H2O2/NaHCO3-mediated enantioselective epoxidation of olefins in NTf2-based ionic liquids and under ultrasound

Article abstract of DOI:10.1016/j.jcat.2012.04.016

The bicarbonate-activated peroxide (BAP) system has been shown to constitute an excellent and versatile combination for the epoxidation of olefins. Herein, we demonstrate that the benign H₂O ₂/NaHCO₃ oxidative combination can be used for the enantioselective epoxidation of olefins using a chiral manganese porphyrin as a catalyst. This unprecedented result was made possible by the use of an ionic liquid with ultrasound, which leads to the activation of Mn-porphyrin catalysis.

Full text of DOI:10.1016/j.jcat.2012.04.016

Journal of Catalysis 291 (2012) 127–132
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
H2O2/NaHCO3-mediated enantioselective epoxidation of olefins in NTf2-based
ionic liquids and under ultrasound
Gregory Chatel a,b, Catherine Goux-Henry b, Anaïs Mirabaud a, Thomas Rossi a, Nathalie Kardos a,
Bruno Andrioletti b, , Micheline Draye a,
⇑
⇑
a Laboratoire de Chimie Moléculaire et Environnement, Université de Savoie, CISM, Campus scientifique, Le Bourget du Lac Cedex 73376, France
b Institut de Chimie et Biochimie Moléculaire et Supramoléculaire (Equipe CASYEN), UMR-CNRS 5246, Université Claude Bernard Lyon 1, Bâtiment Curien (CPE), 43,
boulevard du 11 Novembre 1918, Villeurbanne Cedex 69622, France
a r t i c l e i n f o
a b s t r a c t
Article history:
The bicarbonate-activated peroxide (BAP) system has been shown to constitute an excellent and versatile
combination for the epoxidation of olefins. Herein, we demonstrate that the benign H2O2/NaHCO3 oxida-
tive combination can be used for the enantioselective epoxidation of olefins using a chiral manganese
porphyrin as a catalyst. This unprecedented result was made possible by the use of an ionic liquid with
ultrasound, which leads to the activation of Mn-porphyrin catalysis.
Received 22 February 2012
Revised 18 April 2012
Accepted 19 April 2012
Available online 23 May 2012
Ó 2012 Elsevier Inc. All rights reserved.
Keywords:
Epoxidation
Ionic liquids
Ultrasound
Enantioselective catalysis
Mn-porphyrin
1. Introduction
er reaction time may be attributed to the action of ultrasound that
favors the fast degradation of the transient oxidant active peroxy-
The epoxidation of olefins catalyzed by iron or manganese
metalloporphyrins [1] has received much attention in the rapid
development of their area of research, particularly that related to
enantioselective catalysis [2]. The combination of H2O2/NaHCO3/
imidazole/Mn(TPP)OAc (TPP: tetraphenyl porphyrin) has been pro-
posed in the literature for the selective oxidation of olefins [3].
Epoxidation of lipophilic alkenes using hydrogen peroxide with
manganese sulfate in an imidazolium-based ionic liquid has been
also described [4].
Recently, we proposed that the H2O2/NaHCO3/imidazole/
Mn(TPP)OAc catalytic system proceeds via the Mn(TPP)OAc por-
phyrin rather than the classical bicarbonate-activated peroxide
(BAP) [5] when the epoxidation reaction is carried out in an ionic
liquid under ultrasonic activation [6]. In addition, this smart com-
bination (1) prevents the metalloporphyrin from fast bleaching by
immobilization in the ionic liquid phase and (2) improves the mass
transfer of the biphasic system (hydrophobic ionic liquid and H2O2
(30%) aqueous phase), hence increasing the kinetics of the reaction
(1 h versus 5 h in classical conditions). We proposed that the short-
monocarbonate intermediate allowing a fast epoxidation via the
porphyrin complex (Fig. 1).
In the present article, we clearly demonstrate our prior hypoth-
esis [6]. Indeed, when using a chiral metalloporphyrin, the chiral
epoxide is detected in up to 40% enantiomeric excess. This observa-
tion clearly indicates that the mechanism involves the porphyrin
catalyst. In addition, we demonstrate for the first time that enan-
tioselective oxidation can be carried out in ionic liquids under
ultrasound activation, using NaHCO3/H2O2 as terminal oxidant.
In the recent years, the use of ionic liquids as reaction medium
has attracted much attention. Indeed, these ionic solvents display
interesting properties that could be used for the advantageous
replacement of volatile organic solvents (low vapor pressure, air
and moisture stabilities, high polarity, etc.). In addition, ionic liq-
uids have been shown to allow an easier separation of catalyst
and an easier recycling, by immobilizing the catalyst without the
need for any post-functionalization [7]. Recently, Jessop reviewed
the uses of ionic liquids in research: almost half of the papers deal-
ing with the various class of solvent published in the ‘‘Green Chem-
istry’’ journal in 2010 concern ionic liquids [8]. However, ionic
liquids can not be considered as ‘‘green solvents’’ in terms of sus-
pected toxicity [9] or in terms of number of steps and importance
of energy consumption [10] during their synthesis. In this context,
⇑
Corresponding authors. Fax: +33 (0)472 448 160 (B. Andrioletti), fax: +33
(0)479 758 674 (M. Draye).
E-mail addresses: bruno.andrioletti@univ-lyon1.fr (B. Andrioletti), micheline.
draye@univ-savoie.fr (M. Draye).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2012.04.016

128
G. Chatel et al. / Journal of Catalysis 291 (2012) 127–132
Reaction conditions: 1-methylpyrrolidine (20.0 mmol, 1.2 equiv) and 1-octylbromide
(16.7 mmol, 1 equiv) at the reflux of ethyl acetate (15 mL), under argon. Isolated
yield obtained after washing with ethyl acetate and drying under reduced pressure.
Fig. 3. Kinetic monitoring of the isolated yield of MOPyrroNTf2 ionic liquid versus
time.
Fig. 1. Hypothetical mechanisms for olefins epoxidation reaction in acetonitrile/
silent conditions and ionic liquid/ultrasound conditions.
solvent according to the EHS indicator proposed by Capello et al.
[10]. The EHS indicator takes environmental, safety, and health
impacts into account to classify the solvents. Moreover, the
nucleophilic substitution reaction, generally carried out in toxic
solvents such as acetonitrile or dichloromethane [17], needs long
reaction times to proceed (up to several days) [18]. Thus, in order
to minimize the environmental and economical impacts of the
synthesis, we also determined the optimal stoichiometry and
reaction time for obtaining the highest conversion in refluxing
ethyl acetate (Fig. 3). We established that introducing the amine
in slight excess compared with the halide derivative generally
affords higher conversion. In addition, the excess of amine
could easily be removed by evaporation in vacuo after reaction
because of its low boiling point. Amounts of solvent were also
optimized, and ethyl acetate used for washing was recycled after
distillation.
we re-investigated and optimized the synthesis of the most com-
mon ionic liquids we use in our reactions.
2. Results and discussion
2.1. Improved ionic liquids synthesis
In a recent publication, Deetlefs and Seddon discussed the
greenness of ionic liquids synthesis [11], with a reflexion based
on the green chemistry principles [12] and on the use of green met-
rics [13]. It is now well established that the syntheses of ionic liq-
uids are not green. However, methods exist for reducing the
environmental impact of synthesis. For example, the use of micro-
wave irradiation [14], ultrasonic activation [15], or the combina-
tion of both [16] may be used for the preparation of ionic liquids.
In addition, Deetlefs and Seddon reported that the purification of
hydrophobic ionic liquids is intrinsically greener than hydrophilic
ionic liquids, because the removal of the halide salt after anion
metathesis is achieved simply by washing the crude ionic liquid
product with water [11]. Moreover, although a stoichiometric
amount of halide salt waste is generated during two-step ionic li-
quid syntheses, it is not a toxic waste.
The following metathesis step was carried out in distilled water
over a few hours, at room temperature, and the final purification
step was carried out using the common activated carbon or decol-
orizing charcoal method [19].
2.2. Greener assessment
Based on the aforementioned principles, we re-investigated the
synthesis of lipophilic piperidinium-, pyrrolidinium-, imidazolium-
and pyridinium-bis(trifluoromethanesulfonyl)imide ionic liquids
(Fig. 2).
To this end, we carried out the preliminary nitrogen atom
quaternization in ethyl acetate that is ranked as one of best
In addition to applying the 12 principles of green chemistry for
the evaluation of the green credentials of ionic liquid syntheses
and purifications, these methods can also be assessed using addi-
tional measures, such as E-factor [20] or Eco-Scale calculations
[21].
The E-factor concept was introduced, where all by-products are
classified as waste. The nature of the E-factor equation, Eq. (1), im-
plies that the greenest chemical reactions have E-factor values
nearing zero.
Amount of waste produced in the process=kg
E-factor ¼
Amount of the desired productðsÞproduced in the process=kg
ð1Þ
When ionic liquids are prepared using classical conditions, E-factor
ranges between 8 and 10 for the first step and between 6 and 8 for
the metathesis step. In our optimized conditions, the E-factor was
reduced to 4–5 for the first step (and 2–3 when ethyl acetate was
recycled) and 5–6 for the metathesis.
Fig. 2. Cations and anions selected for this study.

G. Chatel et al. / Journal of Catalysis 291 (2012) 127–132
129
EcoScale is a semi-quantitative tool used to select an organic
preparation based on economical and ecological parameters and
calculated by lowering the maximum value of 100 by any applica-
ble penalty points (Eq. (2)). An ideal reaction has the EcoScale value
of 100.
of the bis(trifluoromethylsulfonyl)imide-based ionic liquids in
oxidation compared with the tetrafluoroborate or hexafluorophos-
phate congeners was often argued [25], in particular, because the
former cannot produce organic volatiles — including the highly
corrosive and toxic HFꢀ that has been detected with the latter
anions [26].
We have also chosen NTfꢀ2 -based ionic liquids, for their ability to
easily solubilize the Mn(TPP)OAc catalyst using an ultrasonic bath
(20 kHz) that enhances the dissolution. Our screening of
NTfꢀ2 -based ionic liquid revealed the following ability for
dissolving the Mn-porphyrin catalyst: MOImNTf2 < OPyNTf2 <
EOPipNTf2 < MBPyrroNTf2 < MOPyrroNTf2. Conversely, Mn(TPP)OAc
was not soluble in any hexafluorophosphate-based ionic liquids.
For our catalytic studies [6], cyclooctene was chosen as model
substrate for the epoxidation reaction conducted in the different
ionic liquids using Mn(TPP)OAc as catalyst (Table 1). Catalytic tests
carried out in acetonitrile resulted in a fast catalyst bleaching and a
low conversion of the starting alkene. The combination of ionic
liquid/ultrasound afforded the selective formation of cyclooctene
oxide in modest 5%, 6%, 21%, and 35% yield, respectively, in MBPyr-
roNTf2, EOPipNTf2, OPyNTf2, and MOImNTf2 ionic liquids. In these
cases, bleaching of the catalyst, observed after 10–30 min of
ultrasonic irradiation, can explain the low yields. Noteworthy,
these results account for a porphyrin-mediated mechanism. The
best result (72% of cyclooctene oxide yield) was obtained in the
more hydrophobic MOPyrroNTf2. In this ionic liquid, no bleaching
occurred even after 1 h of ultrasonic irradiation as proven by
UV–visible monitoring of the reaction mixture. This observation
indicates that the MOPyrroNTf2 ionic liquid prevents the oxidative
degradation of the porphyrin, possibly because the high hydropho-
bicity of MOPyrroNTf2 limits the cavitation phenomenon in the
ionic liquid phase, hence preventing the porphyrin catalyst from
bleaching.
EcoScale ¼ 100 ꢀ sum of individual penalties
ð2Þ
EcoScale is unchanged with our optimized preparation since this
indicator takes into account experimental steps of synthesis
(extraction, temperature conditions, etc.) and not reactants
amounts optimization. In our case, the EcoScale number is about
85 for the first quaternarization step and 85 for the metathesis.
These values are satisfactory compared to EcoScale numbers ob-
tained in the literature [21].
2.3. Evaluation of different NTfꢀ2 -based ionic liquids in olefin
epoxidation
The hydrophilicity of the aqueous bicarbonate-activated perox-
ide system and the lipophilicity of alkenes urged us to develop a
favorable ionic liquid system that could provide a suitable reaction
environment for both hydrophilic and hydrophobic molecules. For
these reasons, we decided to carry out the catalytic epoxidation in
ionic liquids based on NTfꢀ2 anion (Fig. 2). Indeed, these ionic
liquids display generally good air, moisture, and electrochemical
stabilities [22]. Moreover, these ionic liquids are less viscous
(g
= 60–330 mm2 sꢀ1 at 20 °C) than some others [23], facilitating
the mass transfer. In addition, Wong et al. demonstrated that pyr-
rolidinium-based ionic liquids with NTfꢀ2 anion provide a stable
environment in oxidation conditions [24]. This enhanced stability
Table 1
Cyclooctene epoxidation in various ionic liquids catalyzed by MnTPPOAc.
2.4. Recycling
In the recent years, ionic liquids were widely used in catalysis
for the interesting properties discussed above and because the cat-
alysts can be entrapped or immobilized in the ionic liquid phase
without post-functionalization. Generally, the affinity of the polar
catalysts for the ionic liquid phase allows liquid–liquid extraction
and/or distillation of the organic product and reuse of the ionic
liquid/catalyst system. Recycling of the ionic liquid/catalyst system
using this strategy is well documented and possesses a serious
advantage in catalysis, especially for expensive or difficult-to-
synthesize catalysts [27]. However, recycling of the ionic liquid/
catalyst is not always straightforward [28]. In our case, after a sim-
ple liquid–liquid extraction, the yield of epoxidation using recycled
MOPyrroNTf2/Mn(TPP)OAc system dropped to 12% after the second
run (Table 2, entry 2). Yet, the UV–visible spectrum of the
Mn(TPP)OAc catalyst confirms that the catalyst was not degraded
after the first run. Moreover, reuse of the recycled Mn(TPP)OAc
Entry
Solvent
Epoxide yielda,b (%)
Selectivity (%)c
1
2
3
4
5
6
CH3CN
8d
5
>99
>99
>99
>99
>99
>99
MBPyrroNTf2
EOPipNTf2
OPyNTf2
MOImNTf2
MOPyrroNTf2
6d
21d
35d
72
a
GC-yield (based on starting cyclooctene).
Optimized reaction conditions:
4 ꢁ 0.25 equiv NaHCO3, 1.5 mol% imidazole, 0.15 mol% Mn(TPP)OAc, 3 mL solvent,
ultrasonic irradiation: f = 20 kHz, Pelec = 11.5 W, 1 h.
Selectivity is defined as the percentage of desired epoxide relative to all prod-
ucts obtained.
b
1
equiv cyclooctene, 4 ꢁ 2.5 equiv H2O2,
c
d
Bleaching of catalyst.
Table 2
Recycling attempts of the MOPyrroNTf2/Mn(TPP)OAc combination.
Entry
Cyclooctene epoxidation
Recycling method
Epoxide yielda (%)
72
12
66
12
35
1
2
3
4
5
1st runb
2nd runb
2nd runb
2nd runb
2nd runb
Extraction of MOPyrroNTf2/Mn(TPP)OAc system recycled
Extraction of Mn(TPP)OAc and reaction in a new batch of MOPyrroNTf2
Extraction of MOPyrroNTf2 and reaction with a new batch of Mn(TPP)OAc
Extraction and washing,c of MOPyrroNTf2 and reaction with a new batch of Mn(TPP)OAc
a
GC-yield (based on starting cyclooctene).
b
Optimized reaction conditions: 1 equiv cyclooctene, 4 ꢁ 2.5 equiv H2O2, 4 ꢁ 0.25 equiv NaHCO3, 1.5 mol% imidazole, 0.15 mol% Mn(TPP)OAc, 3 mL solvent, ultrasonic
irradiation: f = 20 kHz, Pelec = 11.5 W, 1 h.
c
MOPyrroNTf2 was washed with water, stirred with activated charcoal for 2 h, filtered, and dried under vacuum for 2 h at 50 °C.

130
G. Chatel et al. / Journal of Catalysis 291 (2012) 127–132
Fig. 4. Strategies for preparation of aabb bis-binaphthyl-strapped Mn porphyrin.
for a second run in a new batch of ionic liquid afforded the epoxide
in 66% yield, a result similar to that obtained for the first run
(Table 2, entry 3).
to the best of our knowledge, the preparation of the manganese
analogue had not been described yet. Accordingly, we attempted
the metallation of the chiral free-base 3 with different manganese
salts such as Mn(OAc)2 and MnBr2, in the presence of various
bases (2,6-lutidine and sodium acetate) and different solvents
(glacial acetic acid and tetrahydrofuran) at reflux or under ultra-
sonic irradiation. Unfortunately, no complexation occurred possi-
bly because of the steric hindrance imposed by the two
binaphthyl handles. Then, we tried the second strategy via the
preliminary metallation of the free base porphyrin 1 followed
by the grafting of the chiral straps under high dilution conditions.
Interestingly, along with the expected porphyrin complex 6, we
also isolated porphyrin 5 that displays only three fastening and
a remaining NH2 function.
After proving that the catalyst had not been degraded, we inves-
tigated the recycling of the ionic liquid by adding a new batch of
the catalyst for a second run of epoxidation in a recycled ionic
liquid (Table 2, entry 4). Unfortunately, using these conditions,
the epoxide was isolated in 12% yield, confirming the results
obtained for the entirely recycled system (Table 2, entry 2).
Accordingly, we assumed that the ionic liquid could contain some
impurity that would poison the catalyst in the second run, leading
to modest yields. After checking by NMR, MS, and UV–visible anal-
yses that the recycled ionic liquid did not contain any detectable
impurities, we decided to treat the ionic liquid with activated
charcoal. Interestingly, this treatment allowed us to isolate the
expected epoxide in 35% yield (Table 2, entry 5) with a new batch
of porphyrin. While promising, this result is still lower than that
obtained with a fresh ionic liquid. Yet, we demonstrated that
recycling of the catalyst/ionic liquid system is feasible. However,
further investigations remain to be carried out for determining
the nature of the impurity that limits the recycling. Interestingly,
Suslick et al. also noted a slight degradation of imidazolium-based
ionic liquids after sonication [29]. More complete studies dealing
with the degradation of NTfꢀ2 -based ionic liquids under ultrasonic
irradiation are currently in progress in our laboratories to under-
stand and solve this recycling issue.
2.6. Catalytic activity: confirmation of the crucial role of the
MOPyrroNTf2/ultrasonic activation combination
The catalytic activity of porphyrins 5 and 6 were evaluated
first using the silent PhIO methodology (Table 3, entries 1 and
2). In dichloromethane, in the presence of PhIO, 5 and 6 effi-
ciently epoxidized styrene into the corresponding epoxide in 82
and 95% yield, respectively. More interestingly, enantiomeric ex-
cesses (ee’s) of 32% and 64% were measured with the reaction
run at 2 °C. Even if those results are not as good as those obtained
with Collman’s iron catalyst [30], they clearly demonstrate that
(1) the chiral manganese complexes induce a satisfying enanti-
oselectivity, (2) despite an open face, catalyst 5 induces a non-
neglectable enantioselectivity for the epoxidation of the challeng-
ing styrene substrate. Secondly, we investigated catalytic perfor-
mance of our system in acetonitrile and in silent conditions
(Table 3, entries 3–6). We then screened both catalysts 5 and 6
in the ‘‘optimized’’ conditions in MOPyrroNTf2 and under ultra-
sonic activation at different temperatures and various reaction
times (Table 3, entries 7–15). If the reaction is carried out over
1 h, the catalyst readily degrades, and the enantioselectivity
remains close to 0. When limiting the reaction time and lowering
the temperature, the enantioselectivity increases up to 26% ee,
using catalyst 6 (Table 3, entry 15). If the reaction time is reduced
to 5 min (Table 3, entries 16 and 17) and the temperature main-
tained at 7 °C, the ee reaches 40% for catalyst 6. These results
clearly confirm that in MOPyrroNTf2, epoxidation proceeds via
2.5. Synthesis of the chiral Mn-porphyrin
Based on Collman’s work dealing with asymmetric epoxidation
by aabb bis-binaphthyl-strapped Fe porphyrin [30], we studied
two chiral analogues, namely the aabb bis-binaphthyl-strapped
Mn porphyrin
6 and a bis-binaphthylporphyrin displaying a
mobile, mono-grafted binol residue 5 (Fig. 4). After isolation of
the aabb-TAPP (TAPP: Tetrakis(2-aminophenyl porphyrin)) free
base porphyrin 1 (Fig. 4), two synthetic approaches were possi-
ble: (1) strapping 2 equiv of the MeO-substituted binaphthyl
diacid chloride 2, followed by Mn insertion, or (2) Mn metallation
followed by the strapping of MeO-substituted binaphthyl diacid
chloride.
Collman et al. described that the first route was not possible
for Fe metallation because of the size of the metal ion. However,

G. Chatel et al. / Journal of Catalysis 291 (2012) 127–132
131
Table 3
Enantiomeric epoxidation of styrene.
Entry
Catalyst
Activationa
Solvent
Epoxidation conditions
Time (min)
Temperatureb (°C)
Epoxide yieldc (%)
eed (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Cat. 5
Cat. 6
Cat. 5
Cat. 6
Cat. 5
Cat. 6
Cat. 5
Cat. 5
Cat. 5
Cat. 5
Cat. 5
Cat. 6
Cat. 6
Cat. 6
Cat. 6
Cat. 5
Cat. 6
Cat. 5
Cat. 6
Cat. 6
None
None
None
None
None
None
))))
))))
))))
))))
))))
))))
))))
))))
))))
))))
))))
))))
))))
CH2Cl2
CH2Cl2
CH3CN
CH3CN
CH3CN
CH3CN
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
MOPyrroNTf2
PhIOe
60
60
60
60
10
10
60
60
10
10
10
60
10
10
10
5
2
2
25
25
2
2
25
7
25
14
7
25
25
14
7
82
95
12
93
71
80
18
36
3
32
64
1
1
5
3
1
2
1
PhIOe
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
H2O2/NaHCO3/imidazolef
8
5
7
12
1
72
68
57
51
3
16
13
18
5
1
20
26
10
40
18
34
16
7
7
7
7
5
g
H2O2
10
10
5
g
H2O2
g
))))
H2O2
7
a
b
c
d
e
f
Ultrasonic irradiation (20 kHz): 3 mm diameter tapered microtip probe, Pelec = 11.5 W, Pacous.vol (in water) = 0.79 W mLꢀ1
Macroscopic temperature measured in reaction medium and maintained using a minichiller cooler.
GC-yield (based on starting cyclooctene).
.
Determined on chiral GC. In all cases, (S)-styrene oxide is the major enantiomer obtained.
1 equiv cyclooctene, 2 ꢁ 9.5 equiv PhIO, 0.15 mol% catalyst, 2.5 mL solvent.
1 equiv cyclooctene, 2.5 equiv H2O2, 0.12 equiv NaHCO3, 1.5 mol% imidazole, 0.15 mol% catalyst, 2.5 mL solvent.
1 equiv cyclooctene, 2.5 equiv H2O2, 0.15 mol% catalyst, 2.5 mL solvent.
g
the porphyrin (good ee’s are obtained) even if they do not match
those obtained with PhIO. This observation may be ascribed to
the difficult control of the temperature under ultrasonic activa-
tion. Indeed, if a macroscopic temperature control can be realized,
one cannot exclude hot spots that are known to be deleterious for
enantioselective reactions. Interestingly, under our optimized
conditions, even in the absence of base (NaHCO3 and imidazole),
H2O2 appears oxidizing enough for converting styrene into the
corresponding epoxide in 18% yield and 34% ee, respectively using
catalyst 6 (Table 3, entry 19).
sound can alter the ionic liquid, hence preventing an easy and
efficient recycling step.
4. Experimental section
4.1. General synthesis for ionic liquids synthesis
A solution of the appropriate amine (275 mmol, 1.2 equiv) in
ethyl acetate (120 mL) was degassed for 15 min by argon bubbling.
Then, 1-alkylbromide (229 mmol, 1 equiv) was added and the
resulting mixture was refluxed for 15 h under argon. Finally, the
solvent was evaporated under reduced pressure (rotary evapora-
tor). The residue was washed thoroughly with ethyl acetate
(3 ꢁ 100 mL) and dried under reduced pressure to afford the
first-generation ionic liquid as an hygroscopic white solid. The
resulting product (210 mmol, 1 equiv) was added to a solution of
LiNTf2 (166 mmol, 0.8 equiv) in distilled water (100 mL). The mix-
ture was stirred at room temperature under argon for 5 h and ex-
tracted with CH2Cl2 (3 ꢁ 50 mL). The combined organic phases
were then washed with water (2 ꢁ 15 mL) and brine (2 ꢁ 15 mL),
and then dried over MgSO4. After filtration, the resultant mixture
was stirred for 2 h with activated charcoal. A filtration on celite
to remove charcoal and an evaporation under reduced pressure
(rotary evaporator, 2 ꢁ 10ꢀ3 bar) afforded the colorless desired
NTfꢀ2 -based second-generation ionic liquid (global yield >80%).
3. Conclusion
In this work, we have demonstrated that the syntheses of
air- and moisture-stable ionic liquids based on bis(trifluorometh-
ylsulfonyl)imide anion could advantageously be optimized as
highlighted by the improved E- and EcoScale factors. In addition,
we unambiguously proved that the environmentally benign
H2O2/NaHCO3 combination can be used for the enantioselective
epoxidation of olefins using a new chiral manganese porphyrin
as a catalyst, if the reaction is carried out in a lipophilic ionic
liquid under ultrasonic activation. Even if the enantioselectivity
obtained is not yet competitive with other methods already re-
ported, and this first example of ultrasonic asymmetric catalysis
in ionic liquids clearly proves that the mechanism involves the
porphyrin catalyst. This novel conditions prevent the bleaching
of the Mn catalyst, which is usually a drawback in peroxide
(and BAP) oxidation catalysis. Further work is currently in pro-
gress in our laboratories to improve the enantioselectivity of the
epoxidation in these conditions and to understand how ultra-
4.2. Preparation of Mn(TPP)OAc
Mn(TPP)OAc was synthesized as described in our previous
paper [6].

132
G. Chatel et al. / Journal of Catalysis 291 (2012) 127–132
4.3. Preparation of chiral Mn porphyrins
reaction mixture was thoroughly washed with an ether/cyclohex-
ane (3:2 v/v) mixture (5 ꢁ 2 mL) to extract the remaining olefin
and the newly formed products. The ether/cyclohexane phase
was analyzed by gas chromatography, based on starting olefin.
4.3.1. Preparation of precursors
aabb-TAPP free base 1 and MeO-substituted binaphthyl diacid
chloride 2 were prepared according to Collman’s method [30].
Appendix A. Supplementary material
4.3.2. aabb-TAPP free base 1 (C44H34N8)
1H NMR (d ppm, CDCl3, 300 MHz): d-2.68 (2 H, s, –NH), 3.55 (8
H, s, NH2), 7.11 (4 H, d, J = 8.0 Hz, phenyl CH), 7.16 (4 H, t, J = 7,4 Hz,
phenyl CH), 7.60 (4 H, t, J = 7.7 Hz, phenyl CH), 7.83 (4 H, d,
J = 7.7 Hz, phenyl CH), 8.90 (8 H, s, CH).
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2012.04.016.
References
[1] (a) B. Meunier, Chem. Rev. 92 (1992) 1411;
4.3.3. MeO-substituted binaphthyl diacid 2 (C24H16O6Cl2)
(b) B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457.
[2] E. Rose, B. Andrioletti, S. Zrig, M. Quelquejeu-Ethève, Chem. Soc. Rev. 34 (2005)
573.
[3] H.H. Monfared, V. Aghapoor, M. Ghorbanloo, P. Mayer, Appl. Catal. A: Gen. 372
(2010) 209.
1H NMR (d ppm, CDCl3, 300 MHz): d 3.42 (6 H, s, OCH3), 7.17 (2
H, d, J = 8.4 Hz, CH), 7.45 (2 H, t, J = 7.0 Hz, CH), 7.56 (2 H, t,
J = 7.8 Hz, CH), 8.08 (2 H, d, J = 8.1 Hz, CH), 8.96 (2 H, s, CH).
[4] K.-H. Tong, K.-Y. Wong, T.H. Chan, Org. Lett. 5 (2003) 3423.
[5] H. Yao, D.E. Richardson, J. Am. Chem. Soc. 122 (2000) 3220.
[6] G. Chatel, C. Goux-Henry, N. Kardos, J. Suptil, B. Andrioletti, M. Draye, Ultrason.
Sonochem. 19 (2012) 390.
[7] (a) R. Rogers, K. Seddon, Ionic Liquids as Green Solvents: Progress and
prospects, An American Chemical Society Publication, 2003.;
(b) K. Seddon, Chem. Commun. (2001) 2399;
(c) P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis, Wiley-VCH,
Weinheim, 2008;
(d) J.P. Hallett, T. Welton, Chem. Rev. 111 (2011) 3508;
(e) C.A.M. Afonso, L.C. Branco, N.R. Candeias, P.M.P. Gois, N.M.T. Lourenço,
N.M.M. Mateus, J.N. Rosa, Chem. Commun. (2007) 2669.
[8] P.G. Jessop, Green Chem. 13 (2011) 1391.
4.3.4. aabinaphtyl-b(binapthyl carboxylic acid)-bNH2 strapped Mn
porphyrin 5 and aabb bis-binaphthyl-strapped Mn porphyrin 6
A
mixture of aabb-TAPP 1 (250 mg, 0.37 mmol), MnBr2
(740 mg, 3.45 mmol), 2,6-lutidine (0.75 mL) in freshly distilled
THF (30 mL) was stirred under argon for 36 h at 23 °C. Metallation
was monitored by TLC. This mixture was brought out and stirred
under air for 3 h. The solvent was removed under vacuum and
the residue was dissolved in CH2Cl2 and filtered to remove the
manganese salt. The CH2Cl2 layer was washed with aqueous NH4Cl
solution, dried over MgSO4, and concentrated to give Mn(aabb-
TAPP)Cl 4.
[9] (a) K.M. Docherty, J.C.F. Kulpa, Green Chem. 7 (2005) 185;
(b) S. Stolte, M. Matzke, J. Arning, A. Böschen, W.-R. Pitner, U. Welz- Biermann,
J. Rankea, Green Chem. 9 (2007) 1170;
In absence of air, anhydrous THF (200 mL, freshly distilled from
sodium) and N,N-diethylaniline (0.36 mL) were added to
Mn(aabb-TAPP)Cl 4. The freshly prepared diacid chloride 2 in
THF (12 mL, freshly distilled) was added to the mixture cooled at
ꢀ20 °C by a syringe pump over 1.5 h. The mixture was allowed
to stir at room temperature overnight. THF was removed by rotary
evaporation, and the residue was taken up in CH2Cl2 and washed
with water (3 ꢁ 25 mL) and brine (2 ꢁ 25 mL). After drying with
Na2SO4, the solution was filtered on celite and reduced to dryness.
The residue was separated into two successive chromatography
columns (eluant: CH2Cl2/MeOH = 100/0 to 80/20). The solution
was stirred with NaCl, filtered, and reduced and dried under vac-
uum to give the desired porphyrin 5 (201 mg, 37%) and porphyrin
6 in 7% yield (40 mg).
(c) M. Alvarez-Guerra, A. Irabien, Green Chem. 13 (2011) 1507;
(d) M. Petkovic, K.R. Seddon, L.P. Rebelo, C. Silva Pereira, Chem. Soc. Rev. 40
(2011) 1383;
(e) C. Pretti, C. Chiappe, D. Pieraccini, M. Gregori, F. Abramo, G. Monni, L.
Intorre, Green Chem. 8 (2006) 238;
(f) D. Zhao, Y. Liao, Z. Zhang, Clean 1 (2007) 42;
(g) R.F.M. Frade, C.A.M. Afonso, Hum. Exp. Toxicol. 29 (2010) 1038;
(h) D. Coleman, N. Gathergood, Chem. Soc. Chem. 39 (2010) 600.
[10] C. Capello, U. Fischer, K. Hungerbühler, Green Chem. 9 (2007) 927.
[11] M. Deetlefs, K.R. Seddon, Green Chem. 12 (2010) 17.
[12] (a) P.T. Anastas, J.C. Warner, Green Chemistry: Theory and Practice, Oxford
University Press, New York, 1998;
(b) P.T. Anastas, M.M. Kirchhoff, Acc. Chem. Res. 35 (2002) 686.
[13] D. Constable, A. Lapkin, Green Chemistry Metrics, Wiley, Chichester, 2008.
[14] M. Deetlefs, K.R. Seddon, Green Chem. 5 (2003) 181.
[15] J.-M. Lévêque, S. Desset, J. Suptil, C. Fachinger, M. Draye, W. Bonrath, G.
Cravotto, Ultrason. Sonochem. 13 (2006) 189.
[16] J.-M. Lévêque, J. Estager, M. Draye, G. Cravotto, L. Boffa, W. Bonrath, Monatsh.
Chem. 138 (2007) 1103.
[17] T. Yim, H.Y. Lee, H.-J. Kim, J. Mun, S. Kim, S.M. Oh, Y.G. Kim, Bull. Korean, Chem.
Soc. 28 (2007) 1567.
[18] A.K. Burrell, R.E. Del Sesto, S.N. Baker, T.M. McCleskeya, G.A. Baker, Green
Chem. 9 (2007) 449.
[19] V. Farmer, T. Welton, Green Chem. 4 (2002) 97.
[20] (a) R.A. Sheldon, Chem. Ind. 23 (1992) 903;
5: MS (ESI+) m/z = 1478.4 (MꢀCl)+; UV–vis (CH2Cl2): kmax/nm
470 (Soret), 570 (Q1) and 759 (Q2) (
500).
e
/L molꢀ1 cmꢀ1 36,400, 4400,
6: HRMS m/z = 1459.3909 (M-Cl)+ for C92H60ClMnN8O8 (ESI+);
UV–vis (CH2Cl2): kmax/nm 480 (Soret), 580 (Q1) and 615 (Q2)
(e
/L molꢀ1 cmꢀ1 38,200, 4800, 3200).
(b) R.A. Sheldon, Green Chem. 9 (2007) 1273.
[21] K. Van Aken, L. Strekowski, L. Patiny, Beilstein J. Org. Chem. 2 (2006) 3.
[22] (a) F. Endres, S. Zein El Abedinw, Phys. Chem. Chem. Phys. 8 (2006) 2101;
(b) T. Yim, H. Yeong Lee, H.-J. Kim, J. Mun, S. Kim, S.M. Oh, Y. Gyu Kim, Bull.
Korean Chem. Soc. 28 (2007) 1567.
[23] L.C. Branco, J.N. Rosa, J.J. Moura Ramos, C.A.M. Afonso, Chem. Eur. J. 8 (2002)
3671.
[24] K. -P Ho, W.-L. Wong, L. Lee, K.-M. Lam, T. Chan, K.-Y. Wong, Chem. Asian J. 5
(2010) 1970.
[25] J.M. Pringle, J. Golding, K. Baranyai, C.M. Forsyth, G.B. Deacon, J.L. Scott, D.R.
MacFarlane, New J. Chem. 27 (2003) 1504.
[26] R.P. Swatloski, J.D. Holbrey, R.D. Rogers, Green Chem. 5 (2003) 361.
[27] V.I. Pârvulescu, C. Hardacre, Chem. Rev. 107 (2007) 2615.
[28] D.J. Flannigan, S.D. Hopkins, K.S. Suslick, J. Organomet. Chem. 690 (2005)
3513.
[29] J.D. Oxley, T. Prozorov, K.S. Suslick, J. Am. Chem. Soc. 125 (2003) 11138.
[30] (a) J.P. Collman, Z. Wang, A. Straumanis, M. Quelquejeu, J. Am. Chem. Soc. 121
(1999) 460;
(b) E. Rose, Q.-Z. Ren, B. Andrioletti, Chem. Eur. J. 10 (2004) 224.
[31] S. Koda, T. Kimura, T. Sakamoto, T. Kondo, H. Mitome, Ultrason. Sonochem. 10
(2003) 14.
4.4. General procedure for alkenes epoxidation under ultrasound
Epoxidation in ionic liquid under ultrasound was carried out as
previously described [6]. Mn(TPP)OAc (1.5
dissolved in the desired ionic liquid (3 mL) under sonication for
2 min. Imidazole (15 mol, 1.0 mg), sodium bicarbonate
lmol, 1.1 mg) was
l
(0.25 mmol, 21 mg, 0.25 equiv), and olefin (1 mmol, 0.10–
0.16 mL, 1 equiv) were added to the solution that was maintained
at 25 °C using a minichiller cooler. Hydrogen peroxide (2.5 mmol,
0.25 mL, 2.5 equiv, 30% solution in water) was added under ultra-
sonic irradiation (3 mm Ø tapered microtip probe, Pelec = 11.5 W,
Pacous.vol = 0.79 W mLꢀ1 in water, determined by calorimetry using
a procedure described in the literature [31]). Additional amounts
of sodium bicarbonate (0.25 mmol, 21 mg, 0.25 equiv) and
hydrogen peroxide (2.5 mmol, 0.25 mL, 2.5 equiv) were added
portion-wise every 15 min over 1 h. After 1 h of sonication, the