Polymer-bound aluminium salen complex as reusable catalysts for CO 2 insertion into epoxides

Article abstract of DOI:10.1016/j.tet.2005.07.114

Two polymeric aluminium salen complexes in where the backbones are either a partially crosslinked polystyrene [(Al(salen)/PS)] or poly(ethylene glycol bismethacrylate) [(Al(salen)/PEA)] have been synthesised and used for the carbon dioxide insertion into epoxides to form cyclic carbonates. The catalytic activity of these polymers is similar to that of the unsupported aluminium salen complexes, and the polymeric catalysts can be easily separated from the reaction mixture and reusable in consecutives runs. The activity and reusability of the polymeric salen complex depends on the nature of the polymer: PEA being a polymer with a high oxygen content in the backbone enhances the initial activity as compared to PS, but Al(salen)/PEA exhibits lower stability as compared to Al(salen)/PS and a Al depletion occurs upon use. The presence of nucleophiles such as N-methylimidazole or N,N-dimethylaminopyridine in excess increases the catalytic activity of the polymeric Al(salen) catalyst. Also polymeric nucleophiles have been found to be suitable reusable co-catalysts for this reaction.

Full text of DOI:10.1016/j.tet.2005.07.114

Tetrahedron 61 (2005) 12131–12139
Polymer-bound aluminium salen complex as reusable catalysts
for CO2 insertion into epoxides
Mercedes Alvaro,a Carlos Baleizao,b Esther Carbonell,a Mostafa El Ghoul,a
a,
Hermenegildo Garcıa and Barbara Gigante
b
*
´
a
´ ´ ´ ´
Instituto Tecnologıa Quımica and Departamento de Quımica, Universidad Politecnica de Valencia, 46022 Valencia, Spain
bINETI-DTIQ, Entrada do Pac¸o do Lumiar, 22 1649-038 Lisboa, Portugal
Received 1 July 2005; revised 28 July 2005; accepted 28 July 2005
Available online 24 October 2005
Abstract—Two polymeric aluminium salen complexes in where the backbones are either a partially crosslinked polystyrene [(Al(salen)/
PS)] or poly(ethylene glycol bismethacrylate) [(Al(salen)/PEA)] have been synthesised and used for the carbon dioxide insertion into
epoxides to form cyclic carbonates. The catalytic activity of these polymers is similar to that of the unsupported aluminium salen complexes,
and the polymeric catalysts can be easily separated from the reaction mixture and reusable in consecutives runs. The activity and reusability
of the polymeric salen complex depends on the nature of the polymer: PEA being a polymer with a high oxygen content in the backbone
enhances the initial activity as compared to PS, but Al(salen)/PEA exhibits lower stability as compared to Al(salen)/PS and a Al depletion
occurs upon use. The presence of nucleophiles such as N-methylimidazole or N,N-dimethylaminopyridine in excess increases the catalytic
activity of the polymeric Al(salen) catalyst. Also polymeric nucleophiles have been found to be suitable reusable co-catalysts for
this reaction.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
precautions to avoid the presence of moisture and oxygen.
Later other reports have also described the catalytic activity
of other metal salen complexes and porphyrins.31–39
The increase in the atmospheric concentration of carbon
dioxide is playing a negative effect in the climatic change
due to the green house effect and there are many
international programs aimed at balancing CO2 emission
and absorption.1–10 The availability of CO2 and the interest
in using it as feedstock has motivated a large effort in
developing industrial processes based on the use of CO2 as
C1 raw material.11–20 Among the possible processes to use
CO2 as reagent, the synthesis of carbonates and poly-
carbonates is very appealing, due to the properties of dialkyl
carbonates as solvents and also as synthetic intermediates
for urethanes, ureas and isocyanates.20–28 On the other hand,
polycarbonates have excellent properties as engineering
plastics.29,30 For this reason the report by Nguyen that
chromium salen complex is a suitable catalysts for the CO2
insertion into epoxides to form cyclic carbonates has
constituted a significant breakthrough in this area.31 The
main reason for this is the ease of the synthesis of metal
salen catalysts and that their use does not require especial
A general trend in catalysis is to transform a successful
homogeneous catalytic process into a heterogeneous
one.40,41 Heterogeneous catalysts can be easily separated
from the reaction products and reused in successive runs,
provided that they do not become deactivated in the
reaction. Also, heterogeneous catalysis is more suitable
for continuous flow operation that is more appropriated than
batchwise processes for large scale industrial synthesis. On
the other hand, from the green chemistry point of view the
use of a non-recoverable catalyst, particularly considering
the toxicity of some transition metals such as chromium,
must be avoided. Recently, we have reported the catalytic
activity of chromium salen complexes anchored on silicas as
solid catalyst for the CO2 insertion into epoxides.42
However, given the negative environmental impact of
chromium metal, and in order to improve the ‘greenness’
of the metal Schiff base complex catalysts for the carbon
dioxide fixation process, we wanted to replace chromium by
aluminium as the Lewis metal atom of the complex.
Precedents in the chemical literature have already shown
that some aluminium complexes are also active for carbon
dioxide insertion.43–46 As a continuation of this research
Keywords: Heterogenoeus catalysis; CO2 insertion; Cyclic carbonates;
Polyethylenglycol as solvent.
*
Corresponding author. Tel.: C34 96387 7809; fax: C34 96387 7807;
e-mail: hgarcia@qim.upv.es
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.07.114

12132
M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
line, herein we describe the synthesis and catalytic activity
of aluminium salen complexes covalently attached to two
different polymeric backbones. We have found that these
polymeric catalysts are also active to effect the CO2
coupling to epoxides, and that the nature of polymer
skeleton plays a remarkable effect in the control of the
catalytic activity of the complex.
covalently attached to the polystyrene polymer. The
polymer-bound ligand was obtained in three steps from
2,6-diformil-4-tert-butylphenol. The last step was metal
complexation by reacting diethylaluminium chloride with
the polymer-bound ligand under inert atmosphere in the
absence of humidity. In this way, Al(salen) complex was
attached to the polymer backbone through imino groups.
Scheme 1 shows the route followed to prepare the
6-methimine substituted Al(salen) complex bonded to
polystyrene.
Suitable polymer-bound catalyst can be recovered and
reused for several runs. In addition, to make the process
completely clean, we have used recoverable polymeric
nucleophile co-catalysts instead of dissolved nucleophilic
molecules. Compared to other strategies in where the
substrates or reagents are supported on a polymer, the
approach of supporting catalyst has several advantages
including: (i) the fact that no high loadings on the polymeric
support are required; (ii) the amplification effect since few
supported catalytic sites can effect the transformation of
considerable amounts of substrate; (iii) and the higher
added-value of catalysts as compared to substrates that
makes advisable their recovery and reuse.
Working with CO2 one important parameter to favour its
reactivity is to use a medium that is able to dissolve a
sufficiently high concentration of CO2. Thus, we anticipated
based on previous literature reports that the affinity for CO2
of polystyrene-derived polymers may be hot as high as
desirable, and therefore, the CO2 insertion reaction should
occur mostly on the interface between the solid polymeric
catalyst and the fluid phase.47–50 In this context we wanted
to compare the activity of polystyrene anchored Al(salen)
complexes with that of an analogous Al(salen) complex
anchored on a different backbone that could show different
affinity for CO2. In this regard it is known that polymers
containing ethylene dioxy groups can be suitable to dissolve
significantly high CO2 concentrations due to a high oxygen
density in the polymer backbone.51–56
2. Results and discussion
2.1. Polymeric catalyst preparation
For the present work we have prepared two different
polymers containing covalently anchored Al(salen)
complex. One of them in where the polymer backbone
was a partially crosslinked poly(styrene-co-p-divinylben-
zene) having some pendant amino methyl groups. These
lateral groups were used to construct a salen ligand
Aimed at the purpose of having an Al(salen) complex
anchored on a polymer with ethylenedioxy units, we
designed a synthetic strategy completely different to that
shown in Scheme 1. For having a polymer containing
ethylendioxy units our approach was not to use a preformed
Scheme 1. Synthetic route for the preparation of Al(salen)/PS.

M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
12133
Scheme 2. Synthetic route for the preparation of Al(salen)/PEA.
polymer, but to perform a copolymerization with a salen
ligand containing reactive styryl units. Thus, in this case, the
key salen derivative is one containing p-styryl units in
peripheral positions of the salen ligand. The corresponding
compound was obtained from 4-styryl-6-tert-butylsalicyl-
aldehyde synthesised through a Pd catalyzed Suzuki cross
coupling between p-styrylboronic acid and the para
brominated salen ligand, as it has been reported pre-
viously.57 With the p-styryl substituted salen ligand we
proceeded to perform a copolymerisation with ethylene
bismethacrylate using AIBN as radical initiator and an inert
atmosphere. The actual synthetic route is illustrated in
Scheme 2. The last step in the synthesis was again the
formation of the aluminium complex by reacting in
dichloromethane an organometallic diethylaluminium
chloride as metal source compound. All the synthetic
intermediates required in the synthesis indicated in
Scheme 2 were characterized by analytical and spectro-
scopic methods.
1.26 and 0.42 mmol!gK1, respectively. IR spectra of the
polymers were in agreement with the simultaneous presence
of Al(salen) complex and the organic polymer. Figure 1
shows the IR spectra of Al(salen)/PS and Al(salen)/PEA
polymers.
Polymers Al(salen)/PS and Al(salen)/PEA were character-
izated by analytical and spectroscopic techniques. The
aluminium content of Al(salen)/PS and Al(salen)/PEA were
Figure 1. IR spectra of Al(salen) (a); Al(salen)/PS (b); Al(salen)/PEA
(c) recorded at room temperature in KBr disks.

12134
M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
Scheme 3. Procedure for the preparation of a PEG solid phase containing Al(salen) and cesium carbonate.
Finally, we included in our study a different recoverable
catalytic system consisting in a CO2-phylic PEG polymer
containing dissolved the Al(salen) complex and Cs2CO3 as
base (Scheme 3). Since PEG (average MW 6000 Da) is a
solid at room temperature, but melted at 40 8C, we have
taken advantage of it and dissolved the Al(salen) and
Cs2CO3 on melt PEG. After dissolving in PEG Al(salen)
and Cs2CO3 remain on the PEG phase during the recovery
of the reaction products.
were flawed by the poor recovery of the reaction mixture
upon reactor discharge. Working under supercritical
conditions in a 125 mL autoclave with 4 g of styrene
oxide the mass balances of the reaction were almost
complete (see Table 1). In addition, we purposely stopped
the reactions at low conversion in order to obtain a valid
information of the relative catalytic activity of the different
polymer-supported aluminium complexes catalysts. It is
obvious that at sufficiently long times for which complete
conversions are achieved, the information about the
distinctive performance and activity of the catalyst will
be lost.
2.2. Catalytic activity
As a model reaction to test the catalytic activity of the two
polymeric Al(salen) complexes, we selected the CO2
insertion into styrene oxide (Scheme 4). The series of
reactions were carried out in a mechanically stirred
autoclave working in supercritical CO2 at 80 8C and
100 bar (CO2 critical point: TcZ31 8C; PcZ74 bar).
Through the series of reactions we varied the amount of
nucleophile co-catalyst to determine the optimum Al(salen)/
nucleophile molar ratio. It has been demonstrated that the
presence of N-methylimidazole and other bases in adequate
metal-to-base molar ratios can enhance considerably the
catalytic activity, the presence of base playing a detrimental
effect if a large excess is used.35,37 For the sake of
comparison we have included also in our study a non-
polymeric Al(salen) complex. The list of reactions
performed and the results obtained are summarized in
Table 1.
The reaction mixtures were analysed by GC and IR, and
the amount of cyclic carbonate quantified by GC using the
external standard method. In most of the experiments, the
only product observed was the corresponding five member
ring cyclic carbonate, but in some cases the corresponding
glycol and polycarbonate were also observed (Scheme 4).
Phenyl ethylene glycol comes from the hydrolytic epoxide
aperture catalyzed by acids.
In the literature it has been shown that epoxides of
cycloalkenes give exclusively or predominantly polycarbo-
nates while epoxides of acyclic alkenes lead predominantly
cyclic carbonates.38 Therefore, in our case we expected to
obtain mostly the cyclic carbonate. Nevertheless, we
assessed in each reaction mixture the presence or absence
of polycarbonate by IR spectroscopy. While pentacyclic
carbonate exhibits a carbonyl stretching vibration band
1820 cmK1, the corresponding carbonyl in the polymeric
carbonate band appears at 1750 cmK1. Figure 2 shows two
Some preliminary experiments were carried out using lesser
amounts of styrene oxide (400 mg) but these experiments
Scheme 4. Reaction products observed for the catalytic CO2 coupling with styrene oxide.

M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
12135
Table 1. Results obtained at 6 h reaction time for the CO2 insertion into styrene oxide (4 mL) catalyzed by Al(salen) complexes and nucleophile (3.6% mol) at
80 8C and 100 bar in a mechanically stirred (300 rpm) autoclave (125 mL)
Run
Catalyst [catalyst-to-substrate mol %]
Mass balance (%)
Conversiona (%)
Yield (%)
TON
1
2
3
4
Al(salen) [0.1]
Al(salen)b [0.1]
Al(salen)c[0.1]
Al(salen)d [0.1]
91
100
98
19
20
4
17
17
3
59
59
10
10
97
4
3
5
6
7
8
9
Al(salen)/PEA (use 1) [0.14]
Al(salen)/PEA (use 2) [0.14]
Al(salen)/PEA (use 3) [0.14]
Al(salen)/PEAe [0.14]
87
95
92
98
95
22
9
9
8
89
19
6
6
8
78
47
15
15
20
39
Al(salen)/PEAf [2]
10
11
Al(salen)/PS (use 1) [0.44]
Al(salen)/PS (use 3) [0.44]
[PEG-Al-CsCO3]g [0.1]
100
97
86f
9
9
9
9
7
7
12
15
11
38
a Based on the disappearance of styrene.
b Dimethylaminopyridine was used instead of N-methylimidazole.
c PZ20 bar.
d 2 equiv of N-methylimidazole
e Dimethylaminopyridine polymer-bound was used as co-catalyst.
f
Reaction time 15 h.
g 1% of polymer it was observed as by-product.
representative IR spectra of the reaction mixture to illustrate
how the presence or absence of polycarbonate can be
determined by this technique.
and used as a tool to control some properties of supercritical
CO2 phase.58–60
With the aim of having a reusable and recoverable
polymeric catalyst we proceeded to perform the same
reaction under supercritical CO2 conditions using Al(salen)/
PS and Al(salen)/PEA solids. The results are also contained
in Table 1.
Before performing experiments with polymeric salen
catalyst, we carried out a preliminary study on the activity
of Al(salen) complex not covalently bonded to any support
(0.1% mol) varying the excess of N-methylimidazole as co-
catalyst. The results are contained in Table 1 and it can be
seen that the presence of bases in a large excess plays a large
beneficial influence on the conversion. When no supercriti-
cal conditions were used (Table 1, entry 3) the conversion
reached was considerably lower. The positive influence of
the supercritical conditions can be interpreted considering
that when no supercritical conditions are used, there should
be two phases, one of them being liquid styrene oxide
saturated in CO2, but most of the CO2 being in a separate gas
phase. In contrast under supercritical conditions a single
phase should be ideally present inside the reactor in where
supercritical CO2 dissolves styrene oxide. The ability of
supercritical CO2 to mix with hydrocarbons is well known
Noteworthy was that the initial activity of fresh Al(salen)/
PEA was significantly higher than that of the analogous
Al(salen)/PS, although the turnover number for unsupported
Al(salen) was still higher than those obtained using
polymeric Al(salen)/PEA. These results can be easily
interpreted considering that the PEA backbone exhibits a
high oxygen density, in contrast to the PS skeleton that is
only a hydrocarbon. Also noteworthy was that the higher
activity of Al(salen)/PEA results in a somewhat lower
selectivity towards the cyclic carbonate since some
polycarbonate is formed in this case. The polymeric
carbonate becomes deposited on the catalyst. The presence
of polycarbonate on Al(salen)/PEA results in a change of
the texture of the initially soft and sticky Al(salen)/PEA
powder that becomes much harder. The presence of
polycarbonate can be spectroscopically determined by
observing in the IR spectrum of the used Al(salen)/PEA
catalyst the presence of a band 1750 cmK1 corresponding to
the stretching vibration to the carbonate group to the
polymer (see Fig. 2). Also, optical microscopy (Fig. 3)
reveals a variation on the Al(salen)/PEA particle mor-
phology with different size and optical properties that those
the fresh Al(salen)/PEA catalyst. In contrast to the
behaviour of Al(salen)/PEA, Al(salen)/PS does no form
detectable amounts of polycarbonate as evidenced by IR
spectroscopy (absence of 1750 cmK1 peak, see Fig. 2).
However, optical microscopy reveals that the initial
spherical shape of the PS beads also disappears upon
catalyst use (Fig. 4). Although polymer attrition can not be
disregarded, the mild mechanical stirring used to perform
the reactions makes more probable that changes in the
polymer morphology are due to the influence of the reaction
and interaction with CO2.
Figure 2. IR spectra of two reaction mixtures obtained after the treatment
of styrene epoxide (35 mmol) with CO2 in autoclave (125 mL) at 80 8C and
100 bar using Al(salen)/PEA (a) or Al(salen)/PS (b) as catalysts. The peaks
at 1750 cmK1 indicates the presence of polycarbonate, while the peak at
1815 cmK1 corresponds to cyclic carbonate.

12136
M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
Figure 3. Optical microscopy of Al(salen)/PEA powder before (left) and after(right) being used as catalyst for the CO2 insertion into styrene oxide. Particle
size and texture of Al(salen)/PEA changes upon its use as catalyst.
Figure 4. Optical microscopy of Al(salen)/PS powder before (left) and after (right) being used as catalyst for the CO2 insertion into styrene oxide. The typical
spherical shape of the Al(salen)/PS beads disappears upon catalyst use.
Concerning the reaction conversion, when higher amounts
of Al(salen)/PEA catalyst is present and the reaction time is
longer (Table 1, entry 9), then, obviously the yield of the
cyclic carbonate increases, although the turnover number
decreases somewhat with respect to those experiments using
higher substrate-to catalyst molar ratio.
chromium (Scheme 5). If this were so, in our case using
polymer-bound Al(salen) complexes, addition of the
supported base co-catalyst should inhibit dramatically its
co-catalyst role by impeding the interaction between the
polymer bound nucleophile and the polymer bound metal
salen complex that will be in different solid phases.
According to this, the expected yield of the CO2 coupling
using a mixture of polymeric Al(salen) and polymeric
nucleophile should be much lower, since the promoting
effect the nucleophile will be lost.
According to the proposed reaction mechanism for the
N-methylimidazole enhancement of the catalytic activity
of Cr(salen) complexes for the CO2 insertion into
epoxides,35–37,61 the key intermediate consists in a distorted
octahedrally coordinated chromium ion interacting at the
apical positions with the nucleophile and the epoxide. In
other words, the role of N-methylimidazole has been
proposed to be the softening of the chromium-epoxide
interaction by N-methylimidazole coordination to the
However, as it can be seen in Table 1, this was not exactly
the case and a mixture of polymeric dimethylaminopyridine
and polymeric aluminium salen complex gives also rise to
the formation of cyclic carbonate with only half the turnover
activity than those obtained using molecular nucleophiles
(compare entries 5 and 8). Polymeric dialkylaminopyridine
(see structure and preparation procedure in Scheme 6) was
prepared as previously reported in the literature.62 A blank
control in where the same experiment using Al(salen)/PEA
as catalyst was used in the absence of any nucleophile
(either molecular or polymeric) under the same conditions
gave essentially no epoxide conversion.
This result using polymeric dialkylaminopyridine seems to
indicate that the reaction mechanism should be probably
revisited to take in consideration that the formation of a
single species between the nucleophile and the metal
complex as shown in Scheme 5 may not be absolutely
necessary. From the practical point of view, the use of both
Scheme 5. Proposed structure of activated epoxide in where the nucleophile
co-catalyst is simultaneously bonded to the epoxide-chromium(salen)
intermediate.

M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
12137
Scheme 6. Route for the preparation of the polystyrene containing pendant dialkylaminopyridine units.
polymeric nucleophile and polymeric metal complex has
even more advantages in terms complete reusability of the
catalytic system and avoidance of waste.
complex IR bands at 1640 cmK1 is significantly dismin-
ished after the first use of the polymeric Al(salen)/PEA
catalyst.
A different reusable system that was also tested for this
reaction was a mixture of Al(salen) and Cs2CO3 dissolved in
polyethyleneglycol (PEG) (Table 1, entry 12). PEG (av.
MW 600 Da) becomes liquid at about 40 8C and is a suitable
solvent for CO2 reactions due to the high CO2 solubility in
PEG. After the reaction and upon cooling, the mixture
becomes solid and the catalyst and base remains in the solid
PEG phase, while the cyclic carbonate can be extracted with
diethyl ether or methanol. Cs2CO3 is a suitable base in PEG
media since large alkali metal ions become solvated through
interaction with ethylendioxy oxygens, increasing the
solubility and boosting the nucleophilic strength of the
base. As it can be seen in the Table 1, also this system based
with PEG in where there is no covalent linkage between the
polymer acting as solvent and the complex could be of
interest for performing the CO2 insertion into epoxides,
while still permitting reusability of the metal complex and
nucleophile.
3. Conclusions
Aluminium salen complex anchored to polymeric back-
bones is an active catalyst for the CO2 insertion into
epoxide. However, as anticipated, polymeric catalysts
exhibit somewhat lower turnover numbers than that of the
non-polymeric complex. The activity of fresh polymeric
catalyst depends on the composition of the backbone and its
affinity for CO2. PEA is more suitable in this regard than
partially crosslinked PS. However, stability and reusability
on the catalyst also depends on the nature of the polymer
and from this point of view Al(salen) bonded to PS appears
as a more stable catalyst than when bonded to PEA, in where
a susbtancial diminution in the number of Al(salen)
complexes is observed after the first use. The use of PEG
as co-solvent and cesium carbonate as a base is also an
interesting medium that permits the reusability of the
system. Finally also the base can be anchored in a polymeric
skeleton without producing a detrimental effect of the
catalytic activity of the system. This is in contrast of could
have been expected in view of the accepted reaction
mechanism and the role of nucleophiles as catalyst
promoter. Given the interest of this reaction and the
possibility to perform enantioselective CO2 insertions,
current work is aimed at expanding the process using
other epoxides as starting materials and to determine the
enantiomeric excess achieved for asymmetric epoxides.
In fact, reusability of polymeric catalyst was performed by
simple filtration of the solid from the reaction mixture and
washing it with dimethylcarbonate and ethanol. After
drying, the solids were used as catalysts for a new batch
under the same conditions. The results obtained are also
shown in Table 1 (entries 6, 7 and 11). It can be seen there
that while the activity of fresh Al(salen)/PS is lower than
that of Al(salen)/PEA, upon reuse the activity of Al(salen)/
PS remains as in the first use. In contrast, the activity of
Al(salen)/PEA undergoes a significant decrease between the
first and second runs, although then the activity is
maintained going from the second to the third use and is
still higher than that of Al(salen)/PS. These results can be
interpreted considering that miscibility CO2 in the poly-
meric PEA skeleton is high, this leading to partial depletion
of the Al(salen) complex, most probably by extracting the
aluminium cation producing the leaching of Al from the
polymer to the liquid phase. In support of this, IR
spectroscopy reveals that the characteristic Al(salen)
4. Experimental
4.1. General
IR spectra of polymer bound Al(salen) were recorded on a
Nicolet 710 FT-IR spectophotometer using KBr disks. The
polymer (10 mg) was mixed in a mortar with dry KBr
(20 mg) and the mixture pressed at 1 ton!cmK2 for 2 min.

12138
M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
Optical microscopy of the polymeric catalysts were
obtained with a Leica DFC 300FX microscope placing the
powder between slides. Images were saved and manipulated
using IMSO program package.
under vacuum. The content of aluminium was determined
by absorption spectroscopy (0.42 mmol Al/g) correspond-
ing to about 95% of the maximum loading capacity
according to the ligand and the N content (0.88 mmol/g).
4.2. Preparation of Al(salen)/PS catalyst
4.4. Preparation of [PEG-Al-CsO3]
2,6-Diformyl-4-tert-butylphenol was prepared from 4-tert-
butylphenol as reported in the literature.63 Al(salen)/PS was
prepared using as support (0.6 g) of commercial amino-
methylated poly(styrene-co-divinylbenzene) (2% wt of
divinylbenzene) that was added to a solution of 2,6-
diformyl-4-tert-butylphenol (216 mg, 1.05 mmol) in etha-
nol (10 mL). Polystyrene co-DVD (2%) aminomethylated
(1.6 mmol N/g) was purchased from Aldrich.
PEG (250 mg) and Cs2CO3 (41 mg) were dissolved in water
(2 mL) and stirred for 15 min. After this time, the water is
removed under vacuum. Independently, PEG (250 mg) and
Al(salen) complex (2.6 mg) were dissolved in dichloro-
methane (2 mL), and the dichloromethane is removed under
vacuum. The two PEG solids, one containing the base and
the other Al(salen) complex, were mixed mechanically and
melted at 40 8C. The resulting solid was dried under vacuum
for 6 h.
The mixture of polystyrene and dialdehyde was magneti-
cally stirred for 2 h at reflux temperature. The solid was
removed by filtration, Soxhlet extracted with dichloro-
methane for 24 h and dried under vacuum. In the second
step, the previous functionalised solid and (R,R)-1,2-
diaminocyclohexane (120 mg, 1.05 mmol) were added to
ethanol (10 mL), stirred magnetically while refluxing for
2 h. The solid was filtered, Soxhlet extracted with
dichloromethane for 24 h, and dried under vacuum. The
solid obtained in the previous step having cyclohexadiamine
units was added to a solution of 3,5-tert-butylsalicylalde-
hyde (246 mg, 1.05 mmol) in ethanol (10 mL), and the
mixture stirred at reflux temperature. After 2 h, the solid was
removed by filtration, Soxhlet extracted with dichloro-
methane for 24 h and dried under vacuum. The synthesis of
Al(salen)/PS was accomplished at room temperature under
nitrogen atmosphere, charging in a flamed flask Al(salen)/
PS and dry dichloromethane (10 mL). Diethyl aluminium
chloride (1 mL, 1.8 mmol, 1.8 M solution in toluene) was
added slowly to the stirred solution. After 4 h, the solid was
removed by filtration, Soxhlet extracted with dichloro-
methane for 24 h and dried under vacuum. The content of
aluminium was 1.26 mmol Al/g as determined by absorp-
tion spectroscopy corresponding to 77% of the maximum
polymer loading according to the density of aminomethyl
groups in PS.
4.5. Preparation of DMAP-polymer
4-[N-methyl-N-(p-vinylbenzyl)amino]pyridine was pre-
pared starting from 4-(methylamino)pyridine sodium salt
and p-(chloromethyl)styrene. This N,N-dialkylaminopyridine
derivative was copolymerised with styrene (1 equiv) and
divinylbenzene (0.02 equiv) to give the polymer-bound
4-(N-benzyl-N-methylamino)pyridine. A more detailed
experimental procedure can be found elsewhere.62
4.6. Reaction procedure
Styrene oxide (4 mL) and the appropriate amount of
Al(salen) catalyst and co-catalyst is placed in a 125 mL
stainless steel autoclave that is subsequently charged with
liquid CO2 until an approximate pressure of 75 bars is
reached. Then, the temperature is increased at 80 8C while
the mixture is mechanically stirred. The approximate
pressure is 100 bar. The reaction mixture was maintained
under these conditions for 6 h. After this time the system
was allowed to cool at ambient temperature and, then, the
reactor valve is opened and the reactor pressure smoothly
decreased over a period of 12 h. The reaction mixture was
extracted with diethyl ether, the polymer washed with
ethanol and the product analysed by IR of the mixture (to
assess the absence of polycarbonate) and GC or GC/MS.
The products were identified by comparison with authentic
samples.
4.3. Preparation of Al(salen)/PEA catalyst
Divinyl salen ligand was prepared as reported by Seebach
and co-workers.57 A mixture of divinyl salen ligand (0.70 g,
1.10 mmol) and ethylene glycol bis(methacrylate)
(2.07 mL, 10.96 mmol) was dissolved in toluene (7 mL)
and ethanol (3 mL). The polymerisation mixture was
bubbled with nitrogen for 30 min. 2,20-Azobisisobutyro-
nitrile (25 mg) was added, and the mixture was heated at
80 8C for 24 h in nitrogen atmosphere. The polymer
(Alsalen/PEA) was removed by filtration, Soxhlet extracted
with dichloromethane for 24 h and dried under vacuum.
Acknowledgements
Financial support by the Spanish DGIC (Project MAT2003-
1226) is gratefully acknowledged. E.C. thanks the
´
Universidad Politecnica de Valencia for a scholarship.
A flamed flask was charged with salen/PEA (1.2 g) and dry
dichloromethane (12 mL) and the solution purged with
nitrogen atmosphere at room temperature for 1 h. Then,
300 mL of diethyl aluminium chloride (1.8 M solution in
toluene, 0.52 mmol) was added slowly to the stirred
solution. After 4 h, the solid was removed by filtration,
Soxhlet extracted with dichloromethane for 24 h and dried
References and notes
1. Weimer, T. Greenhouse Gas Control Technologies, Proceed-
ings of the International Conference on Greenhouse Gas
Control Technologies, 4th, Interlaken, Switz., Aug. 30–Sept. 2,
1998 1999, 403–408.

M. Alvaro et al. / Tetrahedron 61 (2005) 12131–12139
12139
2. Poliakoff, M.; Fitzpatrick, J. M.; Farren, T. R.; Anastas, P. T.
Science 2002, 298, 542.
36. Darensbourg, D. J.; Yarbrough, J. C.; Ortiz, C.; Fang, C. C.
J. Am. Chem. Soc. 2003, 125, 7586–7591.
3. Poliakoff, M.; Fitzpatrick, J. M.; Farren, T. R.; Anastas, P. T.
Science 2002, 297, 807–810.
37. Darensbourg, D. J.; Rodgers, J. L.; Mackiewicz, R. M.; Phelps,
A. L. Catal. Today 2004, 98, 485–492.
4. Anastas, P. T.; Kirchhoff, M. M. Acc. Chem. Res. 2002, 35,
686–694.
38. Darensbourg, D. J.; Mackiewicz, R. M.; Rodgers, J. L.; Fang,
C. C.; Billodeaux, D. R.; Reibenspies, J. H. Inorg. Chem. 2004,
43, 6024–6034.
5. Poliakoff, M.; Anastas, P. Nature 2001, 413, 257.
6. Hjeresen, D. L.; Anastas, P.; Ware, S.; Kirchhoff, M. Environ.
Sci. Technol. 2001, 35, 114A–119A.
39. Yarbrough, J. C. Kinetic and Mechanistic Investigations of the
Copolymerization Reaction of Carbon Dioxide and Epoxides,
using Chromium Chloride Salen Catalysts, Texas A and M
Univ., College Station, TX, USA. FIELD URL:, 2003.
40. Corma, A.; Garcia, H. Chem. Rev. 2003, 103, 4307–4365.
7. Aresta, M. ACS Symp. Ser. 2003, 852, 2–39.
8. Aresta, M.; Dibenedetto, A. ACS Symp. Ser. 2002, 809, 54–70.
9. Aresta, M. Quimica Nova 1999, 22, 269–272.
10. Aresta, M. ASI NATO Ser. C: Math. Phys. Sci. 1987, 206,
1–22.
´
41. Corma, A.; Garcıa, H. Chem. Rev. 2002, 102, 3837–3892.
42. Alvaro, M.; Baleizao, C.; Das, D.; Carbonell, E.; Garcia, H.
J. Catal. 2004, 228, 254–258.
11. Aresta, M.; Dibenedetto, A. Catal. Today 2004, 98, 455–462.
12. Yoshida, M.; Ihara, M. Chem. Eur. J., 2004, 10, 2886–2893.
13. Grunwaldt, J.-D.; Wandeler, R.; Baiker, A. Catal. Rev. 2003,
45, 1–96.
43. Darensbourg, D. J.; Billodeaux, D. R. Inorg. Chem. 2005, 44,
1433–1442.
44. Chen, P.; Chisholm, M. H.; Gallucci, J. C.; Zhang, X.; Zhou, Z.
Inorg. Chem. 2005, 44, 2588–2595.
14. Cutler, A. R.; Hanna, P. K.; Vites, J. C. Chem. Rev. 1988, 88,
1363–1403.
45. Lu, X.-B.; Feng, X.-J.; He, R. Appl. Catal. A 2002, 234, 25–33.
46. Lu, X.-B.; He, R.; Bai, C.-X. J. Mol. Catal. A 2002, 186, 1–11.
47. Han, X.; Baxter, A. R.; Koelling, K. W.; Tomasko, D. L.; Lee,
L. J. Annual Technical Conference–Society of Plastics
Engineers 2002, 60th, 1910–1914.
15. Cheng, M.; Lobkovsky, E. B.; Coates, G. W. J. Am. Chem.
Soc. 1998, 120, 11018–11019.
16. Ricci, M. Carbon Dioxide Recovery Utilization 2003,
395–402.
17. Lemonidou, A. A.; Valla, J.; Vasalos, I. A. Carbon Dioxide
Recovery Utilization 2003, 379–394.
48. Lalanne, P.; Rey, S.; Cansell, F.; Tassaing, T.; Besnard, M.
J. Supercrit. Fluids 2001, 19, 199–207.
18. Aresta, M. Carbon Dioxide Recovery Utilization 2003,
293–312.
49. Lee, M.; Park, C. B.; Tzoganakis, C. Annual Technical
Conference—Society of Plastics Engineers 1998, 56th
1902–1906.
19. Aresta, M.; Dibenedetto, A. Carbon Dioxide Recovery
Utilization 2003, 211–260.
50. Lee, M.; Park, C. B., Tzoganakis, C. Annual Technical
Conference—Society of Plastics Engineers 1997, 55th,
1991–1995.
20. Aresta, M.; Quaranta, E. Chemtech 1997, 27, 32–40.
21. Aresta, M.; Dibenedetto, A.; Tommasi, I. Appl. Organomet.
Chem. 2000, 14, 799–802.
51. DeSimone, J. M. Abstracts of Papers, 223rd ACS National
Meeting, Orlando, FL, United States, April 7–11, 2002 2002,
IEC-313.
22. Selva, M.; Tundo, P.; Foccardi, T. J. Org. Chem. 2005, 70,
2476–2485.
23. Selva, M.; Tundo, P.; Perosa, A.; Dall’Acqua, F. J. Org. Chem.
2005, 70, 2771–2777.
52. Powell, K. R.; McCleskey, T. M.; Tumas, W.; DeSimone,
J. M. Ind. Eng. Chem. Res. 2001, 40, 1301–1305.
53. DeSimone, J. M. Abstracts of Papers, 220th ACS National
Meeting, Washington, DC, United States, August 20–24, 2000
2000, PHYS-069.
24. Tundo, P.; Rossi, L.; Loris, A. J. Org. Chem. 2005, 70,
2219–2224.
25. Selva, M.; Tundo, P.; Perosa, A. J. Org. Chem. 2003, 68,
7374–7378.
54. Mishima, K.; Matsuyama, K.; Nagatani, M. Fluid Phase
Equilib. 1999, 161, 315–324.
26. Tundo, P.; Selva, M. Acc. Chem. Res. 2002, 35, 706–716.
27. Tundo, P.; Perosa, A. Chem. Rec. 2002, 2, 13–23.
28. Lu, X.-B.; Xiu, J.-H.; He, R.; Jin, K.; Luo, L.-M.; Feng, X.-J.
Appl. Catal. A 2004, 275, 73–78.
55. West, K. N.; Hallett, J. P.; Jones, R. S.; Bush, D.; Liotta, C. L.;
Eckert, C. A. Ind. Eng. Chem. Res. 2004, 43, 4827–4832.
56. Fink, R.; Hancu, D.; Valentine, R.; Beckman, E. J. J. Phys.
Chem. B 1999, 103, 6441–6444.
29. Ramasri, M.; Shit, S. C.; Mathur, A. B.; Ramamurthy, K. Pop.
Plast. Packag. 1994, 39, 41–46.
57. Sellner, H.; Karjalainen, J. K.; Seebach, D. Chem. Eur. J.
2001, 7, 2873–2887.
30. Spencer, F. J. Hydrocarbon Process. 1969, 48, 123–130.
31. Paddock, R. L.; Nguyen, S. T. J. Am. Chem. Soc. 2001, 123,
11498–11499.
58. Stahl, C. D. Producers Monthly 1964, 22, 17–19.
59. Holm, L. W.; Josendal, V. A. Soc. Petrol. Eng. J. 1982, 22,
87–98.
32. Paddock, R. L.; Hiyama, Y.; McKay, J. M.; Nguyen, S. T.
Tetrahedron Lett. 2004, 45, 2023–2026.
60. Metcalfe, R. S. Soc. Petrol. Eng. J. 1982, 22, 219–225.
61. Darensbourg, D. J.; Mackiewicz, R. M.; Billodeaux, D. R.
Organometallics 2005, 24, 144–148.
33. Jing, H.; Edulji, S. K.; Gibbs, J. M.; Stern, C. L.; Zhou, H.;
Nguyen, S. T. Inorg. Chem. 2004, 43, 4315–4327.
34. Paddock, R. L.; Nguyen, S. T. Chem. Commun. 2004,
1622–1623.
62. Tomoi, M.; Akada, Y.; Kakiuchi, H. Makomol. Chem., Rapid
Commun. 1982, 3, 537–542.
35. Darensbourg, D. J.; Yarbrough, J. C. J. Am. Chem. Soc. 2002,
124, 6335–6342.
63. Chang, H. R.; Larson, S. K.; Boyd, P. D. W.; Pierpont, C. G.;
Hendrickson, D. N. J. Am. Chem. Soc. 1988, 110, 4565–4576.