Aqueous-microwave synthesized carboxyl functional molecular ribbon coordination framework catalyst for the synthesis of cyclic carbonates from epoxides and CO2

Article abstract of DOI:10.1039/c3gc41833b

A carboxyl-containing coordination polymer catalyst {Cu(Hip) ₂(Bpy)}_n (CHB) was synthesized rapidly in an aqueous medium using microwave energy and characterized for its structure, morphology, acid-base sites and heterogeneity using experimental and physicochemical techniques. The microwave route was established as competent with the hydrothermal pathway. Exploration of the coordination modes of metal-organic framework (MOF) ligands, especially carboxyl spacers, in achieving reactive functional groups is herein illustrated as crucial rather than focusing merely on the porosity of MOF catalysts. The work represents the first report of a carboxyl-containing MOF class material employed for synthesizing cyclic carbonates from epoxide and CO₂. The metal-carboxyl mediated cycloaddition of allyl glycidyl ether and CO₂ by the cooperative influence of Cu²⁺ and the COOH of CHB was synergistically enhanced by an ionic liquid co-catalyst to obtain allyl glycidyl carbonate in 89% yield at optimised reaction conditions. The epoxide substrate scope, effect of reaction parameters and catalyst recyclability (up to 5 cycles) were also studied.

Full text of DOI:10.1039/c3gc41833b

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Aqueous-microwave synthesized carboxyl
functional molecular ribbon coordination
framework catalyst for the synthesis of cyclic
carbonates from epoxides and CO2†
Cite this: Green Chem., 2014, 16,
1607
Amal Cherian Kathalikkattil, Dong-Woo Kim, Jose Tharun, Han-Geul Soek,
Roshith Roshan and Dae-Won Park*
A carboxyl-containing coordination polymer catalyst {Cu(Hip)2(Bpy)}n (CHB) was synthesized rapidly in an
aqueous medium using microwave energy and characterized for its structure, morphology, acid-base
sites and heterogeneity using experimental and physicochemical techniques. The microwave route was
established as competent with the hydrothermal pathway. Exploration of the coordination modes of
metal–organic framework (MOF) ligands, especially carboxyl spacers, in achieving reactive functional
groups is herein illustrated as crucial rather than focusing merely on the porosity of MOF catalysts. The
work represents the first report of a carboxyl-containing MOF class material employed for synthesizing
cyclic carbonates from epoxide and CO2. The metal–carboxyl mediated cycloaddition of allyl glycidyl
ether and CO2 by the cooperative influence of Cu2+ and the COOH of CHB was synergistically enhanced
by an ionic liquid co-catalyst to obtain allyl glycidyl carbonate in 89% yield at optimised reaction
conditions. The epoxide substrate scope, effect of reaction parameters and catalyst recyclability (up to
5 cycles) were also studied.
Received 3rd September 2013,
Accepted 20th November 2013
DOI: 10.1039/c3gc41833b
www.rsc.org/greenchem
Despite their ease of reusability, most heterogeneous catalysts
are fraught with insufficient activities, or the requirement
Introduction
The increasing level of carbon dioxide in the atmosphere is for solvent assistance, or harsh conditions. This necessitates
considered as a major contributor to global warming. The the rational design of a recyclable acid-base multifunctional
non-toxicity and easy availability of carbon dioxide (by-product catalyst that works even under mild reaction conditions.2,5
of chemical and power industries), on the other hand, opens
Metal–organic frameworks (MOFs) or coordination poly-
up its potential as a C1 source in the production of com- mers (CPs), the crystalline compounds constructed from
pounds such as such as dimethyl carbonate, N,N′-disubstituted organic linkers that bridge metal atoms or clusters, are an out-
ureas, cyclic carbonates, cyclic urethanes, formic acid, etc.1 standing class of materials with versatile applications.6 The
The cycloaddition of CO2 with epoxide substrates to yield ability to impart task-oriented properties through the judicious
cyclic carbonates1a,b,h,i,2 is attractive due to the wide appli- choice of constituent materials and the rigidity of their frame-
cation of cyclic carbonates in polar aprotic solvents, electrolytic works make them highly desirable in catalysis.1h,2,7 A series of
elements of batteries, intermediates or precursors in fine multivariate MOF catalysts like IRMOF, MOF-5, MOF-74, MIL
chemicals, pharmaceuticals, and polymer synthesis.3 Even etc. based on exo-dentate di-carboxylic acid template ligands
though homogeneous catalysts represent the most active cata- were recently employed in cycloaddition.7a–e However, reports
lytic systems, they suffer from unviable catalyst recovery process.4 including ours, showing the promoting role of hydrogen-
bonding moieties, especially carboxyl groups, at the solid/
liquid interface of various catalysts in the activation of epoxi-
School of Chemical and Biomolecular Engineering, Pusan National University, Busan
des1b,8 surmised the catalytic possibility of an MOF that con-
609-735, Korea. E-mail: dwpark@pusan.ac.kr
tains reactive carboxyl groups for effectuating cycloaddition.
Unfortunately, reactive COOH-containing MOFs have not been
reported in cycloaddition reactions to date; the major obstacle
being the scarcity of such MOFs, attributable to the preferential
binding tendency of linear exo-dentate dicarboxylates towards
the metal during a typical MOF assembly.9a Angular COOH
†Electronic supplementary information (ESI) available: Experimental details of
Knoevenagel condensation and acetalization reactions, experimental and discus-
sion of CO2-TPD analysis and results, the GC-MS results of cycloaddition reac-
tants and products, effects of CO2 pressure with CHB(M)/TBAB system, and the
schematic representation of cycloaddition mechanism by CHB(M). CCDC
657329. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c3gc41833b
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 1607–1616 | 1607

View Article Online
Paper
Green Chemistry
Scheme 1 Binding modes of carboxylic acid in the frameworks of (a) HKUST-1, (b) CHB and (c) ZHB.
(120°) constituted MOFs employed in cycloaddition viz., synergistically with the hydrogen bonding capable COOH
Cu3(BTC)2 (HKUST-1, benzene tricarboxylicacid, BTC, and OH groups in promoting catalysis. Besides, a quaternary
linker),9b,c and ZnHipBpy (ZHB, from 5-hydroxyisophthalic ammonium salt was engaged as co-catalyst for the optimi-
acid, Hip)7g were also deprived of reactive COOH groups. sation of reaction conditions.
Surprisingly, we found that a copper coordination polymer
{Cu(Hip)2(Bpy)}n (CHB) synthesized from the same organic
constituents of ZHB possessed reactive COOH in its framework
Results and discussion
(Scheme 1).8d CHB consists of the carboxylate end of its
mono-deprotonated Hip ligand appended to the Cu–Bpy Mixed-linker or multivariate MOFs reported for cycloaddition
chain, retaining the second COOH end for intra-molecular
H-bonds. Thus CHB offer itself as a potential candidate
rely on multi-variant metal clusters and/or template carboxy-
late spacers with analogous functional groups. Porosity and
which contribute reactive carboxyl group. Further, the CO2 adsorption capacities, the presence of a co-catalyst, or syn-
avoidance of high-boiling organic solvents (DMF, DEF etc.) for
material synthesis, and the absence of water molecules in the
thetic modifications (PSM) were considered equally important
in choosing an efficient MOF catalyst.1h,7a–c However, unlike
crystal lattice and coordination sphere of CHB are advan- MOF-74 or IRMOFs, multi-variant or iso-structural MOFs are
tageous in relieving the tedious or prolonged activation and
purification steps for typical MOFs.7e Other than purifi-
not as simply achievable with every MOF, since the coordi-
nation modes of carboxyl group is largely dependent on variety
cation steps, the circumvention of a long material synthesis of factors ranging from the nature of metal salt precursor to
the reaction conditions employed for synthesis.9a HKUST-1,
time is another drawback for many MOFs. Slow evaporation or
diffusion methods may require up to several weeks to achieve
MOFs despite the crystallinity and phase purity offered. Further, area of 1270 m2 g−1 and high CO2 adsorption capabilities,
the tri-carboxylic acid derived porous MOF with a BET surface
these methods are unfeasible for large-scale applications owing
to their scaling-up limitations. A direct mixing method, though,
whose Lewis acid sites (framework node) were recently
reported to catalyze the cycloaddition of CO2 and epichlorohy-
allows an effective scale up but phase purity and crystallinity drin with a >30% yield of cyclic carbonate.7i,9c Even though
could be largely affected.10a Solvothermal methods also need a
CHB (the catalyst under study) and HKUST-1 were synthesized
considerably long time for MOF synthesis. In contrast to such
conventional catalyst synthesis techniques, microwave energy is ligands, their resultant structures and geometries were totally
from a cupric nitrate precursor and isostructural tripodal
renowned as a viable alternative for complex thermally-
different. While HKUST-1 had all its carboxylic acids deproto-
mediated synthetic processes. This efficiency of microwave
nated and coordinated to Cu, CHB was assembled with a
heating as an advantageous alternative to conventional mono-deprotonated mode of Hip ligand, thus retaining a
heating in obtaining catalysts with uniform particle mor-
phologies, and excellent phase purities in a rapid and energy-
COOH as well as an OH group in the framework. An iso-struc-
tural variant of the CHB framework may not be obtained by
efficient manner encouraged us to synthesize CHB using simply altering the metal atom in the precursor salt. The exact
microwave energy.7h,8e,f In addition, microwave chemistry in
zinc analogue of CHB in its constituents, viz., ZHB (ZnHipBpy)
aqueous media is attractive and efficient due to the popularity
is non-isostructural with CHB, whose both the COOH groups
of water as a green solvent, the vapour pressure it imparts, and are deprotonatively coordinated with the metal center, thus
deficit in reactive COOH groups (Scheme 1).8d Therefore, it is
its capability to behave as a pseudo-organic solvent at elevated
temperatures.11 Herein, we report the microwave-assisted syn-
also clear that the generation of a reactive carboxyl group in
thesis of the carboxyl functional coordination polymer CHB in CHB in comparison with HKUST-1 is not merely resultant of
an aqueous medium for assessing its catalytic activity in the
solvent-free synthesis of cyclic carbonate from CO2 and
the pH or coordinative influence of the Bpy ligand, or even a
variation in the geometry of the copper precursor salt (mono
epoxide. The copper center of CHB is expected to operate versus tri-hydrate). In controlling the geometry for achieving a
1608 | Green Chem., 2014, 16, 1607–1616
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
task-specific MOF, a meticulous understanding of the nature
of the metal ions, ligands, solvents, pH and reaction con-
ditions are equally important.
The hydrothermal synthesis of the CHB catalyst (herein
denoted as CHB(S)) requires two days at 125 °C.8d Our own
reports of microwave-mediated methods as rapid means of
catalyst synthesis with better catalytic activities encouraged us
to adopt a microwave-assisted route for CHB synthesis.8e–g
However, for synthesizing a typical MOF under microwave radi-
ation, the literature suggests temperature programmed syn-
thetic techniques.7i Such syntheses however extend even up to
an hour for yielding the material, probably owing to the con-
trolled irradiation of microwaves while following the tempera-
ture program. Since the CHB synthesis was performed in an
aqueous medium, the maximum temperature of the reactant
mixture was auto-controlled around 100 °C, and hence a temp-
erature program would not be needed. Thus, the maximum
utilization of microwaves within a minimum time frame
was achieved by performing a CHB synthesis under sealed
conditions and constant microwave power in the aqueous
medium. To our delight, a very short reaction time of 6 min
with 100 W of microwave power was sufficient to facilitate the
synthesis of CHB crystals (denoted as CHB(M)) in exceptional
yields (96%).
Fig. 2 FT-IR spectra of CHB(S) and CHB(M).
Yet another shortcoming encountered in the catalysis of
conventional MOFs is the evacuation of solvent molecules
from the pores of MOFs. DMF or DEF, the most common sol-
vents for MOF synthesis, demand prolonged post synthetic
treatment, such as repeated boiling in methanol for purifi-
cation of the as-synthesized MOFs before activation and
catalysis.7e However, since CHB was synthesized totally in an
aqueous medium, purification was accomplished by soni-
cation in the aqueous medium, and activation by vacuum
drying. The absence of any lattice or coordinated water mole-
cules in the crystal of CHB further simplified the activation
process.
Fig. 3 Thermal gravimetric analyses (TGA) of CHB(M) and CHB(S).
crystallinity with respect to CHB(S), calculated from the relative
intensities of the central reflections, in the 2θ range 15–30°
(assuming 100% crystallinity of the solvothermal catalyst).
However, the higher intensity peak of CHB(M) than CHB(S) at
39° 2θ might be ascribed to some minor amounts of metal
oxides.10
The FT-IR spectrum of CHB(M) resembled that of CHB(S)
displaying their compositional similarity (Fig. 2). The carboxyl
Phase homogeneity of the CHB(M) and CHB(S) catalysts
was confirmed by comparing their powder X-ray diffraction
(PXRD) patterns with the single-crystal-simulated pattern
(Fig. 1). XRD analysis also revealed that CHB(M) exhibited 77%
and phenol O–H stretching frequencies of Hip (3450 cm−1
)
shifted to lower wavenumbers (3295 cm−1) in CHB owing to
the extensive H-bonds in CHB. The CvO stretching frequency,
characteristic of carboxylic acids, appeared at 1690 cm−1
.
However, two asymmetric (1600 and 1427 cm−1) and sym-
metric (1566 and 1390 cm−1) carboxylate stretches were
observed. The former carboxylate stretches, existent also in
Hip, correspond to COOH, whereas the latter ones having
Δυ(υasym − υsym) = 176 cm−1 belong to the chelated bidentate
carboxylate. The C–O stretching frequencies at 1300 cm−1
(reactive carboxyl) and 1237 cm−1 (coordinated carboxylate)
supplemented them further.
The thermal stability of CHB up to 300 °C from the TGA
curve points at similarities in the thermal stabilities of CHB(S)
and CHB(M) (Fig. 3). However a closer analysis of the degra-
dation curves around 250–300 °C reveals the better stability of
CHB(M). The residual mass of ∼18% at 600 °C could be corre-
lated to the oxidized inorganic part of the sample, presumably
Fig. 1 XRD patterns of CHB(S) and CHB(M) compared to the single
crystal simulated pattern.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 1607–1616 | 1609

View Article Online
Paper
Green Chemistry
Scheme 2 Representation of the cycloaddition of allyl glycidyl ether
(AGE) and CO2 to yield allyl glycidyl carbonate (AGC).
Table 1 Cycloaddition of allyl glycidyl ether and CO2 using various
catalysts
Surface
Selectivity Yield of
area (BET, Conversiona AGCa
AGCa
(%)
Entry Catalyst
m2 g−1
)
(%)
(%)
Fig. 4 FE-SEM images of (a) CHB(S) and (c) CHB(M); and their respect-
ive magnified images (b) and (d).
1
2
3
4
5
6
7
8
9
None
Cu(OAc)2·H2O
Hip
Bpy
CHB(S)
CHB(M)
—
—
—
—
1.3
0.4
0
1.7
0
66.1
—
84.6
84.4
90.3
63.5
56.2
95.7
00
1.1
—
24.5
71.7
77.4
40.7
6.3
Trace
29.0
84.9
85.7
64.1
11.3
91.3
cupric oxide.10a Further, CHB(M) had 1% higher residual mass
than CHB(S). The surface area measurements of CHB(M) and
CHB(S) and their respective isotherms are depicted in the ESI
(Fig. S2 and S3†). The BET surface areas of CHB(M) and
CHB(S) (0.4 m2 g−1 and 1.3 m2 g−1, respectively) were negligible
in comparison with HKUST-1 (1270 m2 g−1) and most of the
other functional MOFs.9c Examination of the textural features
using FE-SEM (Fig. 4) revealed that the CHB(M) catalyst (Fig.
4(a)) exhibits a morphology and particle dimensions compar-
able to CHB(S) (Fig. 4(b)). Magnified views of the crystals are
shown in Fig. 4(b) and (d).
ZnHipBpy (ZHB) 2.7
HKUST-1
1270b
0.4
CHB(M)c
87.4
Reaction conditions: Allyl glycidyl ether (AGE) = 18.6 mmol (2.2 mL),
0.3 mmol catalyst, catalyst to substrate ratio = 1.6 mol%, batch
reaction (except entry 9), 1.2 MPa CO2, 120 °C, 6 h, 600 rpm.
a Determined by GC. b Ref. 9b,c. c Semi-batch reaction. AGC
=
allyl glycidyl carbonate. Side product = 3-allyloxy-1,2-propanediol.
In order to compare the acid-base sites of the catalysts, the
initial rate calculation from the Knoevenagel and acetalization
reactions is a generally accepted experimental procedure.13
While comparing two similar catalysts, the higher the initial
rate of reaction, the higher the number of active sites. The
Knoevenagel condensation reaction is a base catalyzed reac-
tion, while the acetalization reaction is an acid catalyzed reac-
tion. Using CHB(S) or CHB(M) as catalyst, a Knoevenagel
condensation was performed between benzaldehyde and malo-
nonitrile to yield benzylidenemalononitrile, whereas an aceta-
lization reaction was carried out using benzaldehyde and
ethanol as the substrates (ESI†). The initial rates (r0) of 43.0 ×
103 and 40.0 × 103 mol min−1 g−1 were recorded for CHB(M)
and CHB(S), respectively. The slightly higher r0 value calcu-
lated for the Knoevenagel condensation with CHB(M) than
CHB(S) implies a slightly higher basicity of CHB(M) (Fig. S4†).
Ahn et al.15 observed using Temperature Programmed De-
sorption (TPD) that lone pair electrons in the metal coordinated
O- and N atoms of MOFs can act as base sites. Therefore, we
performed a CO2-TPD experiment with CHB(S) (representa-
tively) and found that it possess 29.5 mmol g−1 of base sites.
Hence, the slight basicity of the metal–carboxylate system
was verified. In the acetalization reaction, a higher r0 value of
16.5 × 103 mol min−1 g−1 was obtained for CHB(M), than the r0
value of 14.0 × 103 mol min−1 g−1 observed for CHB(S). These
observations could be attributed to slight amounts of metal
oxide in CHB(M).
The catalytic abilities of CHB were investigated in the cyclo-
addition reaction of CO2 using allyl glycidyl ether (AGE) as the
epoxide substrate (Scheme 2). In the absence of a catalyst,
the cycloaddition reaction (batch reaction) did not afford any
product even after 6 h of reaction using 18.6 mmol of AGE and
1.2 MPa of CO2 at 120 °C (Table 1, entry 1). Under similar con-
ditions, 1.6 mol% of the catalyst precursors viz., cupric acetate
monohydrate and 5-hydroxyisophthalic acid (Hip) yielded
almost no AGC (Table 1, entries 2 and 3). However, 29% con-
version of AGE with 84.6% AGC selectivity was observed with
the third precursor of CHB, viz., Bpy, probably due to its Lewis
basic pyridyl-nitrogens and homogeneity of reaction due to its
solubility in AGE (Table 1, entry 4). The heterogeneous cata-
lysts CHB(S) and CHB(M) provided a manifold increase of AGC
yield in comparison with any of its precursor materials
(Table 1, entries 5 and 6). Furthermore, among the CHB cata-
lysts in the batch conditions, the microwave-synthesized catalyst
CHB(M) was slightly superior in AGC yield than the solvo-
thermally synthesized CHB(S), probably owing to the metal oxide
impurities, as evident from the slightly higher residual mass
of CHB(M) observed in the TGA (Fig. 3).12 Thus, microwave
energy facilitates a more rapid catalyst synthesis route than
solvothermal, while maintaining the essential catalytic activity.
For experimental evidence of the promoting role of the
COOH group in cycloaddition, we compared CHB to the cata-
lytic activity of ZHB, a similar catalyst reported by our group
which is devoid of any COOH groups.7g ZHB, under similar
1610 | Green Chem., 2014, 16, 1607–1616
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Fig. 5 Effect of catalyst CHB(M) amount on the reactivity of AGE and
Fig. 6 Effect of reaction time on the reactivity of AGE and CO2 using
CO2 at 120 °C, 6 h, 600 rpm and PCO = 1.2 MPa.
1.6 mol% CHB(M) at 120 °C and 1.2 MPa PCO
.
2
2
reaction conditions to CHB, exhibited only 64.1% AGE conver-
sion and 63.5% selectivity of AGC (Table 1, entry 7). In
addition, a copper containing catalyst HKUST-1, which also
lacked COOH groups,14 was employed for the same reaction
under the same conditions. Despite its high porosity and CO2
adsorption capacities, HKUST-1 exhibited an AGE conversion
of only 11.3%, with 56.2% selectivity for AGC (Table 1, entry
8), which could be attributed to the absence of reactive COOH
groups. Hence the positive influence of the COOH group in
the cycloaddition reaction of AGE and CO2 is established. The
importance of the awareness of the coordination geometries,
ionization states and functional groups of a MOF rather than
merely relying on porosity, surface area or gas adsorption is
therefore signified.
Fig. 7 Effect of reaction temperature on the reactivity of AGE and CO2
using 1.6 mol% CHB(M) at 1.2 MPa PCO for 6 h.
2
A study of the reaction parameters with the CHB(M) catalyst
in batch conditions were conducted for understanding the
optimal conditions for AGC yield. A study of the effects of cata-
lyst amount (Fig. 5) was conducted in the range of
0.1–0.5 mmol of CHB(M) with 18.6 mmol of AGE (catalyst
mol% 0.5–2.68%) at 120 °C and 1.2 MPa CO2 pressure for 6 h.
A steady incremental increase in the yield of AGE was observed
from 0.53 to 1.6 catalyst mol% (0.1 to 0.3 mmol of catalyst). No
significant increase in the conversion was observed with
higher catalyst amounts of 0.4 and 0.5 mmol (catalyst mol%
2.15 and 2.68, respectively), probably because of the hindrance
in mass transfer through the H-bonded layers between the
active site and reagent, caused by the lesser dispersion of the
excess catalyst in the reaction mixture,1g or because of
the attainment of reaction equilibrium. 0.3 mmol (1.6% cata-
lyst to substrate ratio) was therefore identified as the optimal
catalyst amount under the aforementioned conditions (85.7%
AGE conversion and 90.3% AGC selectivity).
Fig. 8 Effect of CO2 pressure on the reactivity of AGE and CO2 using
1.6 mol% CHB(M) at 120 °C for 6 h.
The effect of reaction time on the CHB(M) catalyzed cyclo-
addition is depicted in Fig. 6. The conversion increased with
respect to reaction time and the highest conversion was obtained
at 6 h. Thereafter, the reaction did not proceed promisingly,
indicating that the optimum reaction time was 6 h for which
85.7% AGE conversion and 90.3% AGC selectivity was
obtained. The catalytic activity increased progressively from 60
to 120 °C, reaching the highest AGC yield (77.3%) at 120 °C
(Fig. 7), thereafter no noticeable increase in activity was found.
The effect of CO2 pressure on the yield of AGE is depicted in
Fig. 8. AGE conversion increased with an increase in CO2
pressure from 0.4 MPa to 1.2 MPa. This could be attributed to
an increased solubility of CO2 in the reaction mixture at higher
pressures and the subsequently improved CO2 concentration
which shifts the reaction equilibrium favoring an accelerated
formation of AGC. CHB(M) exhibited its highest catalytic
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 1607–1616 | 1611

View Article Online
Paper
Green Chemistry
Table 2 Cycloaddition of CO2 with various epoxides using the CHB(M)
Table 3 Influence of TBAB in the cycloaddition of AGE and CO2
catalyst
Entry
Catalyst
% Conversion
% Selectivity
Conv
[Sel] (%)
Yield
(%)
1
2
3
4
5
6
7
8
None
0
0
—
Entry
1
Epoxide
Cyclic carbonate
CHB(M)a
Trace
2.4
24.3
24.1
89.2
36.9
88.5
TBABa
99.9
99.9
68.2
99.9
91.3
99.9
75.2 [83.3]
80.7 [89.5]
62
TBAB/CHB(M)a
CHB(M)b
TBAB/CHB(M)b
TBAC/CHB(M)b
TBAI/CHB(M)b
2
3
72
Reaction conditions: 18.6 mmol AGE, 1.2 MPa PCO , 6 h, semi-batch.
2
65.6 [86.3]
56
a At 27 °C, 0.3/0.3 mmol CHB(M)/TBAB. b 80 °C, 1.6 mol% catalyst,
TBAB = 1.6 mol%, wherever used.
4
5
85.9 [84.4]
85.1 [83.5]
73
71
The activation of epoxide at mild reaction conditions is
widely regarded as an attractive research goal.7e Therefore, the
possibility of a synergistic enhancement in catalytic activity of
CHB(M) by a co-catalyst containing a strong nucleophilic anion
for a better cyclic carbonate yield was estimated worthwhile.15
Accordingly, quaternary ammonium salts were engaged as a co-
catalyst with CHB(M) under semi-batch conditions (Table 3). At
room temperature, a ten-fold increase in yield of AGC was
observed with the CHB(M)/TBAB system, in comparison to the
summated AGC yields with individual CHB(M) and TBAB cata-
lysts (entries 2 and 3). This clearly demonstrates the synergistic
catalysis by the CHB(M)/TBAB system. Yet another surprising
result was the more superior AGC yield (in terms of conversion
and selectivity) with the CHB(M)/TBAB system at room temp-
6
7
69.8 [80.6]
19.7 [52.4]
56
10
Reaction conditions: 0.3 mmol CHB(M), 18.6 mmol epoxide, PCO
1.2 MPa; 120 °C, 6 h, 600 rpm. Conv. = Conversion, Sel = Selectivity,
Yield = Isolated yield.
=
2
activity under 1.2 MPa of CO2 pressure and it decreased erature than CHB(M) alone at 80 °C (entries 4 and 5). Above
slightly at 1.4 MPa. It must be inferred that too high a CO2 all, the high AGC yield of 89.2% (99.9% selectivity) at 80 °C
pressure may retard the interaction between AGE and the cata- with the CHB(M)/TBAB system clearly demonstrates its high
lyst by a dilution effect, resulting in a lower conversion.1g Thus, catalytic activity under optimised reaction conditions (entry 6).
the optimum conditions for obtaining maximum yields using
In an attempt to study the effect of anions in this cyclo-
the CHB catalyst were identified as 6 h, 120 °C, and 1.2 MPa addition, the reactions were performed with tetrabutyl
CO2, contributing to an AGE conversion of 85.7% and a selecti- ammonium chloride (TBAC) and tetrabutylammonium iodide
vity of 90.3% for AGC under batch conditions. Cycloaddition (TBAI) co-catalysts. While the effect of nucleophilicity is
at the same conditions in a semi-batch operation is desirable expected to exhibit activities in the order of I > Br > Cl, steric
for an understanding of the AGC yield with CO2 pressure factors may reverse the order to Cl > Br > I.16 However in
maintained throughout. Interestingly, an AGC yield of 87.4% a typical MOF, say for example MOF-5, the combined effect
was obtained in 6 h at 120 °C and 1.2 MPa CO2 in semi-batch equilibrates to the order Br > I > Cl; explained by the diffusion
condition (Table 1, entry 9).
of the large iodide being hampered by the restriction of the
The scope of CHB(M) in catalyzing epoxide–CO2 cycloaddi- pore size of the MOF.7b In the present system also, TBAC
tions was explored by extending the catalysis to a variety of exhibited a conversion of 36.9% (91.3% selectivity for AGC),
epoxides; the results are summarized in Table 2. The internal which is much less than that with the bromide anion (entry 7).
epoxide cyclohexene oxide gave only comparatively low conver- Surprisingly, the reaction with the iodide containing co-
sions, which could be ascribed to the hindrance in anionic catalyst (TBAI, entry 8) exhibited an AGC yield similar to that
attack posed by the steric crowding of the cyclohexene ring with the bromide anion (of TBAB). This sheds light on the
(entry 7). The comparatively lower conversion of styrene open active sites of the CHB catalyst that offer the bromide
oxide is attributable to the low reactivity at its β-carbon center species easy access in comparison to the difficultly of iodide in
(entry 6). AGE and phenyl glycidyl ether (PGE), in contrast, its diffusion into the pores of higher dimensional MOFs.
gave comparatively better conversions (entries 4 and 5). However, the expected superior catalytic activity of iodide over
Recently our group, using experimental and DFT calculations, bromide might have been nullified by the steric hindrance
explained that the enhanced activity of AGE with COOH- experienced by iodide around the active sites.16
containing catalysts is due to the H-bonding interactions
In order to get a clearer understanding of the CHB(M)/
between the side chain ether oxygen of AGE and the COOH TBAB catalysis, the yield of AGC at various temperatures in
group of the catalyst.8c,g
different time intervals was studied (Fig. 9). The reaction
1612 | Green Chem., 2014, 16, 1607–1616
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Fig. 10 Mercury diagram of the molecular ribbon in CHB. O atoms of
carboxyl and hydroxyl groups, and Cu atom represented as ball model
(H-atoms omitted for clarity).
Fig. 9 Time variant yield of AGC at various temperatures (room temp-
erature, 50 °C and 80 °C) with 18.6 mmol AGE, 1.2 MPa PCO , and
2
interaction between the COOH group of CHB and the ether
oxygen of AGE (H-bonding), thereby holding AGE more
strongly to the proximity of the reactive site (Scheme S1,
ESI†).8c,g Concomitantly, the Br− of TBAB attacks the least hin-
dered carbon atom, thereby resulting in epoxide ring opening.
Following this, the O− of the ring-opened intermediate attacks
the carbon atom of CO2, forming the carbonate complex. In
the succeeding step, ring closure occurs by intra-molecular car-
bonate O− attack at the C–Br carbon whereby AGC is gener-
ated, together with the liberation of TBAB and the regeneration
of the original structure and geometry of CHB(M).
1.6 mol% CHB(M)/TBAB semi-batch operation.
In the absence of TBAB, the mechanism proceeds via a
CHB(M)–AGE interaction that undergoes a ring-opening by the
O− atom of the polarized CO2 molecule (polarized either by
base sites of CHB or epoxide substrate) at the β-carbon of the
epoxide (Scheme S2, ESI†). The higher activity of CHB in com-
parison with the MOF precursor materials in isolation is justi-
fied by the aforesaid CHB(M)–AGE interaction (Table 1, entries
2 and 3). The CHB(M) catalyzed cycloaddition that demands
elevated temperatures may be ascribed to the barrier in cross-
ing the transition state for activation of CO2 either by the
carbonyl group of CHB or AGE.
Scheme 3 Proposed mechanism for the cycloaddition of AGE and CO2
using CHB/tetrabutylammonium halide. (Bu4NX = tetrabutylammonium
halide, where X = Cl/Br/I).
In several MOFs, like Cu2(BTC)3/HKUST-1 and ZIF-8, re-
cyclability issues were documented for the cycloaddition of
epoxide and CO2,2,7i resulting from local structural disorders
and/or active site/pore blocking by residual carbonaceous
deposits. Despite such observations for CHB(M) at 120 °C, it is
worthwhile to note that recycling was successful with the CHB(M)/
TBAB system at 80 °C and 1.2 MPa CO2 pressure. Powder
XRD patterns of the recycled catalyst were similar to the fresh
CHB(M), with 97% retention of crystallinity, implying stability
of CHB(M) under the aforementioned conditions (Fig. 11).
proceeds considerably even at room temperature. However,
at 80 °C, where the AGC yield was quite high, 6 hours was
the optimal time for reaction. Supplemented with the CO2
pressure study (Fig. S5, ESI†), it could be concluded that with
1.6 mol% TBAB and CHB(M), the optimum conditions were
80 °C, 1.2 MPa CO2 pressure and 6 h, yielding 89.1% AGC.
A plausible mechanism of the cycloaddition reaction of
AGE in the presence of CHB(M), co-catalyzed by TBAB, is illus-
trated in Scheme 3. The Mercury diagram of CHB (Fig. 10) gen-
erated from the crystallographic information file (CIF) of the
Cambridge Crystallographic Data Center (CCDC deposit
number 657329), depicts the reactive functional groups and
Cu in CHB (depicted using a ball-and-stick model). Initially,
the epoxide oxygen of AGE is activated by the Cu(II) center of
CHB(M). This attack takes place with substitution, breaking
the vO–Cu bond, maintaining the octahedral geometry.
The AGE–CHB interaction is further strengthened by the
Interestingly, an increase in BET surface area (to 7.5 m2 g−1
)
was observed for CHB(M) after the cycloaddition (ESI,
Fig. S6†). This may be ascribed to the partial denaturing
of intra-molecular H-bonds between 1-D chains of CHB
upon cycloaddition, an imperfect elasticity of hydrogen bonds
during substrate diffusion/effusion between the H-bonded
layers,17 or catalyst milling arising from stirring during the
course of the reaction. Nonetheless, the covalent or coordinate
bonds of CHB were unaffected throughout the cycloaddition,
as is evident from PXRD.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 1607–1616 | 1613

View Article Online
Paper
Green Chemistry
Fig. 12 Catalyst recycling at 80 °C, 1.2 MPaCO , 6 h; 1.6 mol% CHB(M)/
2
TBAB in AGE (99% selectivity maintained).
Fig. 11 Powder XRD patterns of fresh CHB(M) and recycled CHB(M).
A hot filtration test by Sheldon et al. was conducted to test
the heterogeneous nature of the catalyst in the cycloaddition
reaction (section 4, ESI†).18 The cycloaddition was carried out
with 1.6 mol% CHB(M), AGE and CO2 at 1.2 MPa CO2 and
100 °C until a 49.6% conversion of AGE was achieved (at 5 h).
The contents were filtered while hot; 0.2 μL of the filtrate was
used for GC analysis (in CH2Cl2) and the remaining filtrate
was fed back to the same conditions of temperature, CO2 and
stirring. Sampling was performed every 1 hour thereafter. No
increase in conversion was observed in the filtered content,
implying the absence of any catalytically active homogeneous
species in the filtrate (Fig. S7, ESI†). Since there is no conver-
sion in the filtered sample compared to the conversion
observed in the original sample, there are no homogeneous
species catalyzing the reaction in solution. The hot filtration
test is regarded as more of a test for homogeneous catalytic
contributions to the rate than a test for catalyst leaching. In
order to have an accurate determination of catalyst leaching,
the filtrate after reaction was therefore analyzed using
ICP-OES, which showed only small traces of Cu, counting as
less as 0.5 ppm in comparison to 30–50 ppm metal detected
with other MOFs such as MILs.7e
The recycled CHB(M) catalyst from the CHB(M)/TBAB reac-
tion (Table 3, entry 6) was further employed for successive
runs with freshly added co-catalyst. A slight rise in AGC yield
was observed after the first run. This could be attributed to the
enhanced AGE access as a consequence of intra-molecular H-
bonding rupture. This is supplemented by the increased
surface area of the recycled CHB with respect to fresh CHB(M).
The selectivity and yield of AGE were not significantly affected
in further cycles (until the 4th reuse, Fig. 12). The chemical
stability of CHB(M) after each cycle was confirmed by FT-IR
spectroscopy (Fig. 13).
Fig. 13 FT-IR spectra of the recycled CHB(M) catalyst.
synergistic role of reactive hydrogen-bonding functional
groups such as the carboxyl group with metal centers in
enhancing the cycloaddition of epoxide and CO2 under
solvent-free conditions. A thorough observation of the topolo-
gies and coordination modes of the coordination polymers
was demonstrated as a successful strategy for accomplishing
an assigned task, instead of functionalizing a higher dimen-
sional, porous, MOF template. The role of the reactive COOH
group in a MOF for the successful cycloaddition of epoxides
and CO2 was proposed and demonstrated for the first time.
Microwave irradiation, used for accomplishing rapid catalyst
synthesis, instigated a better catalytic activity than the hydro-
thermally synthesized catalyst in the synthesis of AGC from
CO2 and AGE. Furthermore, excellent selectivities of AGC were
obtained even under ambient reaction temperatures in the
presence of the quaternary ammonium salt co-catalyst TBAB.
Substantial yields of AGC were obtained at 80 °C and 1.2 MPa
CO2 pressure in 6 h, with 99.9% selectivity. The catalyst was
successfully recycled up to four times without the degradation
of any covalent/coordinate linkages of the catalyst framework.
The high catalytic activity exhibited by CHB in comparison
with a typical MOF with high porosity, surface area and gas
adsorption property points at the significance of a careful
Conclusions
The aqueous microwave-assisted synthesized coordination
polymer {Cu(Hip)2(Bpy)}n was employed to demonstrate the
1614 | Green Chem., 2014, 16, 1607–1616
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
watch of coordination geometries and ionization in MOFs. A CHB(M), were obtained in a >96% yield. The activation was
meticulous design or judicious selection of MOFs that could accomplished by a similar procedure used for CHB(S). The
aptly accommodate active sites and functional groups will material was ground prior to use in catalysis.
serve a prospective role in attaining excellent catalytic pro-
perties, underlining the so-called versatility of coordination
Cycloaddition of epoxide and CO2
polymers and MOFs in task-oriented catalytic applications.
All syntheses were carried out in a 25 mL stainless steel reactor
Employing water in association with microwave energy may
equipped with a magnetic stirrer. Procedure for the synthesis
serve as a green substitute for organic solvents in MOF
of allyl glycidyl carbonate (AGC) from allyl glycidyl carbonate
syntheses.
(AGE) is as follows. An appropriate amount of activated, finely
ground catalyst was added to the reactor containing AGE
(18.6 mmol). The reactor was pressurized with CO2 to the
Experimental section
required pressure at room temperature, brought to the desired
Materials
temperature, and stirred at 600 rpm. After reaction com-
pletion, the reactor was cooled and excess CO2 was carefully
vented off. Dichloromethane (2 mL) was added to the product
mixture and filtered. The filtrate was subsequently analyzed
using gas chromatography (GC, HP 6890, Agilent Techno-
logies) to determine conversion, selectivity, and yield. The
Copper(II) acetate monohydrate (Cu(OAc)2·H2O; >98%), and
4,4′-dipyridyl (Bpy; 98%) were purchased from Aldrich, Korea
and used as received. 5-Hydroxyisophthalic acid (Hip; 99%,
Acros Organics) was purchased from Fisher Scientific.
Doubly distilled water was used for catalyst synthesis. Allyl gly-
cidyl ether (AGE) and all other epoxides were procured from
Aldrich. CO2 of 99.999% purity was used. Chemicals for the
Knoevenagel condensation and acetalization viz. benzaldehyde
(>99.5%), malononitrile (>99%), ethanol (>99.5%) were
purchased from Aldrich, whereas 1,4-dioxane was purchased
from Junsei Chemical Co. Ltd.
1
product AGC was further confirmed by H NMR and 13C NMR
experiments. 1H NMR (500 MHz, CDCl3) δ (ppm): 5.23–5.14
(2H), 5.86 (1H), 4.82–4.76 (1H), 4.49–4.45 (2H), 4.5 (2H),
4.82–4.9 (2H); 13C NMR (75 MHz, CDCl3) δ (ppm): 61.7, 71.8,
72, 75.1, 115.4, 131.7, 154.2; GC-MS: m/z calcd for C7H10O4:
158.06. Found: m/z 157.9 (ESI, Fig. S1†). The ring-opened side
product, 3-allyloxy-1,2-propanediol, was also identified using
GC-MS (ESI, Fig. S1†).
Synthesis of catalyst
Solvothermal synthesis: CHB(S). Cu(Hip)2(Bpy) was hydro-
thermally synthesized by a procedure similar to that reported
by Feller and Cheetham.8d To an aqueous solution of Cu-
(OAc)2·H2O (9 mmol) in water (30 mL, doubly distilled) were
added 5-hydroxyisophthalic acid (18 mmol) and 4,4′-bipyridine
(9 mmol); the resultant mixture was stirred for 30 min in room
temperature (600 rpm). The contents were then transferred to
a 100 mL Teflon-lined autoclave, sealed and heated at 125 °C
for 48 h in a convection oven. The contents were then regres-
Acknowledgements
This study was supported by the National Research Foundation
of Korea (2012-001507) and the Global Frontier Project. The
authors are grateful to KBSI and the Central Lab facility
of Pusan National University for performing characterization
studies.
sively cooled to room temperature at a rate of 0.1 °C min−1
.
Deep purple crystals of CHB, denoted as CHB(S), were
obtained in a 90% yield from the product mixture by washing
with water under sonication. The catalyst was activated to
remove any physically adsorbed water molecules by vacuum
drying at 100 °C for 4 h (heating rate 5 °C min−1). The crystals
were ground to a fine powder for catalytic reaction.
Microwave synthesis: CHB(M). A rapid synthesis of CHB was
accomplished by microwave irradiation of the reaction mixture
in a 40 mL Pyrex glass reactor tube. The multimode microwave
reactor (KMIC-2 KW), having a continuously adjustable power
source (range 0–2 kW) with a 3-stub tuner, operates at a fre-
quency of 2.450 GHz. Cu(OAc)2·H2O (0.3 mmol) was comple-
tely dissolved in water (10 mL, doubly distilled), followed by
the addition of 4,4′-bipyridine (0.3 mmol) and 5-hydroxy-
isophthalic acid (0.6 mmol) in water (5 mL). The contents were
transferred to the Pyrex glass tube, sealed and irradiated with
a microwave power of 100 W for 6 min. The reaction mixture
then remained in the reactor for cooling down spontaneously
to room temperature, and purple crystals of CHB, denoted as
Notes and references
1 (a) H. S. Kim, J. J. Kim, B. G. Lee, O. S. Jung, H. G. Jang and
S. O. Kang, Angew. Chem., Int. Ed., 2000, 39, 4096;
(b) D. Tian, B. Liu, Q. Gan, H. Li and D. J. Darensbourg,
ACS Catal., 2012, 2, 2029; (c) K. R. Roshan, G. Mathai,
J. Kim, J. Tharun, G. A. Park and D. W. Park, Green Chem.,
2012, 14, 2933; (d) Z. Zhang, Y. Xie, W. Li, S. Hu, J. Song,
T. Jiang and B. Han, Angew. Chem., Int. Ed., 2008, 47, 1127;
(e) J. C. Choi, L. N. He, H. Yasuda and T. Sakakura, Green
Chem., 2002, 4, 230; (f) J. Bayardon, J. Holz, B. Schäffner,
V. Andrushko, S. Verevkin, A. Preetz and A. Börner, Angew.
Chem., Int. Ed., 2007, 46, 5971; (g) A. C. Kathalikkattil,
J. Tharun, R. Roshan, H. G. Soek and D. W. Park, Appl.
Catal., A, 2012, 447–448, 107; (h) C. M. Miralda,
E. E. Macias, M. Zhu, P. Ratnasamy and M. A. Carreon, ACS
Catal., 2012, 2, 180; (i) M. Yoshida and M. Ihara, Chem.–
Eur. J., 2004, 10, 2886; ( j) Y. Himeda, N. Onozawa-
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 1607–1616 | 1615

View Article Online
Paper
Green Chemistry
Komatsuzaki, H. Sugihara and K. Kasuga, J. Am. Chem.
Soc., 2005, 127, 13118.
8 (a) J. Sun, J. Y. Ren, S. J. Zhang and W. G. Cheng, Tetrahe-
dron Lett., 2009, 50, 423; (b) J. Qu, C. Y. Cao, Z. F. Dou,
H. Liu, Y. Yu, P. Li and W. G. Song, ChemSusChem, 2012, 5,
652; (c) J. Tharun, G. Mathai, A. C. Kathalikkattil,
R. Roshan, J. Y. Kwak and D. W. Park, Green Chem., 2013,
15, 1673; (d) R. K. Feller and A. K. Cheetham, Dalton Trans.,
2008, 2034; (e) J. Tharun, D. W. Kim, R. Roshan, Y. Hwang
and D. W. Park, Catal. Commun., 2013, 31, 62;
(f) D. W. Jung, D. A. Yang, J. Kim, J. Kim and W. S. Ahn,
Dalton Trans., 2010, 39, 2883; (g) J. Tharun, G. Mathai,
R. Roshan, A. C. Kathalikkattil, K. Bomi and D. W. Park,
Phys. Chem. Chem. Phys., 2013, 15, 9029.
9 (a) X.-T. Wu, Controlled Assembly and Modification of
Inorganic Systems, Springer, Heidelberg, Germany, 2009;
(b) J. Moellmer, A. Moeller, F. Driesbach, R. Glaeser and
R. Staudt, Microporous Mesoporous Mater., 2011, 138, 140;
(c) K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald,
E. D. Bloch, Z. R. Herm, T. H. Bae and J. R. Long, Chem.
Rev., 2012, 112, 724.
2 J. Song, Z. Zhang, S. Hu, T. Wu, T. Jiang and B. Han, Green
Chem., 2009, 11, 1031.
3 (a) M. North, R. Pasquale and C. Young, Green Chem., 2010,
12, 1514; (b) P. Jessop, T. Ikariya and R. Noyori, Chem. Rev.,
1995, 95, 259; (c) G. Shaikh and S. Sivaram, Chem. Rev.,
1996, 96, 951.
4 (a) R. L. Paddock and S. T. Nguyen, J. Am. Chem. Soc., 2001,
123, 11498; (b) J. Wang, J. Wu and N. Tang, Inorg. Chem.
Commun., 2007, 10, 1493.
5 (a) M. M. Dharman, J.-I. Yu, J.-Y. Ahn and D. W. Park, Green
Chem., 2009, 11, 1754; (b) A. Ghosh, P. Ramidi, S. Pulla,
S. Z. Sullivan, S. L. Collon, Y. Gartia, P. Munshi, A. S. Biris,
B. C. Noll and B. C. Berry, Catal. Lett., 2010, 137, 1;
(c) K. Yamaguchi, K. Ebitani, T. Yoshida, H. Yoshida and
K. Kaneda, J. Am. Chem. Soc., 1999, 121, 4526.
6 (a) A. C. Kathalikkattil, K. K. Bisht, N. Aliaga-Alcalde and
E. Suresh, Cryst. Growth Des., 2011, 11, 1631; (b) J. Y. Lee,
O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and 10 (a) C. M. Calleja, J. A. Botas, M. G. Orcajo and
J. T. Hupp, Chem. Soc. Rev., 2009, 38, 1450;
(c) R. J. Kuppler, D. J. Timmons, Q.-R. Fang, J.-R. Li,
T. A. Makal, M. D. Young, D. Yuan, D. Zhao, W. Zhuang
and H.-C. Zhou, Coord. Chem. Rev., 2009, 253, 3042.
M. S. Sanchez, J. Porous Mater., 2010, 17, 91;
(b) A. Lamberti, M. Destrob, S. Biancoa, M. Quaglioa,
A. Chiodoni, C. F. Pirri and C. Gerbaldi, Electrochim. Acta,
2012, 70, 62.
7 (a) H. Deng, C. J. Doonan, H. Furukawa, R. B. Ferreira, 11 V. Polshettiwar and R. S. Varma, Aqueous Microwave Chem-
J. Towne, C. B. Knobler, B. Wang and O. M. Yaghi, Science, istry, RSC Publishing, Cambridge, 2010.
2010, 327, 846; (b) W. Kleist, F. Jutz, M. Maciejewski and 12 J. D. Chen and R. A. Sheldon, J. Catal., 1995, 153, 1.
A. Baiker, Eur. J. Inorg. Chem., 2009, 3552; (c) Y. J. Kim and 13 M. J. Climent, A. Corma, P. D. Frutos, S. Iborra, M. Noy,
D. W. Park, J. Nanosci. Nanotechnol., 2013, 13, 2307;
(d) M. Zhu, D. Srinivas, S. Bhogeswararao, P. Ratnasamy 14 J. L. Zhuang, D. Ceglarek, S. Pethuraj and A. Terfort, Adv.
and M. A. Carreon, Catal. Commun., 2013, 32, 36; Funct. Mater., 2011, 21, 1442.
(e) O. V. Zalomaeva, A. M. Chibiryaev, K. A. Kovalenko, 15 J. Kim, S. N. Kim, H. G. Jang, G. Seo and W. S. Ahn, Appl.
O. A. Kholdeeva, B. S. Balzhinimaev and V. P. Fedin, Catal., A, 2013, 453, 175.
J. Catal., 2013, 298, 179; (f) Y. Ren, Y. Shi, J. Chen, S. Yang, 16 H. Y. Ju, M. D. Manju, K. H. Kim, S. W. Park and
C. Qia and H. Jiang, RSC Adv., 2013, 3, 2167; D. W. Park, J. Ind. Eng. Chem., 2008, 14, 157.
(g) A. C. Kathalikkattil and D. W. Park, J. Nanosci. Nanotech- 17 Y. Cheng, H. Kajiro, H. Noguchi, A. Kondo, T. Ohba,
A. Velty and P. Concepcion, J. Catal., 2010, 269, 140.
nol., 2013, 13, 2230; (h) D. A. Yang, H. Y. Cho, J. Kim,
S. T. Yang and W. S. Ahn, Energy Environ. Sci., 2012, 5,
Y. Hattori, K. Kaneko and H. Kanoh, Langmuir, 2011, 27,
6905.
6465; (i) E. E. Macias, P. Ratnasamy and M. A. Carreon, 18 R. A. Sheldon, M. Wallau, I. W. C. E. Arends and
Catal. Today, 2012, 198, 215.
U. Schuchardt, Acc. Chem. Res., 1998, 31, 485.
1616 | Green Chem., 2014, 16, 1607–1616
This journal is © The Royal Society of Chemistry 2014