K. Kim, S.-W. Park, B. Liu and
D.-W. Park, J. Mol. Catal. A: Chem., 2011, 338, 58–64.
28 L. Han, H. Li, S.-J. Choi, M.-S. Park, S.-M. Lee, Y.-J. Kim and
D.-W. Park, Appl. Catal., A, 2012, 429–430, 67–72.
29 K. Motokura, S. Itagaki, Y. Iwasawa, A. Miyaji and T. Baba,
Green Chem., 2009, 11, 1876–1880.
30 T. Sakai, Y. Tsutsumi and T. Ema, Green Chem., 2008, 10,
337–341.
31 Y. Xie, Z. Zhang, T. Jiang, J. He, B. Han, T. Wu and K. Ding,
Angew. Chem., Int. Ed., 2007, 46, 7255–7258.
32 J. Sun, S. Zhang, W. Cheng and J. Ren, Tetrahedron Lett.,
2008, 49, 3588–3591.
Notes and references
1 M. Peters, B. Köhler, W. Kuckshinrichs, W. Leitner, P. Markewitz
and T. E. Müller, ChemSusChem, 2011, 4, 1216–1240.
2 N. MacDowell, N. Florin, A. Buchard, J. Hallett, A. Galindo,
G. Jackson, C. S. Adjiman, C. K. Williams, N. Shah and P. Fennell,
Energy Environ. Sci., 2010, 3, 1645–1669.
3 M. Mikkelsen, M. Jorgensen and F. C. Krebs, Energy Environ.
Sci., 2010, 3, 43–81.
4 P. Markewitz, W. Kuckshinrichs, W. Leitner, J. Linssen,
P. Zapp, R. Bongartz, A. Schreiber and T. E. Muller, Energy
Environ. Sci., 2012, 5, 7281–7305.
5 E. A. Quadrelli, G. Centi, J.-L. Duplan and S. Perathoner,
ChemSusChem, 2011, 4, 1194–1215.
33 Y.-M. Shen, W.-L. Duan and M. Shi, Eur. J. Org. Chem., 2004,
2004, 3080–3089.
6 M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann and
F. E. Kühn, Angew. Chem., Int. Ed., 2011, 50, 8510–8537.
7 M. Drees, M. Cokoja and F. E. Kühn, ChemCatChem, 2012, 4,
1703–1712.
34 Y.-M. Shen, W.-L. Duan and M. Shi, Adv. Synth. Catal., 2003,
345, 337–340.
35 C. J. Whiteoak, A. Nova, F. Maseras and A. W. Kleij,
ChemSusChem, 2012, 5, 2032–2038.
8 T. Sakakura and K. Kohno, Chem. Commun., 2009, 1312–1330.
9 W.-L. Dai, S.-L. Luo, S.-F. Yin and C.-T. Au, Appl. Catal., A,
2009, 366, 2–12.
10 S. Fukuoka, M. Kawamura, K. Komiya, M. Tojo, H. Hachiya,
K. Hasegawa, M. Aminaka, H. Okamoto, I. Fukawa and
S. Konno, Green Chem., 2003, 5, 497–507.
36 J. Sun, L. Han, W. Cheng, J. Wang, X. Zhang and S. Zhang,
ChemSusChem, 2011, 4, 502–507.
37 J.-Q. Wang, K. Dong, W.-G. Cheng, J. Sun and S.-J. Zhang,
Catal. Sci. Technol., 2012, 2, 1480–1484.
38 I. I. E. Markovits, W. A. Eger, S. Yue, M. Cokoja,
C. J. Münchmeyer, B. Zhang, M.-D. Zhou, A. Genest, J. Mink,
S.-L. Zang, N. Rösch and F. E. Kühn, Chem. – Eur. J., 2013,
19, 5972–5979.
11 A. Decortes, A. M. Castilla and A. W. Kleij, Angew. Chem., Int. Ed.,
2010, 49, 9822–9837.
12 S. Klaus, M. W. Lehenmeier, C. E. Anderson and B. Rieger,
Coord. Chem. Rev., 2011, 255, 1460–1479.
13 M. North, R. Pasquale and C. Young, Green Chem., 2010, 12,
1514–1539.
39 M. Lee, U. H. Choi, S. Wi, C. Slebodnick, R. H. Colby and
H. W. Gibson, J. Mater. Chem., 2011, 21, 12280–12287.
40 M. Debdab, F. Mongin and J. P. Bazureau, Synthesis, 2006,
4046–4052.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 1749–1758 | 1757

View Article Online
Paper
Catalysis Science & Technology
41 P. Lucas, N. E. Mehdi, H. A. Ho, D. Bélanger and L. Breau,
Synthesis, 2000, 1253–1258.
42 C. C. Weber, A. F. Masters and T. Maschmeyer, J. Phys.
Chem. B, 2012, 116, 1858–1864.
43 I. Arellano, J. Guarino, F. Paredes and S. Arco, J. Therm.
Anal. Calorim., 2011, 103, 725–730.
44 C. Betti, D. Landini and A. Maia, Tetrahedron, 2008, 64,
1689–1695.
45 F. D'Anna, S. Marullo, P. Vitale and R. Noto, ChemPhysChem,
2012, 13, 1877–1884.
46 S. N. V. K. Aki, B. R. Mellein, E. M. Saurer and J. F. Brennecke,
J. Phys. Chem. B, 2004, 108, 20355–20365.
47 J. D. Holbrey, W. M. Reichert, M. Nieuwenhuyzen, S. Johnson,
K. R. Seddon and R. D. Rogers, Chem. Commun., 2003,
1636–1637.
48 J. M. Serrano-Becerra, S. Hernandez-Ortega, D. Morales-Morales
and J. Valdes-Martinez, CrystEngComm, 2009, 11, 226–228.
49 V. Caló, A. Nacci, A. Monopoli and A. Fanizzi, Org. Lett.,
2002, 4, 2561–2563.
53 M. C. Corvo, J. Sardinha, S. C. Menezes, S. Einloft, M. Seferin,
J. Dupont, T. Casimiro and E. J. Cabrita, Angew. Chem., Int. Ed.,
2013, 52, 13024–13027.
54 O. Hollóczki, Z. Kelemen, L. Könczöl, D. Szieberth, L. Nyulászi,
A. Stark and B. Kirchner, ChemPhysChem, 2013, 14, 315–320.
55 H. Zhou, W.-Z. Zhang, C.-H. Liu, J.-P. Qu and X.-B. Lu,
J. Org. Chem., 2008, 73, 8039–8044.
56 Y. Kayaki, M. Yamamoto and T. Ikariya, Angew. Chem., Int. Ed.,
2009, 48, 4194–4197.
57 K. Fumino, A. Wulf and R. Ludwig, Angew. Chem., Int. Ed.,
2008, 47, 8731–8734.
58 Y. Zhang and E. J. Maginn, Phys. Chem. Chem. Phys., 2012,
14, 12157–12164.
59 K. Noack, P. S. Schulz, N. Paape, J. Kiefer, P. Wasserscheid
and A. Leipertz, Phys. Chem. Chem. Phys., 2010, 12, 14153–14161.
60 C. Hardacre, J. D. Holbrey, S. E. J. McMath, D. T. Bowron
and A. K. Soper, J. Chem. Phys., 2003, 118, 273–278.
61 M. Deetlefs, C. Hardacre, M. Nieuwenhuyzen, A. A. H. Padua,
O. Sheppard and A. K. Soper, J. Phys. Chem. B, 2006, 110,
12055–12061.
62 C. Hardacre, J. D. Holbrey, C. L. Mullan, T. G. A. Youngs and
D. T. Bowron, J. Chem. Phys., 2010, 133, 074510.
63 E. E. L. Tanner, H. M. Yau, R. R. Hawker, A. K. Croft and
J. B. Harper, Org. Biomol. Chem., 2013, 11, 6170–6175.
64 E. E. L. Tanner, R. R. Hawker, H. M. Yau, A. K. Croft and
J. B. Harper, Org. Biomol. Chem., 2013, 11, 7516–7521.
50 M. J. Muldoon, S. N. V. K. Aki, J. L. Anderson, J. K. Dixon
and J. F. Brennecke, J. Phys. Chem. B, 2007, 111, 9001–9009.
51 C. Cadena, J. L. Anthony, J. K. Shah, T. I. Morrow,
J. F. Brennecke and E. J. Maginn, J. Am. Chem. Soc., 2004,
126, 5300–5308.
52 E. Torralba-Calleja, J. Skinner and D. Gutirrez-Tauste, J. Chem.,
2013, 2013, 1–16.
1758 | Catal. Sci. Technol., 2014, 4, 1749–1758
This journal is © The Royal Society of Chemistry 2014