Easily recyclable polymeric ionic liquid-functionalized chiral salen Mn(III) complex for enantioselective epoxidation of styrene

Article abstract of DOI:10.1016/j.jcat.2009.02.020

A polymeric ionic liquid (IL)-functionalized chiral salen ligand (PICL) was synthesized by covalent polymerization between amino ({single bond}NH₂) group of 1,3-dipropylamineimidazolium bromide with chloromethyl ({single bond}CH₂Cl) group at two sides of 5,5′ positions in the typical chiral salen ligand. Treatment of the synthesized PICL with Mn(OAc)₂s4H₂O and LiCl under aerobic oxidation yielded the corresponding polymeric IL-functionalized chiral salen Mn(III) complex (PICC). The typical IR bands at 1613, 1540, 570, and 412 cm^-1, as well as the maximum UV-vis absorbed peaks around 433 nm of the PICC were proposed as characteristics of the monomeric salen Mn(III) complex. The PICC was used as a catalyst in the enantioselective epoxidation of styrene. Comparable catalytic activity and enantioselectivity relative to the monomeric chiral salen Mn(III) complex were observed. Furthermore, recovery of the polymeric catalyst was readily accomplished by simple precipitation in n-hexane, and subsequently reused (10 times) without significant loss of reactivity and enantioselectivity.

Full text of DOI:10.1016/j.jcat.2009.02.020

Journal of Catalysis 263 (2009) 284–291
Contents lists available at ScienceDirect
Journal of Catalysis
www.elsevier.com/locate/jcat
Easily recyclable polymeric ionic liquid-functionalized chiral salen Mn(III) complex
for enantioselective epoxidation of styrene
Rong Tan a, Donghong Yin a,b,∗, Ningya Yu a,∗, Haihong Zhao a, Dulin Yin a
a
Institute of Fine Catalysis and Synthesis, Key Laboratory of Chemical Biology and Traditional Chinese Medicine Research (Ministry of Education), Hunan Normal University,
Changsha 410081, Hunan, China
b
Technology Center, China Tobacco Hunan Industrial Corporation, No. 426 Laodong Road, Changsha, Hunan, 410014, China
a r t i c l e i n f o
a b s t r a c t
polymeric ionic liquid (IL)-functionalized chiral salen ligand (PICL) was synthesized by covalent
Article history:
A
Received 17 January 2009
Revised 15 February 2009
Accepted 16 February 2009
Available online 10 March 2009
polymerization between amino (–NH2) group of 1,3-dipropylamineimidazolium bromide with chloro-
methyl (–CH2Cl) group at two sides of 5,5 positions in the typical chiral salen ligand. Treatment of
ꢀ
the synthesized PICL with Mn(OAc)2·4H2O and LiCl under aerobic oxidation yielded the corresponding
polymeric IL-functionalized chiral salen Mn(III) complex (PICC). The typical IR bands at 1613, 1540, 570,
and 412 cm−1, as well as the maximum UV–vis absorbed peaks around 433 nm of the PICC were
proposed as characteristics of the monomeric salen Mn(III) complex. The PICC was used as a catalyst
in the enantioselective epoxidation of styrene. Comparable catalytic activity and enantioselectivity
relative to the monomeric chiral salen Mn(III) complex were observed. Furthermore, recovery of the
polymeric catalyst was readily accomplished by simple precipitation in n-hexane, and subsequently
reused (10 times) without significant loss of reactivity and enantioselectivity.
Keywords:
Chiral salen Mn(III) complex
Ionic liquid
Styrene
Enantioselective epoxidation
© 2009 Elsevier Inc. All rights reserved.
1. Introduction
fortunately, comparing with the homogeneous counterparts, some
of the immobilized complexes often suffer from various disadvan-
tages even easy recovery, such as poor activity, leaching of the
active species into the reaction medium, and low accessibility of
substrates. Therefore, for industrial practical merit and academic
interest of homogeneously catalyzed reactions, the development
of an efficient strategy for catalyst recovery is still challenging.
Cavazzini et al. [15] reported that sterically hindered chiral salen
Mn(III) complexes bearing long perfluoroalkyl substituents showed
similar ee and yield of epoxides compared with standard ho-
mogeneous salen Mn(III) complexes. The advantage was that the
catalysts could be readily recovered from the fluorous layer by
simple phase separation techniques at room temperature. How-
ever, the toxic compounds which derived from the decomposition
of fluorous solvents and fluorous derivatives at high temperature
are often unfriendly to environment. Recently, we had reported a
polymeric chiral salen Mn(III) complex acting as solvent-regulated
phase transfer catalyst in the asymmetric epoxidation of styrene,
which can be performed in “one-phase catalysis and two-phase
separation” by changing solvent [16]. The complex showed dual
advantages of facile separation and high catalytic efficiencies rela-
tive to the corresponding homogeneous chiral salen Mn(III) com-
plex.
Asymmetric epoxidation of prochiral alkenes presents a pow-
erful strategy for the synthesis of enantiomerically enriched epox-
ides. Chiral salen Mn(III) complexes are recognized as one of the
most effective and selective catalysts for enantioselective epoxida-
tion of unfunctionalized olefins since the first chiral C2-asymmetric
salen-type complex synthesized by Zhang et al. in the 1990s [1].
With bulky chiral groups near the metal center, such catalysts give
high catalytic activity and enantioselectivity (ee) for a wide vari-
ety of unfunctionalized alkenes [2,3]. However, the homogeneous
chiral salen Mn(III) catalysts are difficult to recover and recycle.
An approach for overcoming this difficulty is to immobilize the
chiral complex on some supports; thereby creating heterogeneous
chiral catalysts that can be readily recovered from reaction mix-
tures. Many attempts were made to develop supported Mn(III)
salen complexes using organic polymer [4], polymeric membrane
[5] encapsulation in zeolite [6,7] and inorganic supports [8–14]
so as to minimize the degradation of the catalysts, allowing easy
recovery and reusability of the catalyst for a number of cycles. Un-
Corresponding authors at: Institute of Fine Catalysis and Synthesis, Key Labo-
*
ratory of Chemical Biology and Traditional Chinese Medicine Research (Ministry of
Education), Hunan Normal University, Changsha 410081, Hunan, China. Fax: +86 731
8872531.
Room temperature ionic liquids (RTILs) have recently attracted
considerable interests as important solvent in the synthesis and
catalysis owing to their unique solvent power for many organic and
E-mail addresses: yindh@hunnu.edu.cn (D. Yin), yuningya@yahoo.com.cn (N. Yu).
0021-9517/$ – see front matter © 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2009.02.020

R. Tan et al. / Journal of Catalysis 263 (2009) 284–291
285
inorganic substances [17,18]. With the attractive feature of tunable
miscibility, many efforts have been made to use RTILs as phase
tags to facilitate recycling and reuse of the catalysts [19], espe-
cially, using biphasic reaction systems containing RTILs [20]. Song
and Roh reported that the chiral salen Mn(III) epoxidation cata-
lyst immobilized in 1-butyl-3-methylimidazolium hexafluorophos-
phate offered comparable enantioselectivity as well as increased
activity in the asymmetric epoxidation of olefins than that ob-
tained without employment of RTILs [21], probably due to the
positive influence of the polarity of the RTILs on the stabilization
of the transition state [13]. However, the high viscosity of RTILs
often limits mass transfer in the fast chemical reaction. More-
over, some catalysts even in trace quantities are often extracted
during the recovery from the reaction system, causing contamina-
tion of the reaction products and resulting in decrease of catalytic
activity. Some attempts were dedicated to minimize the disad-
vantages by supporting RTILs phases to immobilize homogeneous
catalysts [22–25]. The supported RTILs catalyst combines the ad-
vantages of RTILs and heterogeneous supports. Active species in
the RTILs phase acted as a homogeneous catalyst. For example, Lou
et al. [23] developed an effective method based on supported RTILs
system to immobilize chiral Mn(III) salen complexes, which ex-
hibited excellent activity and enantioselectivity in the asymmetric
epoxidation of unfunctionalized olefins. The immobilized catalysts
were stable and could be recycled three times without loss of ac-
tivities. Lately, we have found that anchoring RTILs moiety onto
chiral salen Mn(III) complex could make the catalysts recoverable
by change of solvent [26]. This method can also provide an ideal
way for combining the advantages of homogeneous catalysts and
reusability.
benzaldehyde were prepared according to the procedures de-
scribed in [27–29], respectively.
2.2. Characterization methods
FT-IR spectra were obtained as potassium bromide pellets with
−1
−1
a resolution of 4 cm
and 32 scans in the range 400–4000 cm
using an AVATAR 370 Thermo Nicolet spectrophotometer. The
ultraviolet–visible light (UV–vis) spectra were recorded on a UV–
vis Agilent 8453 spectrophotometer. The solution of samples in
dichloromethane (ca. 1.0 mM) was poured into a 1 cm quartz
cell for UV–vis adsorption with dichloromethane as the reference.
1H NMR spectra of samples were recorded with a Varian-400 spec-
trometer using tetramethylsilane (TMS) as internal reference. Ther-
mogravimetric and differential thermogravimetric (TG-DTG) curves
were obtained on a NETZSCH STA 449C thermal analyzer. Samples
◦
were heated from room temperature up to 800 C under flow-
ing air using alumina sample holders. The sample weight was ca.
10 mg and the heating rate was 10 K/min. A reference sample
holder contained alumina. Elemental analyses of C, H and N were
carried out on Vario EL III Elemental analyses made in Germany.
The optical rotation of catalysts was measured in dichloromethane
on a WZZ-2A Automatic Polarimeter. Mn ion content was mea-
sured by the method of compleximetry with ethylenediamine
tetraacetic acid (EDTA) according to [30]. The molecular weight
was measured by Ubbelohde viscosimeter. The ee value for styrene
epoxide and analysis of the product were determined by an Agilent
Technologies 6890N gas chromatography equipped with 19091G-
B213 chiral capillary column (30 m × 0.32 mm × 0.25 um) and
flame ionization detector (FID).
Herein, we described a successful synthesis of a PICL by co-
valent polymerization between amino (–NH2) group of 1,3-dipro-
pylamineimidazolium bromide with chloromethyl (–CH2Cl) group
2.3. Preparation of the polymeric IL-functionalized chiral salen Mn(III)
complex (PICC)
ꢀ
at two sides of 5,5 positions in the salen ligand of (R,ꢀR)-{N-
ꢀ
ꢀ
(3-tert-butyl-5-chloromethyl-salicylidine)-N -(3 -tert-butyl-5 -chlo-
romethyl-salicylidine)}-1,2-cyclohexanediamine. One of the most
appealing features of the obtained polymeric complex was that
the ordered distribution of chiral catalytic sites for the single di-
mension polymer matrix offered unique microenvironments for the
reactions, thus, might lead to an enhanced activity. It is expected
that high activity and comparable enantioselectivity can be ob-
tained over the PICC for the asymmetric epoxidation of styrene
because of its special ‘ionophilicity’ and the positive effect of the
polarity of the IL moiety on the stabilization of the transition
state. Also, the PICC with larger molecular weight is miscible with
dichloromethane and insoluble in n-hexane. It therefore will work
under homogeneous conditions during the reaction stage but will
be precipitated at the end of the reaction by simple addition of
n-hexane extraction resulting in facile separation and reuse.
The preparation of the PICC was outlined in Scheme 1.
2.3.1. Synthesis of 1,3-dipropylamineimidazolium bromide (IL)
1,3-Dipropylamineimidazolium bromide was synthesized ac-
cording to the modified procedure described in [31], which was
outlined in Scheme 2.
95% Sodium hydride powder (10.0 mmol) dissolved in tetrahy-
drofuran (8 ml) was cooled in an ice bath, followed by dropwise
addition of a solution of imidazole (10.0 mmol) in tetrahydrofuran
(8 ml). The ice bath was removed and the resulting mixture was
stirred for 2 h at room temperature. Then 3-bromopropylamine
(20.0 mmol) dissolved in tetrahydrofuran (25 ml) was slowly added
at room temperature. The mixture was stirred under reflux for
another 7 h. The lower IL layer was thoroughly rinsed with tetrahy-
drofuran and dried in vacuum to give a pale yellow viscous liq-
uid of 1,3-dipropylamineimidazolium bromide (IL) (2.1 g, yield of
78%). IL: 1H NMR (CDC13, 400 MHz): δ ppm 8.70 (s, 1 H, ring
2. Experimental
NCHN), 7.48 (m, 2 H, ring NCH), 4.30 (t, 4 H, CH2–Nring
,
J =
2.1. Materials
6.6 Hz), 4.18 (m, 4 H, CH2), 2.72 (t, 4 H, CH2–Namine, J = 6.6 Hz),
2.51 (m, 4 H, NH2). FT-IR (KBr): 3426, 2977, 2643, 2013, 1600,
1580, 1491, 1462, 1285, 1168, 1085, 1047, 996, 957, 900, 862, 762,
l-(+)-Tartaric acid, 1,2-diaminocyclohexane, meta-chloropero-
xybenzoic acid (m-CPBA), 3-bromopropylamine, 4-methylmorpho-
line N-oxide (NMO), and 4-phenylpyridine N-oxide (4-PhPyNO)
were purchased from Acros. Pyridine N-oxide (PyNO) was bought
from Fluka. 2-tert-Butyl phenol and 4-(3-phenylpropyl) pyridine
N-oxide (4-PPPyNO) were obtained from Alfa Aesar and Aldrich,
respectively. Other commercially available chemicals were lab-
oratory grade reagents from local suppliers. All solvents were
purified by standard procedures before being used. (R,R)-1,2-
diammoniumcyclohexane mono-(+)-tartrate salt, 3-tert-butyl-2-
hydroxybenzaldehyde, and 3-tert-butyl-5-chloromethyl-2-hydroxy-
−1
621 cm
.
2.3.2. Synthesis of chiral half-unit of ligand (HL)
(R,R)-1,2-diammoniumcyclohexane mono-(+)-tartrate salt (11.2
mmol) was treated with potassium carbonate (22.5 mmol) in
ethanol-distilled water (20 mL, ethanol:H2O = 4:1 (v/v)) under
vigorously stirring. The resulted cloudy mixture was stirred for
another 2 h under reflux. The liberated diamine was extracted
with chloroform (4 × 5 mL), and then 3-tert-butyl-5-chloromethyl-
2-hydroxybenzaldehyde (11.2 mmol) in chloroform (20 mL) was

286
R. Tan et al. / Journal of Catalysis 263 (2009) 284–291
Scheme 1. Synthesis of the PICC.
Scheme 2. Synthesis of 1,3-dipropylamineimidazolium bromide (IL).
◦
added dropwise to the solution of diamine at 0 C. The re-
sulted mixture was stirred under reflux for 48 h. Upon filtra-
tion, a precipitate was collected and recrystallized from ethanol
to get a pale-yellow powder of N-(3-tert-butyl-5-chloromethyl-2-
hydroxybenzaldehyde)-1-amino-2-cyclohexeneimine (HL) (3.23 g,
2.3.3. Synthesis of chiral salen ligand (CL)
The above-obtained HL (8 mmol) dissolved in anhydrous etha-
nol (20 ml) was added dropwise to 3-tert-butyl-5-chloromethyl-
2-hydroxybenzaldehyde (8 mmol) in anhydrous ethanol (20 ml) at
◦
room temperature. The mixture was heated to 60 C for 8 h under
yield of 89%). HL: Calc. for
C
18 H27ClN2O: C, 66.96; H, 8.43;
stirring, and then cooled in ice-water bath over 3 h. The resulted
solid was filtered and recrystallized from ethanol to give a pale-
yellow powder of (R,R)-{N-(3-tert-butyl-5-chloromethyl-salicyli-
dine)-N -(3 -tert-butyl-5 -chloromethyl-salicylidine)}-1,2-cyclohex-
anediamine (CL) (3.38 g, yield of 82%). CL: Calc. for C30H40Cl2-
N2O2: C, 67.79; H, 7.58; N, 5.27%. Found: C, 67.71; H, 7.69; N,
5.32%. 1H NMR (CDC13, 400 MHz): δ ppm 13.68 (s, 2 H, OH), 8.42
(s, 2 H, CH=N), 7.31 (d, 2 H, ArH, J = 2.0 Hz), 7.26 (d, 2 H, ArH,
J = 2.0 Hz), 4.58 (s, 4 H, CH2Cl), 3.55–3.30 (m, 2 H, C=NCH), 1.89–
N, 8.68%. Found: C, 66.89; H, 8.52; N, 8.57%. 1H NMR (CDC13,
400 MHz): δ ppm 13.78 (s, 1 H, OH), 8.29 (s, 1 H, CH=N), 7.28
(d, 1 H, ArH, J = 2.0 Hz), 7.11 (d, 1 H, ArH, J = 2.0 Hz), 4.58 (s,
2 H, CH2Cl), 3.32 (m, 1 H, C=NCH), 2.60 (m, 1 H, C=NCH), 2.21
(m, 2 H, NH2), 2.05–1.44 (m, 8 H, cyclohexyl-H), 1.42 (s, 9 H, t-
butyl-H). FT-IR (KBr): 3449, 2956, 2858, 1630, 1582, 1547, 1480,
ꢀ
ꢀ
ꢀ
1468, 1442, 1386, 1351, 1340, 1325, 1291, 1270, 1240, 1224, 1200,
1173, 1153, 1081, 1054, 1043, 942, 859, 823, 798, 717, 646 cm
−1
.

R. Tan et al. / Journal of Catalysis 263 (2009) 284–291
287
(1 mL) containing 4% of the catalyst (based on monomeric unit of
the PICC) under stirring. The mixture was pre-cooled to the in-
dicated temperature, and then m-CPBA (1 mmol) was added in 4
equal portions at interval of 15 min in a reaction period. Gas chro-
matograph was employed to monitor the progress of the epoxida-
tion reaction. After completion of the reaction, hexane was added
to extract the reaction product. The catalyst was separated from
the reaction system as precipitate and subsequently used without
further purification. The supernatants separated from the reaction
system were concentrated, and further purification of the extrac-
tive by flash column chromatography afforded the pure styrene
epoxides. The reaction products were analyzed by Agilent Tech-
nologies 6890N gas chromatography (FID, 19091G-B213 chiral cap-
illary column (30 m × 0.32 mm × 0.25 um)) using nitrogen as a
carrier gas with flow rate 30 ml/min. The injector temperature,
detector temperature, and oven temperature were 250, 250, and
Scheme 3. Structure of monomeric chiral salen Mn(III) complex.
1.44 (m, 8 H, cyclohexyl-H), 1.42 (s, 18 H, t-butyl). FT-IR (KBr):
3449, 2952, 2866, 1630, 1594, 1545, 1466, 1439, 1390, 1361, 1308,
1266, 1229, 1200, 1167, 1142, 1091, 1030, 974, 930, 871, 778, 753,
−1
669 cm
.
◦
100 C, respectively. The retention times of styrene, R-configuration
styrene epoxide, and S-configuration styrene epoxide were 3.37,
10.1, and 10.5 min, respectively. The authentic samples of R- and
S-configuration styrene epoxide were used as the standard product
to determine the yields by comparison of peak height and area.
2.3.4. Synthesis of PICL
The obtained IL (5 mmol) was added to the synthesized CL
(5 mmol) in dry toluene (15 mL), and then the mixture was re-
◦
fluxed for 48 h under nitrogen protection. After cooled to 5 C
overnight, the obtained PICL was collected by removal of toluene,
and then washed completely with hexane several times, dried in
vacuum to obtain a deep yellow solid of PICL (2.26 g, yield of
91%). PICL: Calc. for (C39H57BrN6O2)17 : C, 64.89; H, 7.96; N, 11.64%.
Found: C, 64.54; H, 8.39; N, 11.63%. 1H NMR (CDC13, 400 MHz):
δ ppm 13.71 (s, 2 H, OH), 8.67 (s, 1 H, ring NCHN), 8.26 (s, 2 H,
CH=N), 7.49 (m, 2 H, ring NCH), 7.39 (d, 2 H, ArH, J = 2.2 Hz), 7.23
(d, 2 H, ArH, J = 2.2 Hz), 4.31 (t, 4 H, CH2–Nring, J = 7.1 Hz), 4.15
(m, 4 H, CH2), 3.97 (m, 4H, ArCH2N), 3.55–3.30 (m, 2 H, C=NCH),
2.70 (t, 4 H, CH2–Namine, J = 7.1 Hz), 2.4 (s, 2 H, CNHC), 1.89–1.44
(m, 8 H, cyclohexyl-H), 1.32 (s, 18 H, t-butyl). FT-IR (KBr): 3434,
2962, 2864, 1630, 1592, 1468, 1439, 1391, 1361, 1307, 1270, 1252,
3. Results and discussion
3.1. Preparation of the PICC
Chiral salen Mn(III) complexes bearing electrondonating groups
has been reported to exhibit higher asymmetric induction than
those bearing electron-withdrawing groups [32]. Ionic liquids de-
rived from N,N-dialkylimidazolium with special polarity can in-
crease the “ionophilicity” of the salen catalyst [13]. In additional,
N,N-dialkylimidazolium cations had positive influence on the sta-
bilization of the transition state during reaction [23]. With these
points in mind, we chose ionic liquid containing imidazolium
moiety to functionalize the chiral salen Mn(III) complex bear-
ing electrondonating groups. As outlined in Scheme 1, a strategy
that we have designed here is to link covalently an IL moiety
with CL by covalent polymerization to make the polymeric ABAB
type ligand resemble RTILs. The covalent linkage between the end
amino (–NH2) groups of 1,3-dipropylamineimidazolium bromide
1240, 1202, 1173, 1135, 1085, 1037, 982, 939, 879, 862, 828, 804,
−1
772, 715, 644, 621 cm
.
2.3.5. Preparation of PICC
◦
Under nitrogen protection and stirring at 50 C, a solution of
Mn(OAc)2·4H2O (8 mmol) in ethanol (15 ml) was added dropwise
to the above-obtained PICL (4 mmol) in ethanol (15 ml). After
refluxing for 5 h, the mixture was cooled to room temperature.
Lithium chloride (24 mmol) in ethanol (10 ml) was added to the
above mixture under stirring for 3 h. After bubbled with a gen-
tle stream of air for another 2 h, the mixture was exposed to air
ꢀ
and chloromethyl (–CH2Cl) groups at two sides of 5,5 positions
ꢀ
ꢀ
of (R,ꢀR)-{N-(3-tert-butyl-5-chloromethyl-salicylidine)-N -(3 -tert-
butyl-5 -chloromethyl-salicylidine)}-1,2-cyclohexanediamine
was
used to form the PICC. At first, the reaction between 3-tert-butyl-5-
chloromethyl-2-hydroxybenzaldehyde and (R,R)-diaminocyclohex-
◦
overnight. The resulting slurry was cooled to 5 C for 2 h, filtered
and washed with 50 ml of water. The obtained solid was dried un-
ane yielded
a compound of HL, and then the obtained HL
◦
der vacuum at 40 C to give a brown powder of PICC (2.85 g, yield
was directly reacted with another 3-tert-butyl-5-chloromethyl-2-
hydroxybenzaldehyde to produce a CL. Subsequently, the polymer-
ization of 1,3-dipropylamineimidazolium bromide reacted with the
obtained CL to afford the PICL. Treatment of the PICL with man-
ganese(II) acetate tetrahydrate under nitrogen gave the dianionic
complex. The acetic ion was replaced by chloride ion with lithium
chloride at room temperature. The dianionic complex was readily
oxidized by oxygen, affording the PICC. The average viscosimetric
molecular weight of the PICC measured by Ubbelohde viscosimeter
was ca. 13727, and the number of repeating units of the polymer
was determined on the basis of the average viscosimetric molec-
ular weight of the PICC and the molecular weight of one building
block (Mv = ∼13727, n = ∼17). In addition, Mn ion content of the
PICC measured by compleximetry was 1.13 mmol/g, which was
close to the theoretical value (1.19 mmol/g). Moreover, it is found
that the PICC with the characteristic of ‘ionophilicity’ is miscible
with polar and weak polar solvent such as dichloromethane, but
insoluble in non-polar solvent such as n-hexane. It is suggested
that the ionophilical PICC should be an easily recoverable catalyst
of 85%). PICC: Calc. for (C39H55BrClMnN6O2)17 : C, 57.82; H, 6.84; N,
10.37%. Found: C, 57.53; H, 7.06; N, 10.39%. FT-IR (KBr): 3384, 2954,
2924, 2856, 1613, 1596, 1542, 1457, 1436, 1390, 1359, 1304, 1265,
1231, 1201, 1145, 1089, 1028, 1000, 926, 869, 797, 752, 699, 661,
621, 570, 412 cm−1; UV–vis (CH2Cl2): 433, 326 nm; Mn ion con-
28
tent: 1.13 mmol/g (theoretical value: 1.19 mmol/g); [α]D = +610
(C = 0.04, CH2Cl2).
2.4. Preparation of the monomeric chiral salen Mn(III) complex
For comparison with the PICC, we also synthesized the mono-
meric chiral salen Mn(III) complex according to the previous re-
port [27], which was outlined in Scheme 3.
2.5. Enantioselective epoxidation of styrene
Enantioselective epoxidation of styrene was typically performed
according to the following procedure. Styrene (0.5 mmol) and a de-
sirable amount of donor ligand were added into dichloromethane

288
R. Tan et al. / Journal of Catalysis 263 (2009) 284–291
Fig. 1. FT-IR spectra of the monomeric chiral salen Mn(III) complex (a), PICC (b) and
IL (c).
Fig. 2. UV–vis spectra of the monomeric chiral salen Mn(III) complex (a) and the
PICC (b).
for asymmetric epoxidation of alkenes by simple phase separation
techniques via change of solvents.
3.2. Sample characterizations
3.2.1. FT-IR and UV–vis spectroscopy
The synthesized PICC, as well as monomeric chiral salen Mn(III)
complex and IL for comparison, was characterized by FT-IR and
UV–vis spectra. As shown in Fig. 1b, the FT-IR spectra of the PICC
−1
exhibited characteristic IR bands at 1613, 1540, 570, and 412 cm
,
which were associated with the stretching vibration modes of
CH=N, C–O, Mn–O, and Mn–N [10], respectively. The typical IR
bands were resemble to that in the spectra of monomeric chi-
ral salen Mn(III) complex (Fig. 1a). Furthermore, Fig. 1b presented
−1
new characteristic band at 3384 cm
assigned to N–H stretching
vibrations of the second amine group [33], which confirmed the
formation of second amine resulting from the reaction between the
end amino (–NH2) groups of IL and chloromethyl (–CH2Cl) groups
of the salen ligand. Comparing with the spectrum of IL (Fig. 1c),
Fig. 3. TG-DTG curves of the PICC.
−1
the PICC showed another band at 621 cm
(Fig. 1b) that is as-
cribed to imidazolium fragments [34]. All of the IR spectra results
proposed that the PICC contained characteristics of salen Mn(III)
complex and imidazolium moiety of IL.
the positive charge of the complex was balanced by chloride ion
but not by the amine group of the IL. A second weight loss ap-
peared at 230 C and was followed by an additional large weight
◦
Fig. 2 showed UV–vis spectra of the PICC and monomeric chi-
ral salen Mn(III) complex. The UV–vis spectra of the PICC and the
monomeric one were almost identical, which indicated that no
change in the local environment of the Mn(III) coordination center
in the complex took place. The main broad absorbed peak ap-
peared near 433 nm of the two complexes ascribed to the charac-
teristic charge transfer band of the ligand-to-metal charge transfer
transitions (MLCT) [35]. The UV–vis spectra further confirmed the
presence of intact catalytic active sites in the PICC, which should
be similar as the monomeric chiral salen Mn(III) complex in the
catalytic epoxidation of alkenes.
◦
◦
loss at 320 C which extended up to ca. 440 C. The two steps
(ca. 56%) were well distinguished in the corresponding DTG curve,
which were assigned to the successive cleavage of the salen lig-
and moieties (theoretical loss: 56.2%) [37]. The final large weight
◦
loss (ca. 32%) centers at 600 C. From the loss in weight, we ten-
tatively assigned these steps to the complete decomposition of
the IL (theoretical loss: 32.4%). Above this temperature, the re-
◦
maining organics were oxidized up to 800 C. The non-removable
residue of ca. 11% belonged to the formation of manganese oxide in
air atmosphere at high temperature. The thermal analysis further
demonstrated that the imidazolium cation moiety was successfully
polymerized with chiral salen Mn(III) complex by covalent poly-
merization to form the PICC, and also the synthesized PICC was
quite stable at room temperature.
3.2.2. Thermoanalysis
Thermal analysis (TG-DTG) had been used to monitor the de-
composition profiles of the PICC, which was presented in Fig. 3.
The polymeric complex showed four distinct steps of weight loss
in the combined TG-DTG curves, when heated under airflow. The
3.3. Catalytic performances
◦
first loss in weight (ca. 4.4%) centers at 130 C. As mentioned,
this weight loss was due to a release of chloride anions as hy-
drogen chloride (theoretical loss: 4.36%) [36]. It was indicated that
As expected according to its structure, the PICC was misci-
ble with dichloromethane and completely insoluble in n-hexane.

R. Tan et al. / Journal of Catalysis 263 (2009) 284–291
289
Table 1
The results of the enantioselective epoxidation of styrene over different chiral salen
Mn(III) complexes.a
−3
Catalyst
Donor ligand
(mmol)
Run
times
Yield
(%)b
ee
TOFd ×10
−1
)
(%)c
(s
No
No
PyNO (1)
NMO (2)
PyNO (1)
/
/
32
36
92
0
0
/
/
Monomeric chiral
salen Mn(III) complexe
PICCe
Fresh
35 (R)
3.19
PyNO (1)
PyNO (1)
PyNO (1)
PyNO (1)
PyNO (1)
Fresh
2nd
4th
7th
10th
99
98
98
98
98
39 (R)
38 (R)
38 (R)
38 (R)
38 (R)
3.44
3.41
3.41
3.41
3.41
a
The epoxidation of styrene was performed with 0.5 mmol styrene and 1 mmol
◦
m-CPBA in 1 mL CH2Cl2 at 0 C for 2 h.
b
Yield of the isolated epoxide.
Determined by GC.
Turnover frequency (TOF) is calculated by the expression of [product]/
c
d
[catalyst] × time (s−1).
e
The amount of catalyst is 4% of styrene.
ꢀ
Fig. 4. FT-IR spectra of the fresh PICC (a), the PICC after the 10th reuse (a ); and the
supernatants by the addition of n-hexane (b).
Therefore, we chose dichloromethane as reaction medium in the
asymmetric epoxidation of styrene. The catalytic activity and enan-
tioselectivity, as well as conditions employed, were contained in
Table 1. For the sake of comparison, the results of a control exper-
iment in dichloromethane in which monomeric chiral salen Mn(III)
complex was used as a catalyst was also included in Table 1. It is
found that low styrene epoxide yield was obtained with racemic
epoxide (0% ee) when catalyst was absent even in the presence
of donor ligand of PyNO or NMO. Therefore, the use of catalyst
becomes very important for improving the yield and enantioselec-
tivity of styrene epoxides. In comparison with monomeric chiral
salen Mn(III) complex, the PICC showed comparable catalytic ac-
tivity (TOF) and ee value. This is due to the fact that the PICC can
dissolve in dichloromethane acting as homogeneous catalyst in the
reaction system, leading to excellent catalytic activity (>99% yield)
and comparable enantioselectivity (39% ee value) when m-CPBA
was used as an oxidant.
After completion of the reaction, the PICC was recovered from
the reaction mixture by simply adding solvent of n-hexane and,
used for the subsequent catalytic runs. Table 1 described the use
of the PICC in ten catalytic cycles. As can be seen in Table 1,
the PICC could be recycled up to 10 times with no appreciable
loss in yield and enantioselectivity of styrene epoxides. The main-
tenance of catalytic activity observed for the PICC should origin
from the unique solubility and the stable structure of the complex.
More importantly, the possibility of formation of undesired inactive
dimeric μ-oxo-manganese(IV) dimmers, a main reason for deac-
tivation of homogeneous salen Mn(III) complex, can be avoided
because of the severe restriction of active site (site isolation) in the
rigid one-dimension polymer framework [37]. The retention of cat-
alytic efficiency suggested that the PICC is perfectly stable during
the epoxidation reaction presented here, and is readily recyclable
from the reaction system.
ꢀ
Fig. 5. UV–vis spectra of the fresh PICC (a), the PICC after the 10th reaction (a ), and
the supernatants by addition of n-hexane (b).
trum (see Fig. 5b). The characteristic IR bands of the supernatants
−1
at 1613, 1540, 570, and 412 cm
of the PICC were not observed
in Fig. 4b. The characteristic UV–vis absorbed peak of the super-
natants near 433 nm was also not presented in Fig. 5b. These ob-
servations suggested that there is no extraction metal complex in
the supernatants, which clearly indicated that the polymeric com-
plex was insoluble completely in n-hexane and could be reused in
consecutive runs without loss of manganese complex during the
enantioselective epoxidation reaction.
After recovery of the PICC by extraction with n-hexane, the
PICC could be precipitated and separated easily by decantation. The
molecular weight of the recovered catalyst (Mv = 13709) was mea-
sured which was very similar to the fresh complex (Mv = 13727).
The chemical analysis of manganese of the recovered PICC gave
manganese content identical to that of the fresh one. And also, the
Mn content in the supernatants was further detected and no Mn
leaching was found for the catalytic system. FT-IR spectra and UV–
vis spectra of the PICC with fresh and reused 10 times (see Fig. 4,
It is well known that the mechanism of olefins epoxidation
catalyzed by salen Mn(III) complexes is oxygen transfer from
the oxo-manganese complex to the olefin [38–40], where the
O=MnV(salen) formation by oxidation is regarded as the catalytic
+
active site. We used UV–vis spectroscopy technique to monitor on-
line the formation of O=MnV(salen) in the PICC catalytic process
+
ꢀ
ꢀ
a vs a and Fig. 5, a vs a ) indicated that no significant changes
took place even after reuse for 10 times. The supernatants ex-
tracted from the catalyst separation after ten recycles were also
characterized by FT-IR spectrum (see Fig. 4b) and UV–vis spec-
with addition of m-CPBA. Herein, 1 mmol of m-CPBA was added
to the reaction mixture in 4 equal portions at interval of 15 min
in a reaction period. The stepwise overlay of UV–vis spectra with

290
R. Tan et al. / Journal of Catalysis 263 (2009) 284–291
Table 2
The catalytic results of the asymmetric epoxidation of styrene with various donor
ligands and performed at different temperature.a
−3
Donor ligand
(mmol)
Temperature
( C)
Yield
(%)b
ee
TOFd ×10
◦
−1
)
(%)c
(s
No
NMO (2)
0
52
96
99
99
93
88
99
99
99
99
98
98
0
1.80
3.34
3.44
3.44
3.23
3.06
3.44
3.44
3.44
3.44
3.41
3.41
−40
−20
0
20
35
39 (R)
48 (R)
44 (R)
37 (R)
34 (R)
41 (R)
39 (R)
40 (R)
37 (R)
39 (R)
35 (R)
PyNO (1)
−20
0
4-PPPyNO (1)
4-PhPyNO (1)
−20
0
−20
0
a
PICC (4% of styrene), styrene (0.5 mmol), m-CPBA (1 mmol), CH2Cl2 (1 mL), 2 h.
b
c
Same as in Table 1.
Same as in Table 1.
Same as in Table 1.
d
ands, low yield (52%) and racemic epoxides (0% ee) of styrene were
obtained even in the case of PICC. Dramatic increases of the yield
and enantioselectivity of the epoxide were achieved by addition of
donor ligands, which suggested that the donor ligands were es-
sential to the attainment of high activity and enantioselectivity. It
is well known that adding donor ligands to chiral salen Mn(III)
complex lead to a well-defined molecular geometry about the Mn
center which is otherwise absent, thus optimizing the reactivity
and conformation of the complex [38]. Moreover, the additives
were coordinated to the metal atom at the axial position through
the oxygen atom, which weakens the O=Mn bond in O=MnV
−7
Fig. 6. UV–vis spectra of the solution of PICC in dichloromethane (5 × 10
M) con-
−5
−5
taining styrene (1.25 ×10
M) and PyNO (2.5 ×10
M) by addition of m-CPBA at
different time, (a1) before addition of m-CPBA, (a5) the last interval addition of m-
CPBA, (a6) after the last interval addition of m-CPBA for another 1 h. Inset shows the
+
(salen) ion and prevented the formation of unreactive Mn(IV)-oxo
successive changes of UV–vis spectra observed upon adding the dichloromethane
dimmers, hence activated and stabilized the catalyst [42]. Compar-
ing with the various donor ligands of NMO, PyNO, 4-PhPyNO and
4-PPPyNO, the most excellent enantioselectivities (48%) could be
obtained with NMO, but the kinds of donor ligands had little ef-
fects on the yield of styrene epoxide. It was reported that NMO
had some additional roles for the Jacobsen’s epoxidation system.
Palucki et al. observed that a 1:1 salt was generated between NMO
and m-CPBA in CH2Cl2 which was unreactive towards alkenes but
oxidizes the chiral salen Mn(III) catalyst even at low temperature
to form the catalytic active species Mn(V)-oxo complexes. Also, ex-
cess NMO was critical in preventing the uncatalyzed epoxidation
pathways that take place in the absence of the additive [43].
Low temperature was well known to be conducive to high effi-
ciency of the salen Mn(III) complexes for the epoxidation of non-
functionalized alkenes [44]. The effect of reaction temperature on
the epoxidation in the presence of the PICC catalyst was listed
in Table 2. Obviously, the catalytic activity and enantioselectivity
of the PICC in the epoxidation of styrene were improved at low
temperature. With a decrease of reaction temperature from 35 to
−5
solution of m-CPBA (2.5 × 10
M, 4 mL) in four equal portions at average 15 min
intervals, (a1) 0 min, (a2) 15 min, (a3) 30 min, (a4) 45 min, (a5) 60 min.
each addition of the oxidant of m-CPBA were recorded in Fig. 6 in-
set. Curve of a1 was the UV–vis spectrum of the PICC dissolved in
dichloromethane containing styrene and donor ligand of PyNO be-
fore addition of m-CPBA. The typical adsorbed peak around 433 nm
assigned to the MLCT of salen Mn(III) complex could be observed.
After the first addition of m-CPBA, the typical peak (MLCT) shifted
from 433 nm to 414 nm (see Fig. 6 inset curve a2), which indicated
+
the beginning formation of O=MnV(salen) species in the process
by addition of oxidant [2]. The curves a3–a5 recorded the UV–vis
spectrum of the PICC after 2nd, 3rd and 4th addition of m-CPBA,
respectively. The gradual shift from 433 nm (Fig. 6a1) to 398 nm
+
(Fig. 6a5) was attributed to the formation of O=MnV(salen) dur-
ing the epoxidation process with addition of oxidant. The absorp-
+
tion attributable to O=MnV(salen) present throughout the reac-
tion until all the m-CPBA has been consumed, which implies that
O-transfer from the complex to styrene is a rate-limiting step un-
der the reaction conditions. After complete reaction for 2 h, the
recorded UV–vis spectrum was very similar to the original one
(Fig. 6, a6 vs a1). These characteristic changes recorded by UV–
vis spectra strongly suggested that the PICC is perfectly stable and
not easy degradation during the epoxidation reaction, and can be
readily reused for the subsequent catalytic runs.
◦
−20 C, a significant increase of styrene epoxide yield (from 88%
to 99%) and enantioselectivity (from 34% to 48%) to the styrene
epoxide with R-configuration were obtained over the donor ligand
of NMO. The increase in styrene epoxide yield with a decrease of
temperature could be due to that low temperature restrained the
decomposition of m-CPBA, enhancing the availability of the oxi-
dant. Furthermore, asymmetric epoxidation of terminal olefins was
subject to a special type of enantiomeric “leakage” pathway, which
has been demonstrated to proceed via stepwise, nonstereospecific
oxo-transfer mechanism [43]. At low temperature, the enantiofa-
cial selectivity in the first C–O bond-forming step of epoxidation
was enhanced and the trans-pathway in the second C–O bond-
forming step was suppressed, thereby, increasing the ee value in
the asymmetric epoxidation of alkenes. Although low tempera-
ture was found to induce catalytic activity and enantioselectivity
3.4. Effect of donor ligands and reaction temperature
Donor ligand plays a crucial role in the asymmetric epoxidation
of alkenes catalyzed by salen Mn(III) complex using m-CPBA as ox-
idants [41]. The influence of various donor ligands viz NMO, PyNO,
4-PPPyNO, and 4-PhPyNO on the enantioselective epoxidation of
styrene were summarized in Table 2. In the absence of donor lig-

R. Tan et al. / Journal of Catalysis 263 (2009) 284–291
291
under the conditions presented here, both the epoxide yield and
enantioselectivity decreased with further decreasing reaction tem-
[5] I.F.J. Vankelecom, D. Tas, R.F. Parton, V.V. de Vyver, P.A. Jacobs, Angew. Chem.
Int. Ed. Engl. 35 (1996) 1346.
[6] S.B. Ogunwumi, T. Bein, Chem. Commun. 1997 (1997) 901.
[7] M.J. Sabater, A. Corma, A. Domenech, V. Fornes, H. Garcia, Chem. Com-
mun. 1997 (1997) 1285.
[8] R.I. Kureshy, N.H. Khan, S.H.R. Abdi, I. Ahmael, S. Singh, R.V. Jasra, J. Catal. 221
(2004) 234.
[9] S. Bhattacharjee, J.A. Anderson, Adv. Synth. Catal. 348 (2006) 151.
[10] B.M. Choudary, T. Ramani, H. Maheswaran, L. Prashant, K.V.S. Ranganath, K.V.
Kumar, Adv. Synth. Catal. 348 (2006) 493.
◦
perature from −20 to −40 C. The reason is that the extremely
low reaction temperature limits the solubility of the PICC. After
completion of the reaction, some solid catalyst granules could be
observed, which indicated that the PICC could not be dissolved in
◦
dichloromethane completely at −40 C. It should be noted that un-
der optimal reaction conditions, 99% of the epoxide yield with 48%
ee value could be obtained using the PICC as a catalyst and NMO
as a donor ligand.
[11] H. Zhang, S. Xiang, C. Li, Chem. Commun. 5 (2005) 1209.
[12] Y. Zhang, J. Zhao, L. He, D. Zhao, S. Zhang, Microporous Mesoporous Mater. 94
(2006) 159.
[13] C. Baleizã, H. Garcia, Chem. Rev. 106 (2006) 3987.
[14] T. Belser, E.N. Jacobsen, Adv. Synth. Catal. 350 (2008) 967.
[15] M. Cavazzini, A. Manfredi, F. Montanari, S. Quici, G. Pozzi, Chem. Commun.
(2000) 2171.
[16] R. Tan, D. Yin, N. Yu, L. Tao, Z. Fu, D. Yin, J. Mol. Catal. A Chem. 259 (2006)
125.
[17] C. Li, Z.K. Zhao, Adv. Synth. Catal. 349 (2007) 1847.
[18] D. Yin, C. Li, B. Li, L. Tao, D. Yin, Adv. Synth. Catal. 347 (2005) 137.
[19] L. Gaillon, F. Bedioui, Chem. Commun. 1 (2001) 1458.
[20] L. Tao, D. Yin, R. Tan, N. Yu, D. Yin, Chin, J. Catal. 27 (2006) 440.
[21] C.E. Song, E.J. Roh, Chem. Commun. 10 (2000) 837.
[22] C. Baleizão, B. Gigante, H. Garcia, A. Corma, Tetrahedron Lett. 44 (2003) 6813.
[23] L. Lou, K. Yu, F. Ding, W. Zhou, X. Peng, S. Liu, Tetrahedron Lett. 47 (2006) 6513.
[24] B.M.L. Dioos, P.A. Jacobs, J. Catal. 243 (2006) 217.
[25] S. Sahoo, P. Kumar, F. Lefebvre, S.B. Halligudi, Tetrahedron Lett. 49 (2008)
4865.
4. Conclusions
In conclusion, by covalent polymerization between the end
amino (–NH2) groups in a imidazolium cation and chloromethyl
(–CH2Cl) groups in a salen ligand, we have obtained a polymeric
IL-functionalized chiral salen ligand and the corresponding poly-
meric chiral salen Mn(III) complex. The obtained polymeric com-
plex was insoluble in non-polar organic solvents but totally mis-
cible with dichloromethane. With the intriguing peculiarity of sol-
ubility, the polymeric chiral complex showed comparable activity
and enantioselectivity as a catalyst in asymmetric epoxidation of
styrene relative to the monomeric chiral salen Mn(III) complex
and readily recovered from reaction system by simple precipita-
tion. Moreover, with the catalytic site isolation of the rigid polymer
framework, the polymeric chiral complex could be used at least 10
times without significant loss of catalytic activity and enantioselec-
tivity. On the basis of styrene enantioselective epoxidation results,
the present study provides a novel and promising chiral salen cat-
alyst for the enantioselective epoxidation of unfunctional olefins.
[26] R. Tan, D. Yin, N. Yu, Y. Jin, H. Zhao, D. Yin, J. Catal. 255 (2008) 287.
[27] J.F. Larrow, E.N. Jacobsen, Y. Gao, Y. Hong, X. Nie, C.M. Zepp, J. Org. Chem. 59
(1994) 1939.
[28] M. Hayashi, Y. Miyamoto, T. Inoue, N. Oguni, J. Org. Chem. 58 (1993) 1515.
[29] F. Minutolo, D. Pini, A. Petri, P. Salvadori, Tetrahedron: Asymmetry 7 (1996)
2293.
[30] H. Ren, Y. Tian, H. Liang, The Handbook of Analytical Chemistry, vol. 2, Chemi-
cal Industry Press, Beijing, 1997, p. 559.
[31] S.V. Dzyuba, R.A. Bartsch, Chem. Commun. 1 (2001) 1466.
[32] E.N. Jacobsen, W. Zhang, M.L. Güler, J. Am. Chem. Soc. 113 (1991) 6703.
[33] K. Su, T. Pan, Y. Zhang, The Analytic Method of Spectrum, East China Univ. of
Science & Technology Press, Shanghai, 2002, p. 99.
[34] E.R. Talaty, S. Raja, V.J. Storhaug, A. Dölle, W.R. Carper, J. Phys. Chem. B 108
(2004) 13177.
Acknowledgments
The project was financial supported by the Natural Science
Foundation of Hunan Province (07JJ3027) and Scientific Research
Foundation for the Returned Overseas Chinese Scholars, Ministry
of Education [2003-406]. The authors also thank the National Nat-
ural Science Foundation of China (Grant No. 20803021).
[35] V.D. Chaube, S. Shylesh, A.P. Singh, J. Mol. Catal. A Chem. 241 (2005) 79.
[36] L. Frunza, H. Kosslick, H. Landmesser, E. Hijft, R. Fricke, J. Mol. Catal.
Chem. 123 (1997) 179.
A
[37] Q. Xia, H. Ge, C. Ye, Z. Liu, K. Su, Chem. Rev. 105 (2005) 1603.
[38] T. Katsuki, Adv. Synth. Catal. 344 (2002) 131.
[39] D.A. Plattner, D. Feichtinger, J. El-Bahraoui, O. Wiest, Int. J. Mass Spec-
trom. 195/196 (2000) 351.
References
[40] E.M. McGarrigle, D.G. Gilheany, Chem. Rev. 105 (2005) 1563.
[41] P. Pietikäinen, J. Mol. Catal. A Chem. 165 (2001) 73.
[42] J.P. Collman, L. Zeng, J.I. Brauman, Inorg. Chem. 43 (2004) 2672.
[43] M. Palucki, P.J. Pospisil, W. Zhang, E.N. Jacobsen, J. Am. Chem. Soc. 116 (1994)
9333.
[1] W. Zhang, J.L. Loebach, S.R. Wilson, E.N. Jacobsen, J. Am. Chem. Soc. 112 (1990)
2801.
[2] K. Srinivasan, P. Michaud, J.K. Kochi, J. Am. Chem. Soc. 108 (1986) 2309.
[3] T. Katsuki, J. Mol. Catal. A Chem. 113 (1996) 87.
[4] L. Canali, E. Cowan, H. Deleuze, C.L. Gibson, D.C. Sherrington, J. Chem. Soc.
Perkin Trans. 1 (2000) 2055.
[44] M. Palucki, G.J. McCormick, E.N. Jacobsen, Tetrahedron Lett. 36 (1995) 5457.