Highly active electrophile-nucleophile catalyst system for the cycloaddition of CO2 to epoxides at ambient temperature

Article abstract of DOI:10.1016/j.jcat.2004.07.018

The cycloaddition of CO₂ to epoxides proceeds effectively under extremely mild temperatures and pressures by using a bifunctional catalyst system of tetradentate Schiff-base aluminum complexes (SalenAlX) as electrophile in conjunction with polyether-KY complexes as nucleophile. The steric factor of the substituent groups on the aromatic rings of SalenAlX and the nucleophilicity and leaving ability of the anion Y^-1 of polyether-KY complexes all have great effects on the activity of the bifunctional catalyst system. The reaction of CO₂ with (S)-propylene oxide in the presence of the SalenAlEt/18-crown-6-KI catalyst system gives (S)-propylene carbonate in > 99 % ee with retention of stereochemistry.

Full text of DOI:10.1016/j.jcat.2004.07.018

Journal of Catalysis 227 (2004) 537–541
www.elsevier.com/locate/jcat
Research Note
Highly active electrophile–nucleophile catalyst system for the
cycloaddition of CO2 to epoxides at ambient temperature
Xiao-Bing Lu ∗, Ying-Ju Zhang, Kun Jin, Li-Mei Luo, Hui Wang
State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116012, People’s Republic of China
Received 16 May 2004; revised 15 July 2004; accepted 16 July 2004
Available online 27 August 2004
Abstract
The cycloaddition of CO to epoxides proceeds effectively under extremely mild temperatures and pressures by using a bifunctional
2
catalyst system of tetradentate Schiff-base aluminum complexes (SalenAlX) as electrophile in conjunction with polyether–KY complexes
as nucleophile. The steric factor of the substituent groups on the aromatic rings of SalenAlX and the nucleophilicity and leaving ability of
−1
the anion Y of polyether–KY complexes all have great effects on the activity of the bifunctional catalyst system. The reaction of CO
2
with (S)-propylene oxide in the presence of the SalenAlEt/18-crown-6–KI catalyst system gives (S)-propylene carbonate in > 99% ee with
retention of stereochemistry.
2004 Elsevier Inc. All rights reserved.
Keywords: Carbon dioxide; Epoxide; Cyclic carbonate; Cycloaddition; Salen–aluminum complexes; 18-crown-6; KI; Synergistic effect
1. Introduction
been significant, all suffer from low catalyst reactivity, the
need for cosolvent, or the requirement for high pressure
and/or high temperature. Very recently, some highly efficient
catalyst systems have been reported to give excellent activ-
ity, but they usually exhibit no or very low catalytic activity
at ambient temperature [20–22]. For example, for a recently
reported catalyst system, Cr(III)Salen/DMAP, which oper-
ated at 100 psig CO2 pressure, the decrease of reaction tem-
perature from 75 to 25 ◦C resulted in TOF from 254 h−1
rapidly decreasing to 3 h−1 [20]. Therefore, the exploration
of highly efficient catalysts for coupling CO2 with epoxides
under low temperatures and low CO2 pressure still remains
a challenging problem, though organometallic Pd(0) com-
plexes have been demonstrated to be effective in coupling
CO2 with the much higher reactivity of vinyl epoxides un-
der mild conditions [23,24]. The key to accomplishing this
goal is the rational design of improved catalysts based on
a mechanistic understanding. A general pattern in the reac-
tion of CO2 with epoxides has emerged wherein catalysis
involves activation of both epoxides and CO2 [13]. Our re-
cent research found that the synergistic actions of an elec-
trophile (SalenMX complexes) in conjunction with a nucle-
Conversion of carbon dioxide to industrially useful com-
pounds has received much attention because of its potential
use as an abundant carbon source and its indirect role as an
environmental pollutant [1]. One of the most promising re-
actions in this area is the synthesis of five-membered cyclic
carbonates via the cycloaddition of CO2 to epoxides [2].
These carbonates can be used as aprotic polar solvents and
as precursors for polycarbonates and other polymeric ma-
terials [3]. In recent decades, numerous catalyst systems,
including amines [4], quaternary ammonium or phospho-
nium salts [5,6], ionic liquids [7], alkali metal salts alone [8]
or in combination with crown ethers [9], halostannanes [10],
organoantimony halide [11], MgO [12] or Mg–Al mixed ox-
ides [13], solid base [14,15], porphyrin [16,17], phthalocya-
nine [18], and transition-metal complexes [19], have been
developed for this transformation. While the advances have
*
Corresponding author. Fax: +86 411 83633080.
E-mail address: lxb-1999@163.com (X.-B. Lu).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2004.07.018
2004 Elsevier Inc. All rights reserved.

538
X.-B. Lu et al. / Journal of Catalysis 227 (2004) 537–541
ophile (quaternary ammonium salts) were more effective in
catalyzing this reaction [25].
(0.24 mmol), 18-crown-6 (0.24 mmol), and then propylene
oxide (192 mmol) by means of a hypodermic syringe in
a nitrogen atmosphere. The 18-crown-6–KI complex was
formed and easily dissolved in propylene oxide, and then
SalenAlEt 1a was added under dry nitrogen. When the
catalysts were completely dissolved, the mixture solution
was charged into the autoclave via a syringe in a CO2 at-
mosphere. The autoclave was put into a bath and heated to
the desired temperature. Then, CO2 was charged into the
autoclave and the pressure was kept constant during the re-
action. After the expiration of the desired time, the excess
gases were vented. The remainder mixture was degassed and
fractionally distilled under reduced pressure for obtaining
pure cyclic carbonates.
In the present paper, we report a new bifunctional cat-
alyst system based on polyether–KY complexes as nucle-
ophile and tetradentate Schiff-base aluminum complexes
(SalenAlX) with varying diamine backbones, or/and the ax-
ial X group or/and substitute groups on the aromatic rings, as
electrophile for cooperatively catalyzing the cycloaddition
of CO2 to epoxides, and further elucidating the structure–
activity relationship and synergistic mechanism of the bi-
functional catalyst system.
2. Experimental
2.4. Characterization of cyclic carbonates
2.1. Chemicals
The 1H NMR (Varian INOVA-400-type spectrometer,
400 MHz, CDCl3/TMS) data of these cyclic carbonates are
listed as follows: ethylene carbonate {δ 4.51 (s, 4H)}; propy-
lene carbonate {δ 1.50 (3H, d, J = 6.4 Hz, CH3), 4.05 (1H,
t, J = 8.0 Hz, CH), 4.55 (1H, t, J = 8.0 Hz, CH), 4.85 (1H,
m, CH)}; 1,2-butene carbonate {δ 1.03 (3H, t, J = 7.6 Hz,
CH3), 1.75–1.86 (2H, m, CH2), 4.11 (1H, dd, J = 8.4,
J = 6.8 Hz, CH), 4.55 (1H, t, J = 8.2 Hz, CH), 4.66–4.73
(1H, m, CH)}; chloropropylene carbonate {δ 3.73–3.89 (2H,
m, CH2), 4.42 (1H, dd, J = 5.6, J = 5.6 Hz, CH), 4.62
(1H, t, J = 8.6 Hz, CH), 5.03–5.09 (1H, m, CH)}; 1,2-
hexene carbonate {δ 0.93 (3H, t, J = 6.6 Hz, CH3), 1.37–
1.49 (4H, m, –CH2CH2–), 1.67–1.85 (2H, m, CH2), 4.08
(1H, t, J = 7.6 Hz, CH), 4.54 (1H, t, J = 8.2 Hz, CH), 4.68–
4.75 (1H, m, CH)}.
Propylene oxide, ethylene oxide, and 1,2-butene oxide
were refluxed over a mixture of KOH and CaH2, and frac-
tionally distilled under a nitrogen atmosphere. Epichloro-
hydrin was refluxed over CaH2 and fractionally distilled
under a nitrogen atmosphere. (S)-Propylene oxide and 1,2-
hexene oxide were used as received from Acros Company.
trans-Deuterio-1,2-hexene oxide was synthesized accord-
ing to the literature [26,27]. CO2 was purified by passing
through a column packed with 4 Å molecular sieve before
use. Triethylaluminum (Et3Al) and chlorodiethylaluminum
(Et2AlCl) were fractionally distilled under reduced pres-
sure in a nitrogen atmosphere. (1R,2R)-N,Nꢀ-bis(3,5-di-
t-butylsalicydene)-1,2-cyclohexanediaminoaluminum chlo-
ride was used as received from Strem Company.
trans-Deuterio-1,2-hexene carbonate: 1H NMR (400
MHz, CDCl3/TMS): {δ 0.93 (3H, t, J = 6.6 Hz, CH3),
1.36–1.41 (4H, m, –CH2CH2–), 1.67–1.83 (2H, m, CH2),
4.05 (1H,d, J = 6.8 Hz, CHD), 4.70 (1H, dd, J = 6.4,
J = 6.4 Hz, CH)}; 13C NMR (100 MHz, CDCl3): {δ 13.95,
22.42, 26.59, 33.73, 69.24 (t, JC–D = 23.5 Hz), 77.13,
155.26}.
2.2. Catalysts preparation and characterization
The Schiff-base aluminum complexes (SalenAlX) were
prepared according to the literature methods [28,29]. These
complexes are all sensitive to air or moisture and should
be stored in a nitrogen atmosphere. (t−Bu)SalenAlEt 1b,
1H NMR (Varian INOVA-400-type spectrometer, 400 MHz,
CDCl3/TMS): δ −0.37 (q, 2H, AlCH2CH3), 0.73 (t, 3H,
AlCH2CH3), 1.29 (s, 18H, C(CH3)3), 3.65 (m, 2H, NCH2),
3.95 (m, 2H, NCH2), 6.85–7.48 (m, 6H, PhH), 8.33 (s,
2H, PhCH). (t-Bu)2SalenAlEt 1c, 1H NMR CDCl3/TMS: δ
−0.38 (q, 2H, AlCH2CH3), 0.73 (t, 3H, AlCH2CH3), 1.30
(s, 18H, C(CH3)3), 1.54 (s, 18H, C(CH3)3), 3.67 (m, 2H,
NCH2), 3.93 (m, 2H, NCH2), 6.97 (d, 2H, PhH), 7.50 (d,
2H, PhH), 8.30 (s, 2H, PhCH). SalophenAlEt 1h, 1H NMR
CDCl3/TMS: δ −0.40 (q, 2H, AlCH2CH3), 0.61 (t, 3H,
AlCH2CH3), 6.77–7.71 (m, 12H, PhH), 8.79 (s, 2H, PhCH).
The enantiomeric excesses (ee’s) of the resulting propy-
lene carbonate was determined by chiral GC analysis (GC
column, 2,6-dibutyl-3-butyryl-β-Cyclodex, 30 m×0.25 mm
id × 0.25 µm film; injecton temperature = 250 ◦C; detec-
tion temperature = 250 ◦C; 160 ◦C, isothermal, tR[(R)–PC]
=
9.37 min, tR[(S)–PC] = 9.63 min)) using a Hewlett Packard
5890 gas chromatograph with N2 as a carry gas.
3. Results and discussion
Since all SalenAl(III) complexes shown in Scheme 1 and
18-crown-6–KI complexes are easily dissolved in neat epox-
ides surveyed, the cycloaddition of CO2 to epoxides does not
require any organic cosolvent. During our investigation, we
found that the five-membered propylene carbonate could be
obtained in TOF of 57.9 h−1 from complex 1a (0.125 mol%)
and 18-crown-6–KI (0.125 mol%) cocatalyzed reactions of
2.3. Cycloaddition reactions
The reaction of CO2 and epoxides was carried out in a 50-
ml stainless-steel autoclave equipped with a magnetic stirrer.
In a typical procedure, to a Schlenk flask (50 ml) equipped
with a three-way stopcock were successively added KI

X.-B. Lu et al. / Journal of Catalysis 227 (2004) 537–541
539
Scheme 1.
CO2 with propylene oxide under extremely mild conditions
(25 ◦C, 0.6 MPa CO2) (Table 1, entry 1). No by-product such
as polycarbonates or polyether was observed in the resulting
products. It is important to note that neither 1a nor the 18-
crown-6–KI complex by itself showed any or only very low
catalytic activity under the employed conditions (Table 1,
entries 2 and 3). The coexistence of 1a and 18-crown-6–KI
complexes is essential to effectively catalyze this reaction.
Several SalenAl(III)X complexes (Scheme 1) with vary-
ing substitute groups on the aromatic rings (1a–c), or the
axial X group (1d–g), or/and diamine backbones (1h, 1i)
were investigated as catalysts for the cycloaddition of CO2
to propylene oxide (Table 1, entries 4–11). In the presence of
18-crown-6–KI, complexes 1a and 1d–g were proved to be
more effective in catalyzing this reaction, and their activities
are at least five times as active as complex 1c. We believe
this may be due to the high electrophilicity and the more ac-
cessible coordination site available for the epoxide of these
complexes. The results also indicate that the epoxide was
predominately activated by its coordination to central Al3+
ions at the opposite of axial X group of SalenAl(III)X com-
plexes. The facts that the addition of n-Bu3N to the 1a/18-
crown-6–KI catalyst system resulted in its catalytic activity
(TOF) from 57.9 h−1 rapidly decreasing to 9.6 h−1 (Table 1,
entry 19), and that the bifunctional catalyst system based on
a chiral aluminum complex 1i and 18-crown-6–KI prefer-
entially consumed (S)-propylene oxide over (R)-propylene
oxide with Krel of 1.8 (Table 1, entry 11), all support this
conclusion.
Table 1
Reaction of CO with propylene oxide in the presence of a SalenAl(III)
2
a
complex in conjunction with a polyether–KY complex
Catalyst
+ CO
→
2
−1
)
Entry
Catalyst
SalenAlX
Cocatalyst
polyether–
KY
TOF (h
On the other hand, the anion Y− of inorganic salts in
the complexes 18-crown-6–KY also significantly affected
the catalytic activity of the bifunctional catalyst system (Ta-
ble 1, entries 12–17). Essentially, only these catalyst systems
comprising I− or Br− anions show a considerably high activ-
ity, probably because these anions have a high nucleophility
and leaving ability. Although acetate and phenolic anions
have high nucleophilicity, complex 1a/18-crown-6–KOAcor
KOPh system shows very low activity, probably resulting
from the low leaving ability of these anions. It is generally
known that p-toluene sulfonate is a better leaving group,
but due to its low nucleophilicity, the 1a/18-crown-6–KOTs
system also could not effectively catalyze this reaction. Un-
fortunately, the PEG 400–KI complex in conjunction with
1d could also catalyze this reaction, but its activation is less
than one-quarter of that of 18-crown-6–KI/1d catalyst sys-
tem (Table 1, entry 18).
The catalyst system was found to be applicable to a va-
riety of terminal epoxides, quantitatively providing the cor-
responding cyclic carbonates in 100% selectivity (Table 2).
Ethylene oxide 2c was found to be the most reactive epox-
ide, while epichlorohydrin 2d exhibits relatively low ac-
tivity among the epoxides surveyed. Notably, the reaction
of CO2 with (S)-propylene oxide 2b in the presence of
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1a
1a
None
1b
1c
1d
1e
1f
18-crown-6–KI
None
57.9
0
0.8
18-crown-6–KI
18-crown-6–KI
18-crown-6–KI
18-crown-6–KI
18-crown-6–KI
18-crown-6–KI
18-crown-6–KI
18-crown-6–KI
18-crown-6–KI
18-crown-6–KBr
18-crown-6–KCl
18-crown-6–KOTs
18-crown-6–KOAc
18-crown-6–KSCN
18-crown-6–KOPh
PEG400–KI
18.4
10.1
60.8
62.3
59.6
63.9
19.3
29.7
56.7
3.0
0.2
2.0
2.4
4.1
13.2
9.6
1g
1h
1i
b
1a
1a
1a
1a
1a
1a
1d
1a
c
c
c
d
19
18-crown-6–KI/n-Bu N
3
a
−4
Reaction conditions: SalenAlX (2.4 × 10 mol), SalenAlX/
◦
polyether–KY/epoxide = 1/1/800 (molar ratio), temperature = 25 C,
t = 8 h.
b
The ee of the resulted propylene carbonate is 24.9%.
c
CH CN (5 ml) was added as cosolvent.
3
d
18-crown-6–KI/n-Bu N = 1/10 (molar ratio).
3

540
X.-B. Lu et al. / Journal of Catalysis 227 (2004) 537–541
Table 2
Very recently, Shen and co-workers reported that binaph-
thyldiamino Salen-type Zn, Cu, and Co complexes could
effectively catalyze the cycloaddition of CO2 to terminal
epoxides in the presence of various catalytic amounts of or-
ganic bases [30]. A Lewis acid- and Lewis base-cocatalyzed
mechanism is proposed, on the basis of the reaction of trans-
deuterioethene oxide with CO2. For the 1a/18-crown-6–KI
catalyst system, it is clear that the electrophilicity of Al3+ in
complex 1a and the nucleophilicity of I− in the 18-crown-6–
KI complex cooperatively catalyze the cycloaddition of CO2
to epoxides. In order to clearify the reaction mechanism, we
also synthesized trans-deuterio-1,2-hexeneoxide 2g and uti-
lized it as the substrate in the reaction with CO2 catalyzed
by the 1a/18-crown-6–KI system (Table 2). We found that
trans-deuterio-1,2-hexene carbonate 3g was formed exclu-
sively. The result suggests that the formation of cyclic car-
bonate in our reaction system proceeds via path a as shown
in Scheme 2, in which the epoxide is first activated by its co-
ordination to the central Al3+of complex 1a and then opened
by nucleophilic attack of free I− of the 18-crown-6–KI com-
plex at the less substituted C–O bond, and further reacts
with CO2 to give the corresponding deuterated cyclic car-
bonate. On the contrary, if the reaction proceeds via path b,
cis-deuterio-1,2-hexene carbonate should be the dominant
product.
The cycloaddition of CO to various epoxides catalyzed by 1a/18-crown-
6–KI system
2
a
b
Substrate
Time (h) Product
12
Yield (%)
98
2a
2b
3a
12
3b
98
2c
8
3c
96
95
2d
48
3d
2e
14
24
3e
99
97
2f
3f
Based on the facts described above, a plausible mecha-
nism for this cycloaddition is proposed as shown in Scheme 3.
We believe epoxides are ring-opened at the less substituted
C–O bond to form 3 in view of the synergistic effect of
the bifunctional catalyst, which probably results from nu-
cleophilicity of free I− of the 18-crown-6–KI complex and
the electrophilic interaction of SalenAl(III) with epoxides.
The activation of CO2 is generally initiated by nucleophilic
attack of the alcoholate [OCH(R)CH2I−K+–18-crown-6] at
the Lewis acid carbon atom of CO2, and weak interaction
2g 24
3g 95
a
−5
Reaction conditions: 1a (16.1 mg, 5 × 10 mol), 1a/18-crown-6–
KI/epoxide = 1/1/400 (molar ratio); p(CO ) = 0.60 MPa; temperature,
2
◦
25 C; the reaction was carried out in 10 ml autoclave.
b
Isolated yield.
the 1a/18-crown-6–KI catalyst system gave (S)-propylene
carbonate 3b in > 99% ee with retention of stereochem-
istry.
Scheme 2.

X.-B. Lu et al. / Journal of Catalysis 227 (2004) 537–541
541
Scheme 3.
between the central metal ion and the lone pairs of one oxy-
gen of CO2. The synergistic effect is sufficient to lead to the
formation of linear carbonate 5, which is transformed into
five-membered cyclic carbonates via intramolecular cyclic
elimination.
[9] G. Rokicki, W. Kuran, B.P. Marciniak, Monat. Chem. 115 (1984) 205.
[10] A. Baba, T. Nozaki, H. Matsuda, Bull. Chem. Soc. Jpn. 60 (1987)
1552.
[11] R. Nomura, A. Ninagawa, H. Matsuda, J. Org. Chem. 45 (1980) 3735.
[12] T. Yano, H. Matsui, T. Koike, H. Ishiguro, H. Fujihara, M. Yoshihara,
T. Maeshima, Chem. Commun. (1997) 1129.
[13] K. Yamaguchi, K. Ebitani, T. Yoshida, H. Yoshida, K. Kaneda, J. Am.
Chem. Soc. 121 (1999) 4526.
[14] M. Tu, R.J. Davis, J. Catal. 199 (2001) 85.
Acknowledgments
[15] H. Yasuda, L.N. He, T. Sakakura, J. Catal. 209 (2002) 547.
[16] T. Aida, S. Inoue, J. Am. Chem. Soc. 105 (1983) 1304.
[17] W.J. Kruper, D.V. Dellar, J. Org. Chem. 60 (1995) 725.
[18] K. Kasuga, S. Nagao, T. Fukumoto, M. Handa, Polyhedron 15 (1996)
69.
Gratitude is expressed to the Science Foundation of
Liaoning Province (Grant 20031074) and the National Sci-
ence Foundation of China (NSFC) program (Grant 20204002)
for financial support.
[19] H.S. Kim, J.J. Kim, H.N. Kwon, M.J. Chung, B.G. Lee, H.G. Jang,
J. Catal. 205 (2002) 226.
[20] R.L. Paddock, S.T. Nguyen, J. Am. Chem. Soc. 123 (2001) 11498.
[21] H. Kawanami, A. Sasaki, K. Matsui, Y. Ikushima, Chem. Commun.
(2003) 896.
References
[22] H.S. Kim, J.J. Kim, H. Kim, H.G. Jang, J. Catal. 220 (2003) 44.
[23] T. Fujinami, T. Suzuki, M. Kamiya, Chem. Lett. (1985) 199.
[24] B.M. Trost, S.R. Angle, J. Am. Chem. Soc. 107 (1985) 6123.
[25] X.B. Lu, B. Liang, Y.J. Zhang, Y.Z. Tian, Y.M. Wang, C.X. Bai, H.
Wang, R. Zhang, J. Am. Chem. Soc. 126 (2004) 3732.
[26] P.S. Skell, P.K. Freeman, J. Org. Chem. 29 (1964) 2524.
[27] A.L. Schwan, M.R. Roche, J.F. Gallagher, G. Ferguson, Can. J. Chem.
72 (1994) 312.
[1] X.L. Yin, J.R. Moss, Coord. Chem. Rev. 181 (1999) 27.
[2] D.J. Darensbourg, M.W. Holtcamp, Coord. Chem. Rev. 153 (1996)
155.
[3] J.H. Clements, Ind. Eng. Chem. Res. 42 (2003) 663.
[4] A. Barbarini, R. Maggi, A. Mazzacani, G. Mori, G. Sartori, R. Sarto-
rio, Tetrahedron Lett. 44 (2003) 2931.
[5] H. Hisch, R. Millini, I.J. Wang, Chem. Ber. 119 (1986) 1090.
[6] L.N. He, H. Yasuda, T. Sakakura, Green Chem. 5 (2003) 92.
[7] H.Z. Yang, Y.L. Gu, Y.Q. Deng, F. Shi, Chem. Commun. (2002) 274.
[8] N. Kihara, N. Hara, T. Endo, J. Org. Chem. 58 (1993) 6198.
[28] X.B. Lu, X.J. Feng, R. He, Appl. Catal. A 234 (2002) 25.
[29] A. Le Borgne, M. Moreau, V. Vincens, Macromol. Chem. Phys. 195
(1994) 375.
[30] Y.M. Shen, W.L. Duan, M. Shi, J. Org. Chem. 68 (2003) 1559.