New iron pyridylamino-bis(Phenolate) catalyst for converting CO2 into cyclic carbonates and cross-linked polycarbonates

Article abstract of DOI:10.1002/cssc.201403323

The atom-efficient reaction of CO₂ with a variety of epoxides has been efficiently achieved employing iron pyridylamino-bis(phenolate) complexes as bifunctional catalysts. The addition of a Lewis base co-catalyst allowed significant reduction in the amount of iron complex needed to achieve high epoxide conversions. The possibility of controlling the selectivity of the reaction towards either cyclic carbonate or polycarbonate was evaluated. An efficient switch in selectivity could be achieved when cyclic epoxides such as cyclohexene oxide and the seldom explored 1,2-epoxy-4-vinylcyclohexane were used as substrates. The obtained poly(vinylcyclohexene carbonate) presents pending vinyl groups, which allowed post-synthetic cross-linking by reaction with 1,3-propanedithiol. The cross-linked polycarbonate displayed a substantial increase in the glass transition temperature and chemical resistance, thus opening new opportunities for the application of these green polymers. CO₂ meets epoxides: Iron pyridylamino-bis(phenolate) complexes are highly active catalysts for the atom-efficient reaction of CO₂ with a variety of epoxides. The selectivity can be switched between the cyclic or polymeric carbonate when using cyclic epoxides such as cyclohexene oxide and 1,2-epoxy-4-vinylcyclohexaneas substrates. Cross-linking of the obtained poly(vinylcyclohexene carbonate) leads to substantial increase in the T_g and chemical resistance of the polymer.

Full text of DOI:10.1002/cssc.201403323

DOI: 10.1002/cssc.201403323
Full Papers
New Iron Pyridylamino-Bis(Phenolate) Catalyst for
Converting CO2 into Cyclic Carbonates and Cross-Linked
Polycarbonates
Masoumeh Taherimehr,[a] Jo¼o Paulo Cardoso Costa Sert¼,[a] Arjan W. Kleij,[b, c]
Christopher J. Whiteoak,[b] and Paolo P. Pescarmona*[a, d]
The atom-efficient reaction of CO2 with a variety of epoxides
has been efficiently achieved employing iron pyridylamino-bi-
s(phenolate) complexes as bifunctional catalysts. The addition
of a Lewis base co-catalyst allowed significant reduction in the
amount of iron complex needed to achieve high epoxide con-
versions. The possibility of controlling the selectivity of the re-
action towards either cyclic carbonate or polycarbonate was
evaluated. An efficient switch in selectivity could be achieved
when cyclic epoxides such as cyclohexene oxide and the
seldom explored 1,2-epoxy-4-vinylcyclohexane were used as
substrates. The obtained poly(vinylcyclohexene carbonate)
presents pending vinyl groups, which allowed post-synthetic
cross-linking by reaction with 1,3-propanedithiol. The cross-
linked polycarbonate displayed a substantial increase in the
glass transition temperature and chemical resistance, thus
opening new opportunities for the application of these green
polymers.
Introduction
Emissions of carbon dioxide (CO2) as a consequence of human
activity over the past two hundred years have resulted in an
accumulation of approximately 1 teraton of this molecule in
the atmosphere.[1] The high concentration of CO2 may influ-
ence natural phenomena such as global temperature and the
acidity of oceans and thus affect the environment.[2] This issue
has prompted the international scientific community to pursue
research attempting to mitigate the effects of the increased
concentration of CO2 in the atmosphere. Capture and seques-
tration of CO2 have been proposed as a solution and this topic
has seen significant investigation.[3] At the same time, the wide
availability, renewability and non-toxicity of CO2 has stimulated
research aimed at employing this molecule as a renewable
carbon resource to produce useful chemicals.[1,4–11] Since the
carbon atom in CO2 is in its highest oxidation state, conversion
of CO2 into chemicals requires a high energy input. The low
free energy level of the molecule can be overcome by using
a high free energy coupling partner such as hydrogen, unsatu-
rated compounds, and small-ring heterocycles.[12–14] In this con-
text, the atom-efficient reaction of CO2 with epoxides to pro-
duce polymeric or cyclic carbonates is an attractive option for
the utilization of CO2 (Scheme 1).[15,16] Both possible products
of the CO2–epoxide coupling reaction have growing potential
applications.[12,17] Cyclic carbonates can be employed as green
solvents with attractive features as high polarity and low vapor
pressure. They also find application as electrolytes in Li-ion bat-
teries, and as intermediates for the synthesis of polymers and
fine chemicals.[12] On the other hand, polycarbonates obtained
from alternating co-polymerization of CO2 and epoxides can
be used in the synthesis of polyurethanes, thermoplastics and
polymer resins.[5,17–19] These polycarbonates are considered
green polymers not only for the use of CO2 as feedstock in
their production, but also for their biodegradability.[17] Several
homogeneous and heterogeneous catalysts have been investi-
gated to promote the CO2–epoxide coupling reaction.[1,20–28]
Typically, the catalytic system includes a nucleophile that is
able to open the epoxide ring and a Lewis acid having the role
of activating the epoxide towards the nucleophilic attack. After
the formation of an alkoxide intermediate, CO2 insertion into
the metal-oxygen bond generates a carbonate intermediate
that may undergo ring-closure to produce cyclic carbonate, or
insert further CO2 and epoxide molecules resulting in the
growth of a CO2–epoxide co-polymer chain.[29] During the co-
polymerization, the consecutive insertion of two or more epox-
ides may also happen, which leads to the formation of ether
linkages and thus affects the polymer properties (Scheme 1).
[a] M. Taherimehr, J. P. C. C. Sert¼, Prof. P. P. Pescarmona
Centre for Surface Chemistry and Catalysis
University of Leuven, Kasteelpark Arenberg 23
3001 Heverlee (Belgium)
[b] Prof. A. W. Kleij, Dr. C. J. Whiteoak
Institute of Chemical Research of Catalonia (ICIQ)
Av. Pa¨ısos Catalans 16
43007 Tarragona (Spain)
[c] Prof. A. W. Kleij
Catalan Institute for Research and Advanced Studies (ICREA)
Pg. Lluꢀs Companys 23
08010 Barcelona (Spain)
[d] Prof. P. P. Pescarmona
Chemical Engineering Department
University of Groningen
Nijenborgh 4
9747 AG Groningen (The Netherlands)
E-mail: p.p.pescarmona@rug.nl
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201403323.
ChemSusChem 0000, 00, 0 – 0
1
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
mass, thermal properties, regio-
and stereochemistry of the ob-
tained polymer was carried out.
In the case of poly(vinylcyclo-
hexene carbonate), post-syn-
thetic cross-linking of the poly-
carbonate with 1,3-propanedi-
thiol was performed. This strat-
egy was applied here for the
first time to CO2-based poly-
Scheme 1. The atom-efficient reaction of CO2 with epoxides.
carbonates and proved very ef-
ficient in improving the thermal
In this work, we present a study of iron(III) pyridylamino-bi-
s(phenolate) complexes, FeX[O2NN’], (X=Cl, Br) (Scheme 2) as
novel homogeneous catalysts for the coupling reaction be-
tween CO2 and a range of epoxides. Particular attention was
and mechanical properties of the polymer. This promising
result opens new perspectives for the application of this class
of green polycarbonates.
Results and Discussion
Two iron(III) pyridylamino-bis(phenolate) complexes with bipyr-
amidal geometry in which Br or Cl occupy one apical site
[FeX(O2NN’) with X=Cl or Br, see Scheme 2] were synthe-
sized[32] and tested as novel homogeneous catalysts for the re-
action of carbon dioxide with a broad scope of epoxides (see
the Supporting Information and Figure S1–S4 therein for more
details about the synthesis). Different organic salts were inves-
tigated as co-catalysts, that is, tetrabutylammonium salts
(Bu4NY, Y=Cl, Br, I, OAc) and bis(triphenylphosphoranylide-
ne)ammonium chloride (PPNCl),[12] and the most relevant reac-
tion conditions were studied by means of a tailored high-
throughput reactor unit for performing reactions in supercriti-
cal CO2 (see Figure S5).[33,34] This approach allowed the rapid
and reliable investigation of the chosen set of parameters and,
therefore, enabled a detailed screening of the catalytic proper-
ties of the iron complexes in the CO2–epoxide coupling reac-
tions. The target is to achieve high conversion of different ep-
oxides with high selectivity to either the cyclic carbonate or
the polycarbonate product. Both products are attractive, but
synthesizing only one of them selectively would save undesira-
ble work-up to separate them.
Scheme 2. The chemical structure of complex FeIIIX[O2NN’].
dedicated to the investigation and control of the reaction se-
lectivity towards either cyclic carbonates or polycarbonates,
which entails careful tuning of the catalytic system and of the
reaction conditions.[12,29] In the complex used in this study, the
coordination of FeIII to the pyridylamino-bis(phenol) ligand
through two amino groups and two oxygen atoms implies
a positive charge on the iron metal center, which is balanced
by an anion that can act as a nucleophile. Therefore, it is antici-
pated that the use of these iron pyridylamino-bis(phenolate)
complexes as catalysts can proceed without the need for
adding a co-catalyst. This feature differentiates the Fe-com-
plexes presented here from the neutral iron amino-tripheno-
late complexes, which have been reported to display excellent
performance in the reaction of CO2 with epoxides in combina-
tion with tetrabutylammonium halide co-catalysts.[29,30] In gen-
eral, iron complexes have the advantage of being based on
a non-toxic, widely available and inexpensive metal like iron.[31]
Moreover, the iron pyridylamino-bis(phenolate) complexes can
be obtained in virtually quantitative yields in a two-step proce-
dure (See Scheme S1). Another important asset of these iron
complexes is their air- and moisture-stability, which means that
no inert conditions are required for their handling and storage.
Additionally, no extra drying of the epoxide substrate is re-
quired and no specific equipment is needed to prevent the
presence of adventitious water.
Cyclohexene oxide (CHO) was chosen as substrate for the
first catalytic screening. This is a challenging substrate for the
reaction with CO2 because it has steric hindrance at both car-
bons of the epoxide ring on which the nucleophilic attack can
occur. This epoxide also has a higher tendency to form poly-
carbonates compared to terminal epoxides (e.g., propylene
oxide or 1,2-epoxyhexane) because the geometric strain of the
two connected rings of cis-cyclohexene carbonate limits the
formation of this product. All the catalytic tests were conduct-
ed in supercritical carbon dioxide (scCO2) medium without any
additional solvent. It is worth pointing out that employing
scCO2 can improve the interaction between reagents and cata-
lysts and generally results in higher epoxide conversion and
high carbonate linkages percentage in the poly(cyclohexene
carbonate) chains.[29] The iron catalyst A (X=Cl) or B (X=Br)
are both able to act as bifunctional catalyst to promote the re-
action between cyclohexene oxide and CO2 (Table 1, entries 1
and 18). The halide in apical position acts as the nucleophile to
For the epoxide substrates that produced polycarbonates in
the coupling reaction with CO2, a detailed study of molar
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
2
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 1. Catalytic screening of complexes A and B in the reaction of cyclohexene oxide (CHO) with CO2.[a]
[g]
Entry
Catalyst/co-catalyst
loading[b] [mol%]
Metal/nucleophile Epoxide
Selectivity[d] [%]
Carbonate
TON[f] Mn
PDI[h]
(Mw/Mn)
type
conversion[c] [%] polycarbonate cyclic carbonate linkages[e] [%]
[gmolÀ1
]
1
2
3
4
5
6
7
8
A/–
’’
A/Bu4NCl
’’
5:0
0.5:0
0.5:0.5
0.5:1
0.5:5
0.5:0.5
0.5:1
0.5:5
0.5:0.5
0.5:1
0.5:5
1:1
1:1
1:2
1:3
1:11
1:2
1:3
1:11
1:2
1:3
1:11
1:2
1:3
1:11
1:2
1:3
1:11
1:1
1:2
1:2
–
79
11
60
59
56
63
76
95
12
44
63
10
11
42
49
52
59
67
13
19
3
88
85
82
64
17
60
32
0
60
56
9
86
69
15
81
49
5
12
15
18
36
83
40
68
>99
40
44
91
14
31
85
19
51
95
21
–
–
96
94
–
–
–
–
–
–
–
–
–
16
22
–
-
–
–
1.1
1.2
–
–
–
–
–
–
–
–
–
120
118
112
126
190
200
24
88
126
20
22
84
98
104
118
13
26
20
1418
1012
–
–
–
–
–
–
–
–
–
–
704
903
–
–
–
’’
A/Bu4NBr
’’
’’
A/Bu4NI
’’
’’
A/Bu4NOAc 0.5:0.5
’’
9
10
11
12
13
14
15
16
17
18
19
20
21
22
0.5:1
0.5:5
0.5:0.5
0.5:1
0.5:5
5:0
0.5:0.5
0.5:0.5
0:0.5
0:0.5
’’
–
–
A/PPNCl
’’
’’
>99
>99
–
–
–
–
–
–
1.1
1.2
–
–
–
–
–
–
B/–
79
41
32
0
B/Bu4NCl
B/Bu4NBr
–/Bu4NCl
–/Bu4NBr
59
68
>99
>99
–
–
–
–
–
–
4
0
[a] Reaction conditions: 608C, 80 bar, 18 h, 15 mmol CHO, 3 mmol mesitylene as internal standard for NMR analysis. [b] Relative to the epoxide. [c] Based
1
1
1
on H NMR analysis of the reaction mixture (see Figure S6 for an example). [d] Based on H NMR and FTIR analysis. [e] Based on H NMR analysis of the pu-
rified co-polymer (see Figure S7 for an example). [f] Turnover number, defined as molconverted epoxide/molFe. [g] Determined by GPC in THF at 308C against
polystyrene standards. [h] Polydispersity index defined as PDI=Mw/Mn.
attack the coordinated epoxide, as it is shown in the postulat-
ed mechanism (Scheme S2). In both cases, the polycarbonate
is the major product, though cyclic carbonate was also ob-
served. Higher activity and higher selectivity towards poly-
carbonates was obtained using catalyst A. The difference in ac-
tivity can be explained by the larger radius of bromide com-
pared to chloride, which may impose a higher degree of steric
repulsion for the incoming epoxide substrate when approach-
ing the iron center, thus reducing the reaction rate. On the
other hand, the observed higher selectivity towards poly-
carbonates with catalyst A is the result of the poorer leaving
ability of ClÀ compared to BrÀ. This means that the growth of
the polycarbonate chain with chloride terminal groups is fa-
vored over ring closure (“back-biting”), which would lead to
20), as already observed in the tests with the complex alone.
With ammonium salts and catalyst A in 1:1 ratio (0.5 mol%)
the dependency of the activity on the type of anion shows the
trend OAc~I<Cl~Br (Table 1). These results suggest that the
catalytic activity increases by decreasing the size of the nucleo-
philic anion, as a result of less steric hindrance in the nucleo-
philic attack in the first step of the reaction. The high selectivi-
ty towards the polycarbonate when chloride and acetate salts
are employed as co-catalysts (Table 1, entries 3 and 12) origi-
nates from the poor leaving ability of these species.[12] Howev-
er, full selectivity towards the polycarbonate could not be ach-
ieved (Table 1).
The selectivity towards the cyclic carbonate can be en-
hanced not only by improving the leaving ability of the nucle-
ophile (vide supra) but also by increasing the nucleophile-to-
metal ratio, which facilitates the displacement of the metal-
bound carbonate intermediate by a new nucleophilic anion,
thus favoring the ring-closure reaction.[12,16] The effect of the
relative amount of the co-catalyst was studied by employing
three different ratios between the catalyst A and each of the
co-catalysts (1:1, 1:2, and 1:10). As anticipated, an increase in
the nucleophile to metal ratio leads to higher selectivity to-
wards cyclic carbonate (see the following groups of entries in
Table 1: 3–5; 6–8; 9–11; 12–14; 15–17). With NBu4Br, an excel-
lent 95% epoxide conversion with complete selectivity to-
wards cyclic carbonate production was achieved (Table 1,
entry 8).
cyclic carbonate formation and liberate
(Scheme S2).[12]
a halide anion
To improve the efficiency of the catalytic system, the ratio
between the iron complex and nucleophile was increased by
adding different (external) organic halides as co-catalysts
(binary catalytic system). The addition of a co-catalyst allows
significant reduction in the concentration of iron complex
needed to achieve high epoxide conversion (Table 1, compare
entries 1, 2 and 3). Tuning the nucleophilicity of the anion of
the organic salt and varying the ratio between the iron com-
plex and the organic salt are tools to optimize the activity and
control the product selectivity in the CO2–epoxide coupling re-
action.[12] Experiments performed with NBu4Br and NBu4Cl as
co-catalysts evidence the higher activity of complex A com-
pared to complex B (Table 1, entries 3 and 19, entries 6 and
The effect of the cation of the co-catalyst was examined by
comparing PPNCl with Bu4NCl. (Table 1 entries 3–5 and en-
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
3
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
tries 15–17). Similar activity and selectivity trends are observed
with these two co-catalysts, but slightly higher selectivity for
cyclic carbonate was obtained with PPNCl. This behavior is op-
posite to what generally observed with these two chloride
salts.[29] It can be hypothesized that the PPN+ cation remains
in the vicinity of the chloride anion that attacked the epoxide
due to the interaction of the phenyl rings in PPN+ with the
phenyl rings of the iron complex. This proximity facilitates the
displacement of chloride from the carbonate intermediate,
thus leading to a higher selectivity towards cyclic carbonate.
The poly(cyclohexene carbonate)s obtained in this work are
characterized by low molar mass (see Mn in Table 1) and for
the reactions with higher concentration of co-catalyst only
small oligomers are generated that cannot be separated by
precipitation with acidified methanol. The low molar mass of
these poly(cyclohexene carbonate)s can be ascribed to the
competition between polymer growth and cyclic carbonate
formation and to the presence of adventitious water in the
system, which can cause chain termination.[17,29] The low molar
mass of these short-chain polycarbonates coupled with their
narrow molar mass distribution (cf. low PDI values in Table 1)
makes them suitable for application as diols in the synthesis of
polyurethanes.[5,17,19]
carbonate (Table 2, entry 3). Similar trends were observed with
PPNCl as co-catalyst. Decreasing the molar loading of catalyst
and PPNCl to half the amount of entry 1 leads to slightly
higher TON, whereas the selectivity is nearly unaltered (Table 2,
entries 6 and 7). When comparing the catalytic performance of
complex A with that of the previously reported iron amino-tri-
phenolate complexes under the same reaction conditions and
the same loading of catalyst and co-catalyst,[29] it is observed
that the iron pyridylamino-bis(phenolate) complex reported
here displays higher activity and selectivity when the target
product is cyclohexene carbonate (Table 2, entry 3), whereas
both activity and selectivity are lower under reaction condi-
tions aimed at obtaining the polycarbonate (Table 2, entry 7).
The lower selectivity of complex A towards the formation of
poly(cyclohexene carbonate) is attributed to the lower metal/
nucleophile ratio (1:2) compared to that with the iron amino-
triphenolate complexes (1:1) when the complex and the co-
catalyst are employed in 1:1 ratio. The higher relative amount
of nucleophile favors the displacement of the metal-bound
carbonate intermediate,[16] thus leading to the observed higher
selectivity towards the cyclic carbonate product.
Water that might be present as impurity can interact with
the metal center and thus affect the activity of the catalyst: it
has been reported that the presence of water can substantially
decrease the activity of binary catalytic systems as well as the
selectivity towards polycarbonate and its molar mass.[35] To
study the effect of H2O on our catalytic system, different
amounts of water were added in the reaction catalyzed by A/
Bu4NBr. The decrease in activity from 63%, (Table 2, entry 2), to
7% with 10% of water (Table 2, entry 4), and to 0% when 50%
of water was added (entry 5) demonstrates that water has
a detrimental effect on the catalytic activity of complex A.
The next step in our study was to evaluate the versatility of
the catalyst with another challenging substrate, namely 1,2-
epoxy-4-vinylcyclohexane (VCHO). Similarly to CHO, this sub-
strate is anticipated to display high selectivity towards poly-
carbonate formation. VCHO is largely unexplored as substrate
for co-polymerization with CO2. We selected it for this work be-
cause the vinyl functionality offers the attractive opportunity
of post-polymerization modifications.[12,36] VCHO was tested
with catalyst A and various co-catalysts at 608C, 80 bar, 18 h
To fully appreciate the catalytic activity of the iron pyridyla-
mino-bis(phenolate) complexes, it is important to note that
when Bu4NBr or Bu4NCl were employed alone under the same
conditions (608C, 80 bar, 18 h), the conversion of CHO was less
than 5% compared to values higher than 60% in the presence
of the Fe-complex (Table 1, compare entries 21 and 22 to en-
tries 3 and 6).
The effect of increasing the reaction temperature from 60 to
858C was evaluated with different catalyst to co-catalyst ratios
in the reaction of CO2 with CHO employing complex A and
NBu4Br (Table 2, entries 1–3). The reaction time was decreased
to 3 h to allow comparison at moderate conversions. Since
five-membered cyclic carbonates are the thermodynamic prod-
ucts of the CO2–epoxide coupling reaction, increasing the tem-
perature leads to an increase in the selectivity towards cyclic
carbonates (compare Table 2, entries 1 and 2 with Table 1, en-
tries 6 and 7). The reaction with a cat./co-cat. ratio of 1:10
gives full conversion with complete selectivity towards cyclic
Table 2. Catalytic screening of complex A in the reaction of cyclohexene oxide (CHO) with CO2.[a]
[g]
Entry
Catalyst/co-catalyst
loading[b] [mol%]
Metal/nucleophile Epoxide
Selectivity[d] [%]
Carbonate
TON[f] Mn
PDI[h]
(Mw/Mn)
type
conversion[c] [%] polycarbonate cyclic carbonate linkages[e] [%]
[gmolÀ1
]
1
2
3
4[i]
5[j]
6
A/Bu4NBr
0.5:0.5
0.5:1
0.5:5
0.5:1
0.5:1
1:2
1:3
1:11
1:3
1:3
1:2
1:2
50
63
>99
7
0
32
40
26
0
0
–
60
74
>99
>99
–
–
>99
–
–
–
100
126
200
14
–
128
108
1351
1145
–
–
–
1.0
1.0
–
–
–
’’
’’
’’
’’
A/PPNCl 0.25:0.25
’’ 0.5:0.5
63
60
37
40
>99
>99
808
1612
1.1
1.1
7
54
1
[a] Reaction conditions: 858C, 80 bar, 3 h, 15 mmol CHO, 3 mmol mesitylene. [b] Relative to the epoxide. [c] Based on H NMR analysis of the reaction mix-
ture. [d] Based on 1H NMR and FTIR analysis. [e] Based on 1H NMR spectra of the purified co-polymer. [f] Turnover number, defined as molconverted
/
epoxide
molFe. [g] Determined by GPC in THF at 308C against polystyrene standards. [h] Polydispersity index defined as PDI=Mw/Mn. [i] 10 mol% H2O (relative to
the epoxide) were added to the reaction mixture. [j] 50 mol% H2O (relative to the epoxide) were added to the reaction mixture.
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
4
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 3. Catalytic screening of complexes A and B in the reaction of 1,2-epoxy-4-vinylcyclohexane (VCHO) with CO2.[a]
[g]
Entry
Catalyst/co-catalyst
loading[b] [mol%]
Metal/nucleophile Epoxide
Selectivity[d] [%]
Carbonate
TON[f] Mn
PDI[h]
(Mw/Mn)
type
conversion[c] [%] polycarbonate cyclic carbonate linkages[e] [%]
[gmolÀ1
]
1
2
3
4
5
6
7
8
A/Bu4NCl
’’
’’
A/Bu4NBr
’’
’’
A/Bu4NOAc 0.5:0.5
’’
0.5:0.5
0.5:1
0.5:5
0.5:0.5
0.5:1
0.5:5
1:2
1:3
1:11
1:2
1:3
1:11
1:2
1:3
1:11
1:2
1:3
48
30
53
27
50
92
18
37
35
39
43
48
14
98
67
49
78
48
0
99
96
48
95
70
45
69
2
33
51
22
52
>99
1
4
52
5
86
–
–
–
–
–
–
–
–
96
60
106
54
100
184
36
74
70
78
86
1995
–
–
–
–
–
–
–
–
1.2
–
–
–
–
–
–
–
–
0.5:1
0.5:5
0.5:0.5
0.5:1
0.5:5
9
’’
10
11
12
13
A/PPNCl
’’
’’
>99
–
–
932
–
–
1.2
–
–
30
55
41
1:11
1:2
96
28
B/Bu4NCl
0.5:0.5
–
–
–
[a] Reaction conditions: 608C, 80 bar, 18 h. 15 mmol VCHO, 3 mmol mesitylene. [b] Relative to the epoxide. [c] Based on 1H NMR analysis of the reaction
1
1
mixture (see Figure S8 for an example). [d] Based on H NMR and FTIR analysis. [e] Based on H NMR of the purified co-polymer (see Figure S9 for an exam-
ple). [f] Turnover number, defined as molconverted epoxide/molFe. [g] Determined by GPC in THF at 308C against polystyrene standards. [h] Polydispersity index
defined as PDI=Mw/Mn.
(Table 3). The catalytic activity and product selectivity in the
CO2–VCHO reaction follow similar trends as those observed for
the CO2–CHO coupling reactions, as exemplified by the in-
crease in selectivity towards the cyclic carbonate with higher
relative amount of co-catalyst. However, higher selectivity to-
wards the polycarbonate was obtained compared to the reac-
tion of CHO with CO2 under each set of conditions. Virtually
complete selectivity towards poly(vinylcyclohexene carbonate)
was achieved with OAcÀ and ClÀ as nucleophiles and a Fe-
complex/nucleophile ratio of 1:1, with the latter nucleophile
giving significantly higher conversion (Table 3, entries 1 and
7).[17] The higher nucleophilicity and smaller size of the chloride
compared to the acetate anion explain the observed difference
in activity. The quality of the obtained polycarbonate in terms
of content of carbonate linkages was higher when PPNCl was
employed as co-catalyst compared to Bu4NCl (Table 3, entries 1
and 10). With PPNCl as co-catalyst the obtained polycarbonate
exclusively contained carbonate linkages (no ether linkages)
whereas the polymer obtained in the presence of Bu4NCl as
co-catalyst contained 14% ether linkages, as detected by
1H NMR and FTIR spectroscopy (Figure S10, S11). Remarkably,
the selectivity of the reaction between VCHO and CO2 could
be completely switched by using Bu4NBr as co-catalyst and
performing the reaction with a catalyst-to-co-catalyst ratio of
1:10. Under these conditions, full selectivity towards the cyclic
carbonate was achieved, with an excellent epoxide conversion
of 92% (Table 3, entry 6). These results show that in the reac-
tion of VCHO with CO2 catalyzed by complex A, tuning the
type and relative amount of co-catalyst allows efficiently
switching the reaction selectivity between exclusive polymeric
or exclusive cyclic carbonate formation. Similarly to what ob-
served with CHO as substrate, higher selectivity towards cyclic
carbonate was attained in the CO2–VCHO coupling reaction by
increasing the reaction temperature from 608C to 858C (com-
pare Table 3 and Table 4). Increasing the conversion of VCHO
in the reaction catalyzed by complex A/PPNCl by performing
the reaction at 858C for 18 h also leads to a substantial in-
crease in the polycarbonate molar mass: from Mn =
2086 gmolÀ1 (PDI=1.2) after 3 h to Mn =3784 gmolÀ1 (PDI=
1.4) (see Table 4, entries 5 and 6). In line with what found with
CHO, the polycarbonate selectivity is largely unaffected by in-
creasing the catalyst and co-catalyst loading while keeping
their ratio constant (Table 4, entries 3–5).
The potential of complex A as homogeneous catalyst for the
reaction of CO2 with epoxides was further evaluated employ-
ing a wide scope of substrates. Styrene oxide (SO) is generally
Table 4. Catalytic screening of complex A in the reaction of 1,2-epoxy-4-vinylcyclohexane (VCHO) with CO2.[a]
[g]
Entry Time
[h]
Catalyst/co-catalyst
loading[b] [mol%]
Metal/nucleophile Epoxide
Selectivity[d] [%]
Carbonate
TON[f] Mn
PDI[h]
(Mw/Mn)
type
conversion[c] [%] polycarbonate cyclic carbonate linkages[e] [%]
[gmolÀ1
]
1
2
3
4
5
6
3
3
3
3
3
A/Bu4NBr 0.5:0.5
’’ 0.5:1
A/PPNCl 0.25:0.25
1:2
1:3
1:2
1:2
1:2
1:2
29
24
18
34
69
84
49
35
78
79
73
77
51
65
22
21
27
23
66
–
77
85
>99
>99
58
48
72
68
69
84
–
–
–
–
–
–
2.0
1.2
1.4
’’
’’
’’
0.5:0.5
1:1
1:1
1840
2086
3784
18
1
[a] Reaction conditions: 858C, 80 bar, 3 h or 18 h.15 mmol VCHO, 3 mmol mesitylene. [b] Relative to the epoxide. [c] Based on H NMR analysis of the reac-
1
1
tion mixture. [d] Based on H NMR and FTIR analysis. [e] Based on H NMR spectra of the purified co-polymer. [f] Turnover number, defined as molconverted ep-
oxide/molFe. [g] Determined by GPC in THF at 308C against polystyrene standards. [h] Polydispersity index defined as PDI=Mw/Mn.
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
5
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
reported as a relatively sluggish substrate in the reaction with
carbon dioxide compared to other terminal epoxides.[17] Com-
plex A was used as catalyst in combination with a tetrabutylam-
monium halide, because these readily available co-catalysts
gave the best performance with CHO and VCHO as substrates
(vide supra). Full selectivity towards styrene carbonate was ob-
tained in all the tests. For this reaction, the activity order as
Table 6. Substrate screening with catalyst A in the reaction with CO2.[a]
Entry Substrate
Epoxide
conversion[b] [%] polycarbonate cyclic carbonate
Selectivity[c] [%]
TON[d]
200
1
2
>99
0
0
>99
>99
a
function of the co-catalyst is Bu4NI>Bu4NBr>Bu4NCl
96
192
(Table 5), which corresponds to the trend of leaving ability of
the halides. An excellent conversion (92%) was achieved with
complex A/Bu4NI at 608C (Table 5, entry 3). For the reaction of
CO2 with a range of other epoxides including two natural
products (the terpene-based substrates in Table 6 entries 8 and
9) and oxetane, complex A was employed in combination with
Bu4NBr (Table 6). This co-catalyst is often reported as a suitable
co-catalyst for the synthesis of cyclic carbonates due to the
balance between nucleophilicity and leaving ability of the bro-
mide anion.[12]
3
4
88
56
0
0
>99
>99
176
112
5
50
40
60
100
6
7
29
7
49
0
51
58
14
Good to excellent conversions were achieved with the termi-
nal epoxides at 858C, 80 bar and using a reaction time of 3 h,
demonstrating the versatility of catalyst A (Table 6, entries 1–
4). The lower ring strain of oxetane represents a drawback in
its reaction with CO2 because it makes the ring opening step
more difficult, and this is illustrated by the low activity in the
conversion of this substrate (Table 6, entry 7). The presence of
a methanetetrayl group on the epoxide ring imposes a signifi-
cant steric barrier to the nucleophilic attack, and this explains
the lack of success when attempting to react limonene oxide
(Table 6, entry 8) and a-pinene oxide (Table 6, entry 9).
>99
8
9
0
0
–
–
–
–
0
0
[a] Reaction conditions: 858C, 80 bar, 3 h, catalyst 0.5 mol%, co-catalyst Bu4NBr
0.5 mol%, epoxide 15 mmol, mesitylene 3 mmol (mol% relative to the epoxide).
[b] Based on 1H NMR analysis of the reaction mixture. [c] Based on 1H NMR and
FTIR analysis. [d] Turnover number, defined as molconverted epoxide/molFe.
Microstructure and properties of polycarbonates
For the substrates that could be converted into polycarbonate
that could be isolated (i.e., in the case of CHO and VCHO) the
physicochemical properties of the polymers were investigated.
13C{1H} NMR analysis of purified polycarbonate allowed to de-
termine the regio- and stereo-selectivity of the co-polymeri-
zation reaction. With CHO as substrate, the 13C{1H} NMR spec-
trum of the carbonate region shows signals corresponding to
syndiotactic junctions (at d=153.2 ppm), isotactic junctions (at
d=153.8 ppm) and a signal at d=154.3 ppm, which is as-
signed to carbonate junctions located in proximity of hydroxyl-
terminated ends of the chain (see Figure S12 in the Supporting
Information).[37,38] The assignment of this peak is in agreement
with the low Mn observed for the polycarbonates reported in
this work. The low intensity peak at d=155.1 ppm might indi-
cate the presence of a small amount of terminal carbonate
groups.[37] These results indicate that atactic poly(cyclohexene
carbonate) was formed, as expected considering the non-chiral
nature of the catalyst, and show that the polymer is mainly ter-
minated with hydroxyl groups.[17]
In the reaction of CO2 with VCHO, the formation of atactic
polymer chains is evidenced by the complex signal in the
region between d=153.2 and 153.5 ppm of the 13C{1H} NMR
spectrum (Figure S13). Head-to-tail (HT) regio-regular sequen-
ces (Scheme 3) are expected to be favored over tail-to-tail (TT)
and head-to-head (HH) regio-regular sequences because they
would lead to lower steric hindrance between the pending
vinyl groups of adjacent repeating units. Since VCHO has four
different stereoisomers (Scheme 3), a polymer with exclusively
HT regio-regular sequences would have ten possible stereo-se-
quences and thus generate a complex signal in the carbonate
region of the 13C NMR spectrum, as the one observed in this
work. However, based on the current data it cannot be exclud-
ed that TT and HH regio-regular sequences are also present. In
analogy with the polycarbonate based on CHO, the low inten-
Table 5. Catalytic screening of complex A in the reaction of styrene
oxide (SO) with CO2.[a]
.
Entry
Catalyst/co-catalyst
Epoxide conversion[b] [%]
TON[c]
1
2
3
A/Bu4NCl
A/Bu4NBr
A/Bu4NI
28
73
92
56
146
184
[a] Reaction conditions: 608C, 80 bar, 18 h, 15 mmol SO, 3 mmol mesity-
lene, catalyst 0.5 mol%, co-catalyst 0.5 mol% (mol% relative to the epox-
ide). [b] Based on 1H NMR analysis of the reaction mixture. [c] Turnover
number, defined as molconverted epoxide/molFe.
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
6
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Scheme 4. Schematic representation of the free-radical cross-linking of
PVCHC.
Scheme 3. A) Stereoisomers of 1,2-epoxy-4-vinyl-cyclohexane, B) Head-to-tail
PVCHC chain structures.
sity peak at d=153.9 ppm is assigned to carbonate junctions
located in proximity of hydroxyl-terminated ends of the chain
(Figure S13).
Improvement of the physical properties of PVCHC by cross-
linking
The presence of functional groups on the polycarbonate back-
bone offers the possibility of post-polymerization modifications
that enable tailoring the physicochemical features of the poly-
mer such as hydrophilicity, solubility, biodegradability and ther-
mal and mechanical properties.[17,39] Cross-linking generates
networks by interconnecting the polymer chains and, there-
fore, can be employed to increase the molar mass, the chemi-
cal resistance, and the stability against elevated temperatures
and mechanical deformation of the polymers. This strategy is
applied here for the first time to poly(vinylcyclohexene carbon-
ate) [PVCHC] by connecting the polymer chains through the
pendant double bond of the vinyl groups. For this purpose,
the purified polycarbonate (obtained form the reaction men-
tioned in entry 1, Table 3) was reacted with 1,3-propanedithiol
in the presence of the radical initiator azobisisobutyronitrile
(AIBN) (thiol-ene reaction, see Scheme 4). The successful cross-
linking was demonstrated by the disappearance of the peaks
of the double bond of the vinyl group in the solid-state
13C{1H} NMR spectrum of the cross-linked material (Figure S14
in the Supporting Information). As a result of the cross-linking,
the glass transition temperature (Tg) of the polymer shows
a considerable improvement of 558C from 75 to 1308C, as
measured by DSC analysis (Figure S15 in Supporting Informa-
tion). The cross-linked polymer also displays enhanced chemi-
cal resistance, as indicated by its insolubility in (hot) THF, DMF,
acetone, ethanol, CHO or propylene carbonate. SEM images
show that the morphology of the cross-linked PVCHC differs
significantly from that of PVCHC (Figure 1 and 2). The new
morphology might arise from nucleation during dissolution of
PVCHC in ethanol followed by particle growth through cross-
linking.[40] The observed properties of the thermosetting poly-
mer obtained by cross-linking PVCHC are promising for a po-
tential application as engineering plastic.[41] Further investiga-
tion will aim at tuning the physicochemical properties of these
polymers by varying the degree of cross-linking by changing
Figure 1. SEM image of PVCHC after re-precipitation and drying in a Schlenk
line.
Figure 2. SEM image of dried cross-linked PVCHC.
the 1,3-propanedithiol loading. This work will also involve
a study of the increase in stiffness and thermal resistance as
a function of the cross-linking degree.
Conclusions
FeIIIX[pyridylamino-bis(phenolate)] complexes (X=Cl, Br), alone
or in combination with ionic Lewis bases, have been evaluated
as homogeneous catalysts for the coupling reaction of CO2
and epoxides. The catalytic system has been shown to be ver-
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
7
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
which involves the reaction of the ligand H2[O2NN’] with the select-
ed iron salt (anhydrous FeCl3 or anhydrous FeBr3) in 1:1 ratio in
methanol solution and in the presence of triethylamine as a HX
scavenger (X=Cl, Br).[32] The formation of the desired FeCl[O2NN’]
and FeBr[O2NN’] complexes was monitored by means of FTIR and
UV/Vis.[32] Full details of the synthesis and characterization of the
complexes can be found in the Supporting Information.
satile in the conversion of a broad scope of epoxides. Besides
displaying high activity, the FeCl[O2NN’] catalyst is remarkably
robust, and unpurified substrates can be used without con-
cerns for the presence of water impurities in the reaction set-
up. The parameters governing the selectivity of the reaction
towards cyclic or polymeric carbonate were studied and opti-
mized to control and maximize the selectivity towards each of
the two valuable products. With terminal epoxides, the cyclic
carbonate was the main product, whereas with CHO and the
less explored VCHO as substrates it was also possible to selec-
tively obtain polycarbonates with high percentage of carbon-
ate linkages. Particularly, for the reaction of VCHO with CO2 it
was possible to fully switch the selectivity of the reaction be-
tween cyclic to polymeric carbonate by tuning the type and
relative amount of organic salt employed as Lewis base. The
obtained PVCHC was efficiently cross-linked with 1,3-propane-
dithiol. This new strategy proved effective in increasing the
glass transition temperature of the polymer by 558C and ren-
dering it insoluble in organic compounds. Cross-linked PVCHC
is a thermosetting polymer that may find applications as engi-
neering plastic.
Catalytic tests, synthesis of cyclic and/or polymeric carbo-
nates
The catalytic tests were performed in a high-throughput unit (Inte-
grated Lab Solution, ILS) consisting of 24 batch reactors (see Fig-
ure S5) that can operate simultaneously at temperature up to
1808C and CO2 pressure up to 180 bar. A detailed description of
this unit can be found elsewhere.[33,34] For each experiment, the
catalyst, co-catalyst, the selected epoxide (15 mmol) and mesity-
lene (3 mmol) as the NMR internal standard were weighed and
added to a glass vial along with a magnetic stirring bar. The vials
were sealed with a plastic cap containing a silicon rubber seal
where two needles were inserted for the CO2 gas to enter and cir-
culate through the vial. After closing the reactor, all the steps to
reach the required reaction conditions were controlled using tail-
ored software (ProControl). First, the reactors were purged with N2
and CO2 to remove air. Then, the exit valves were closed and the
reactors were pressurized with CO2 (99.995% purity, supplied by
Air Liquide). For the reactions at a pressure of 80 bar and a temper-
ature of 858C, carbon dioxide was pumped until the pressure
inside the reactor reached ~70 bar while the temperature was in-
creased to 858C. Next, the reactor was further pressurized until the
target value of 80 bar was reached inside the reactors, at which
point the valve connecting the reactor and the main line was
closed. Similar protocols were followed for the reactions at 608C.
The start of each reaction was defined as the moment at which
the selected reaction conditions were reached and the magnetic
stirring was turned on at 900 rpm. After the required reaction time,
the stirring was turned off and the reactor was cooled down until
it reached 308C (which takes about 2 h). Then, the depressurization
process was initiated and continued until the pressure inside the
reactor was below 2 bar (these conditions were reached in about
further 2 h). Finally, the reactor block was opened and the glass
vials were removed from the block. 50 mL of each reaction mixture
Experimental Section
Materials
2,4-di-tert-butylphenol (99% purity), 2-picolylamine (99% purity),
aqueous formaldehyde (37 wt% in H2O), anhydrous iron(III)chloride
(FeCl3 99.99% purity), anhydrous iron(III) bromide (FeBr3 99.99%
purity), tetrabutylammonium iodide (Bu4NI, 98% purity), tetrabuty-
lammonium bromide (Bu4NBr, 98% purity), tetrabutylammonium
chloride (Bu4NCl, 98% purity), Bis(triphenylphosphoranylidene)am-
monium chloride (PPNCl) (98% purity), mesitylene (98% purity), cy-
clohexene oxide (CHO, 98% purity), styrene oxide (SO, 97% purity),
1,2-epoxy-4-vinylcyclohexane (VCHO, 98% purity), glycidyl metha-
crylate (97% purity), tert-butyl glycidyl ether (99% purity), glycidyl
isopropyl ether (98% purity), 1,3-trimethylene oxide (97% purity),
a-pinene oxide (97% purity), (+)-limonene oxide (97% purity), azo-
bisisobutyronitrile (AIBN) (99% purity), and 1,3-propanedithiol
(99% purity), unstabilized THF (99.8% purity) and solvents [diethyl
ether, tetrahydrofuran (THF), methanol (MeOH), toluene] were pur-
chased from Sigma–Aldrich and used without further purification.
Deuterated chloroform (CDCl3) (>99.6 atom%), as solvent for
1H NMR and 13C{1H} NMR measurements, was purchased from
Acros Organics.
1
were diluted with 600 mL of CDCl3 to measure the H NMR spectra
1
to calculate the conversion (examples of these H NMR spectra are
presented in the Supporting Information). To determine the prod-
uct selectivity, one drop of the reaction mixture was diluted with
ethanol and placed on a KBr plate, which was used for analysis by
infrared spectroscopy (FTIR). The selectivity towards cyclic or poly-
meric carbonate was determined on the basis of the C=O absorp-
tion band, which is observed at ~1800 cmÀ1 for the cyclic carbon-
ate and at ~1750 cmÀ1 for the polycarbonate.[17,29] Acidified metha-
nol (1m of HCl in methanol) was added to the samples containing
polycarbonate, as determined by FTIR, to precipitate the polymer.
The liquid was separated using a pipette and the remaining vis-
cous polycarbonate was placed in a Schlenk-line to vacuum-dry
overnight. Around 10 mg of the obtained powder was dissolved in
Synthesis of the pyridylamino-bis(phenol), H2[O2NN’] ligand
The ligand was synthesized following a previously reported proce-
dure involving the reaction of 2,4-di-tert-butylphenol, 2-picolyla-
mine and formaldehyde.[32] The identity of the compound was de-
termined by 1H NMR and FTIR spectroscopy.[32] The details of the
synthesis and characterization are provided in the Supporting In-
formation.
1
600 mL CDCl3 and analyzed by H NMR spectroscopy to determine
the number of repeating units, and by 13C{1H} NMR spectroscopy
to determine the stereoselectivity. 20 mg of polycarbonate were
dissolved in 500 mL of unstabilized THF to measure the mass aver-
age molar mass (Mw) and the number average molar mass (Mn) by
gel-permeation chromatography (GPC).
Synthesis of the iron(III) pyridylamino-bis(phenolate),
FeCl[O2NN’] and FeBr[O2NN’] complexes
The synthesis of the iron(III) pyridylamino-bis(phenolate) com-
plexes was performed following a previously reported procedure,
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
8
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[8] R. Nakano, S. Ito, K. Nozaki, Nat. Chem. 2014, 6, 325–331.
[9] I. Omae, Coord. Chem. Rev. 2012, 256, 1384–1405.
[10] F. Manjolinho, M. Arndt, K. Gooßen, L. J. Gooßen, ACS Catal. 2012, 2,
2014–2021.
[11] K. M. K. Yu, I. Curcic, J. Gabriel, S. C. E. Tsang, ChemSusChem 2008, 1,
893–899.
[12] P. P. Pescarmona, M. Taherimehr, Catal. Sci. Technol. 2012, 2, 2169–2187.
[13] M. Aresta, Carbon Dioxide as Chemical Feedstock, Wiley-VCH, Weinheim,
2012.
Poly(4-vinyl-1,2 cyclohexene carbonate) cross-linking reac-
tion
PVCHC obtained form the reaction mentioned in Table 3, entry 1
(1000 mg corresponding to ~6 mmol of the 1,2-epoxy-4-vinylcyclo-
hexane), 1,3-propanedithiol (3 mmol), azobisisobutyronitrile
(90 mg) as radical initiator, and ethanol were weighed in a three-
neck round-bottom flask. Argon was bubbled for 10 min and the
mixture was stirred for 20 h under Ar at 788C. After cooling to
room temperature, the foam-like solid (see Figure S16 in Support-
ing Information) was filtered, washed with hot methanol and then
dried in a vacuum oven at 408C overnight.[34]
[14] P. W. N. M. van Leeuwen, Science of Synthesis: C-1 Building Blocks in Or-
ganic Synthesis, Thieme, Stuttgart, 2013.
[15] M. North, R. Pasquale, C. Young, Green Chem. 2010, 12, 1514–1539.
[16] D. J. Darensbourg, Chem. Rev. 2007, 107, 2388–2410.
[17] M. Taherimehr, P. P. Pescarmona, J. Appl. Polym. Sci. 2014, 131, 41141.
[18] J. Langanke, A. Wolf, J. Hofmann, K. Bohm, M. A. Subhani, T. E. Muller,
W. Leitner, C. Gurtler, Green Chem. 2014, 16, 1865–1870.
[19] M. R. Kember, C. K. Williams, J. Am. Chem. Soc. 2012, 134, 15676–15679.
[20] S. Klaus, M. W. Lehenmeier, C. E. Anderson, B. Rieger, Coord. Chem. Rev.
2011, 255, 1460–1479.
[21] D. J. Darensbourg, S. J. Wilson, Green Chem. 2012, 14, 2665–2671.
[22] M. R. Kember, A. Buchard, C. K. Williams, Chem. Commun. 2011, 47,
141–163.
[23] M. E. Wilhelm, M. H. Anthofer, M. Cokoja, I. I. E. Markovits, W. A. Herr-
mann, F. E. Kꢁhn, ChemSusChem 2014, 7, 1357–1360.
[24] C. J. Whiteoak, A. Nova, F. Maseras, A. W. Kleij, ChemSusChem 2012, 5,
2032–2038.
[25] T. Ema, Y. Miyazaki, J. Shimonishi, C. Maeda, J.-Y. Hasegawa, J. Am.
Chem. Soc. 2014, 136, 15270–15279.
[26] T. Werner, H. Bꢁttner, ChemSusChem 2014, 7, 3268–3271.
[27] A. M. Hardman-Baldwin, A. E. Mattson, ChemSusChem 2014, 7, 3275–
3278.
[28] G. W. Coates, D. R. Moore, Angew. Chem. Int. Ed. 2004, 43, 6618–6639;
Angew. Chem. 2004, 116, 6784–6806.
Characterization
1H NMR and 13C{1H} NMR spectra were measured on a Bruker Ad-
vance 300 MHz spectrometer. FTIR spectra were obtained using
a Bruker IFS66v/S spectrometer equipped with a vacuum chamber
chargeable with liquid nitrogen. Gel permeation chromatography
(GPC) measurements were performed utilizing a Shimadzu 10 A ap-
paratus with a mixed D-column: 5 mm (medium to high Mn) main-
tained at 308C, and with two tunable absorbance detectors (set at
a wavelength of 250 nm and 300 nm) or a refractive index detector
(RID). The instrument was calibrated with polystyrene standards.
The morphology of polymers was studied by scanning electron mi-
croscopy (SEM) on a Philips XL 30 FEG with an acceleration voltage
of 10 kV and working temperature of the emitter of 1800 K. Differ-
ential scanning calorimetry (DSC) analysis was performed on
a Q2000 DSC (TA Instruments) by cycling between 20 and 2008C
with heating/cooling rates of 208CminÀ1 under N2 atmosphere.
[29] M. Taherimehr, S. M. Al-Amsyar, C. J. Whiteoak, A. W. Kleij, P. P. Pescarmo-
na, Green Chem. 2013, 15, 3083–3090.
[30] C. J. Whiteoak, E. Martin, M. M. Belmonte, J. Benet-Buchholz, A. W. Kleij,
Adv. Synth. Catal. 2012, 354, 469–476.
Acknowledgements
[31] A. Buonerba, A. De Nisi, A. Grassi, S. Milione, C. Capacchione, S. Vagin,
B. Rieger, Catal. Sci. Technol. 2014, 5, 118–123.
[32] K. Hasan, C. Fowler, P. Kwong, A. K. Crane, J. L. Collins, C. M. Kozak,
Dalton Trans. 2008, 2991–2998.
[33] M. Taherimehr, A. Decortes, S. M. Al-Amsyar, W. Lueangchaichaweng,
C. J. Whiteoak, E. C. Escudero-Adꢂn, A. W. Kleij, P. P. Pescarmona, Catal.
Sci. Technol. 2012, 2, 2231–2237.
[34] C. Aprile, F. Giacalone, P. Agrigento, L. F. Liotta, J. A. Martens, P. P. Pescar-
mona, M. Gruttadauria, ChemSusChem 2011, 4, 1830–1837.
[35] X.-B. Lu, L. Shi, Y.-M. Wang, R. Zhang, Y.-J. Zhang, X.-J. Peng, Z.-C.
Zhang, B. Li, J. Am. Chem. Soc. 2006, 128, 1664–1674.
This work is supported by a START1 funding from KU Leuven.
Prof. Guy Koeckelberghs is kindly acknowledged for the support
provided in the GPC analyses. Leander Verbelen and Prof. Peter
Van Puyvelde are kindly acknowledged for the DSC measure-
ments. Dr. Lik Hong Wee is kindly acknowledged for the SEM
measurements. Finally, we would like to thank Ehsan Pesaran for
fruitful discussion.
[36] D. J. Darensbourg, F.-T. Tsai, Macromolecules 2014, 47, 3806–3813.
[37] K. Nakano, K. Nozaki, T. Hiyama, Macromolecules 2001, 34, 6325–6332.
[38] S. Abbina, G. Du, Organometallics 2012, 31, 7394–7403.
[39] D. Vink, “CO2 aids Bayer’s sustainablility efforts,” can be found under
http://www.plasticsnews.com/article/20131016/NEWS/131019950/co2-
aids-bayers-sustainablility-efforts, accessed on Jul 15, 2014.
[40] K. Dusek, W. J. MacKnight, Cross-Linked Polymers Chemistry, Properties,
and Applications, American Chemical Society, Washington, DC, 1988.<
[41] E. Lokensgard, Industrial Plastics: Theory and Applications, 5th ed.,
Delmar, Cengage Learning, Clifton Park, NY, 2008.
Keywords: atom efficiency · CO2 fixation · cross linking ·
epoxide · iron pyridylamino-bis(phenolate) · polycarbonates
[1] M. Mikkelsen, M. Jørgensen, F. C. Krebs, Energy Environ. Sci. 2010, 3, 43–
81.
[2] P. N. Pearson, M. R. Palmer, Nature 2000, 406, 695–699.
[3] L. Yang, H. Wang, ChemSusChem 2014, 7, 962–998.
[4] A. T. Najafabadi, Int. J. Energy Res. 2013, 37, 485–499.
[5] M. Peters, B. Kçhler, W. Kuckshinrichs, W. Leitner, P. Markewitz, T. E.
Mꢁller, ChemSusChem 2011, 4, 1216–1240.
[6] M. Aresta, A. Dibenedetto, A. Angelini, Chem. Rev. 2014, 114, 1709–
1742.
Received: November 27, 2014
[7] B. Yu, L.-N. He, ChemSusChem 2015, 8, 52–62.
Published online on && &&, 0000
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
9
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
M. Taherimehr, J. P. C. C. Sert¼, A. W. Kleij,
C. J. Whiteoak, P. P. Pescarmona*
CO2 meets epoxides: Iron pyridylami-
no-bis(phenolate) complexes are highly
active catalysts for the atom-efficient re-
action of CO2 with a variety of epoxides.
The selectivity can be switched between
the cyclic or polymeric carbonate when
using cyclic epoxides such as cyclohex-
ene oxide and 1,2-epoxy-4-vinylcyclo-
hexaneas substrates. Cross-linking of
the obtained poly(vinylcyclohexene car-
bonate) leads to substantial increase in
the Tg and chemical resistance of the
polymer.
&& – &&
New Iron Pyridylamino-Bis(Phenolate)
Catalyst for Converting CO2 into Cyclic
Carbonates and Cross-Linked
Polycarbonates
&
ChemSusChem 0000, 00, 0 – 0
www.chemsuschem.org
10
ꢀ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!