Amino acid/KI as multi-functional synergistic catalysts for cyclic carbonate synthesis from CO2 under mild reaction conditions: A DFT corroborated study

Article abstract of DOI:10.1039/c3dt52830h

Naturally occurring amino acids were identified as efficient co-catalysts for the alkali metal halide-mediated synthesis of cyclic carbonates from carbon dioxide and epoxides under mild, solvent free reaction conditions. The binary system of histidine/pot

Full text of DOI:10.1039/c3dt52830h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Amino acid/KI as multi-functional synergistic catalysts
for cyclic carbonate synthesis from CO under mild
reaction conditions: a DFT corroborated study†
2
Cite this: Dalton Trans., 2014, 43,
2023
a
a
a
Kuruppathparambil Roshith Roshan, Amal Cherian Kathalikkattil, Jose Tharun,
a
b
a
Dong Woo Kim, Yong Sun Won and Dae Won Park*
Naturally occurring amino acids were identified as efficient co-catalysts for the alkali metal halide-
mediated synthesis of cyclic carbonates from carbon dioxide and epoxides under mild, solvent free reac-
tion conditions. The binary system of histidine/potassium iodide gave an appreciable turnover number of
535 for propylene oxide in 3 h. Detailed studies evaluating a variety of amino acids revealed that the basic
amino acids afforded better conversion rates. The formation of a seven membered ring involving the zwit-
terionic ends of the amino acid, the metal halide, and the epoxide was considered to accelerate the cata-
lysis rate. Density functional theory calculations were performed for the first time on amino acid co-
catalyzed cycloaddition to provide further evidence for this hypothesis. The iodide ions of the alkali metal
halide displayed excellent synergism with the hydrogen bonding groups of the amino acids in the pro-
duction of cyclic carbonates, whereas bromide and chloride anions functioned less efficiently. The utili-
zation of amino acids to enhance the catalytic activity of the cheap and eco-friendly alkali metal halides
for cyclic carbonate synthesis represents a cost-effective, greener route towards the chemical fixation of
carbon dioxide.
Received 9th October 2013,
Accepted 3rd November 2013
DOI: 10.1039/c3dt52830h
www.rsc.org/dalton
Introduction
2
Even though non-toxic, the overabundance of CO is proving
to be a significant challenge to the ecological balance of the
planet, in particular through global warming. Interestingly,
this most oxidized form of carbon could be employed as a
user-friendly alternative C1 feedstock to toxic chemicals such
Scheme 1 Cycloaddition of carbon dioxide with epoxides.
2 2
as COCl and CO. Thus, the chemical fixation of CO has
become apparently an area of competent research, fuelled by
producing cyclic carbonates has been successfully developed
the multiple objectives of reducing atmospheric CO concen-
2
in the last two decades. The most effective of these were found
tration on a greener perspective and providing a cheap raw
material for the synthesis of industrially important chemicals.
3
a–d
4a–c
to be organometallic and salen complexes,
metal oxides,
5
a–m
6a–f
supported phase catalysts,
ionic liquids,
Even though these catalyst systems
and metal
The reaction of CO with epoxides, an atom-efficient synthesis
2
7a–g
organic frameworks.
for cyclic carbonates, is a dynamic field of interest in the above
1
a–d
have shown appreciable activities for this particular reaction,
the quest to effectively exploit naturally occurring raw
materials as catalysts is on the verge of seeking an extra edge.
The extensive research that has been carried out with regard to
exploring the mechanistic aspects behind the epoxide–CO2
cycloaddition has identified the synergistic role played by
halide ions and hydrogen bonding groups in bringing about
context (Scheme 1).
Cyclic carbonates are industrially
important chemicals widely employed as the monomer units
of polycarbonates, as aprotic solvents, and as electrolytes in
1
d,2a
lithium ion batteries.
A plethora of catalytic systems for
a
Division of Chemical and Biomolecular Engineering, Pusan National University,
Busan 609-735, Republic of Korea. E-mail: dwpark@pusan.ac.kr;
Fax: +82 51 512 8563; Tel: +82 510 2399
5
c,8a–h
high selectivity towards cyclic carbonates.
This has led to
the deployment of a large number of bi-functional catalysts
b
Division of Chemical Engineering, Pukyung National University, Busan 608-737,
8a
such as the hydroxyl rich renewable biopolymer, cellulose,
Republic of Korea
8
b
and its oligomeric analogues, gamma and beta cyclodextrins,
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c3dt52830h
8
d
1
curcubit, etc. with alkali metal halides.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 2023–2031 | 2023

View Article Online
Paper
Dalton Transactions
Aspiring to expand the scope of this eco-friendly CO
2
fix- conditions to yield propylene carbonate (PC). Hence, herein,
ation, chemists have delved into nature’s tool box and selected we have employed a series of naturally occurring AAs with MX
amino acids (AAs) as environmentally benign catalysts to effect (M = Li, Na, K, and X = Cl, Br, I) as a catalytic system, and
9
f
the chemical transformation of CO
2
.
AAs are the monomer studied the reaction parameters under solvent free conditions.
units of enzymes, the biological catalysts, and are enriched In addition, an attempt to explore the theoretical aspects
with its natural hydrogen bonding groups. Owing to the versa- behind the mechanistic pathways was performed using
tility of their specific stereochemical orientations, naturally quantum mechanical calculations (DFT).
occurring AAs have been reported to facilitate highly efficient
9
g
9a
asymmetric organic synthesis. Qi et al. first reported the
Results and discussion
The synergistic role of I− ions and the hydrogen bonding
groups of naturally occurring AAs in the cycloaddition of PO
2
with CO was investigated (Table 1). On the basis of their
inherent charge and polar-non polar nature, glycine and four
different kinds of AAs: (i) basic, (ii) acidic, (iii) AAs with polar
uncharged side chains, and (iv) AAs with hydrophobic side
chains, were subjected to the analysis (Scheme 2). All of the
catalytic ability of such molecules for cyclic carbonate syn-
thesis from CO and epoxides. Although they achieved a propy-
2
lene carbonate (PC) yield of almost 100%, comparatively harsh
2
reaction conditions were required (6 MPa CO pressure, 48 h,
and an organic solvent, CH Cl ). In subsequent work, the
2
2
9
b
same group
reported a modification of the procedure,
whereby the need for a solvent was circumvented by the use of
supercritical CO2 (8 MPa), with histidine as the catalyst.
However, this method still required 48 h for completion.
Similar activity was achieved using polystyrene immobilization
9
c
of AAs, which rendered reusability. This process again Table 1 Catalytic test runs on the solvent less cycloaddition of PO
a
with CO
2
required relatively harsh reaction conditions of 9 MPa CO
pressure, 130 °C, and 24 h, although only a small amount of
catalyst was necessary (0.6 mol%). An interesting advance in Entry Catalyst
2
b
b
PO conversion
(%)
PC yield
(%)
the AA-catalyzed cycloaddition of epoxides with CO
2
was dis-
1
2
3
4
5
6
7
8
9
None
—
4
—
3
played by the synthesis of ionic liquids, engaging AA moieties
KI
9
d
c
as the anions. Wu et al. reported that AA ionic liquids
AAILs), possessing an alkyl imidazolium cation and AA anion,
Glycine
3
3
c
Glycine/KI
29
2
28
2
(
d
d
d
d
e
Alanine
could serve as a single component bi-functional catalyst for
the production of cyclic carbonates. The reaction proceeded
under relatively mild conditions of 4 MPa CO for 8–12 h reac-
2
Alanine/KI
36
21
19
51
57
31
27
82
75
99
57
63
59
35
20
19
50
56
30
26
81
74
98
56
62
58
Tyrosine/KI
Tryptophan/KI
Serine/KI
e
f
tion time; however, a large amount of catalyst was required 10
Threonine/KI
Aspartic acid/KI
Glutamic acid/KI
Lysine/KI
Arginine/KI
Histidine/KI
Imidazole/KI
Boc-histidine/KI
Histidinemethyl ester
dihydrochloride/KI
9
e
1
1
1
1
2
3
(3 mol%) to obtain a high PC yield. Recently, Gong et al. suc-
f
cessfully developed an L-proline-based ionic liquid, which
efficiently catalyzed cyclic carbonate synthesis under atmos- 14
pheric CO2 pressure in 24 h. Most recently our group has
reported the catalytic ability of quaternized glycine in effecting
g
g
g
9
f
15
1
1
6
8
h
h
the CO
Alongside these developments, alkali metal halides (MX)
have been identified as cheap catalysts for epoxide–CO cyclo-
2
–epoxide cycloaddition.
19
a
2
Catalyst amount: 0.08 mmol each (0.2 mol%), amino acid : KI = 1 : 1,
b
4
2.8 mmol PO, 120 °C, 1 MPa, 3 h. Calculated from GC (biphenyl
addition although they have been shown to be less effective
under mild reaction conditions (CO2 pressures below
c
d
as internal standard). Neutral amino acid. Neutral amino acid with
e
hydrophobic amino acid chain. Neutral amino acid with polar side
1
0a–c
f
g
h
2
MPa).
extent through the use of organic bases that possess either a
This has been circumvented to an appreciable chain. Acidic amino acid. Basic amino acid. Protected amino acids.
1
1a–f
secondary or tertiary amine moiety as co-catalysts.
Insights
gathered from previous reports were suggestive of the synergis-
tic roles of the basic moieties of the organic base and the
nucleophilic anion of the MX to be the reason for the
enhanced catalytic activity for the cycloaddition. Naturally
occurring organic acids such as formic acid and its higher ana-
logues have also been positively identified to act as efficient
8
f
2
co-catalysts for KI in the CO –epoxide cycloaddition. This
prompted us to investigate the versatility of natural AAs to act
as eco-friendly co-catalysts for MX by virtue of their acidic and
basic hydrogen bonding groups which could act synergistically
with the halide ions of the alkali metal salt to catalyze the
Scheme 2 Amino acid varieties employed as co-catalysts. a. neutral
amino acid, b. neutral amino acid with hydrophobic side chain, c. neutral
amino acid with polar side chain, d. acidic amino acids, e. basic amino
CO –propylene oxide (PO) cycloaddition under mild reaction acids, and f. protected amino acids.
2
2024 | Dalton Trans., 2014, 43, 2023–2031
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
AAs employed in this study were of natural origin and so pos- Table 2 Influence of different molar ratios of histidine to KI in PC
a
synthesis
sessed “L stereochemistry.” No product was detected in the
absence of a catalyst (entry 1), and neither KI nor any of the
b
Histidine
(mmol)
KI
(mmol)
Histidine : KI
ratio
PC yield
(%)
AAs employed gave a conversion higher than 5% (entries 2, 3,
) when used alone under the employed reaction conditions.
Entry
5
1
2
3
4
5
6
7
0.08
0.08
0.08
0.08
0.04
0.02
—
—
—
5
11
51
98
87
23
4
However, the binary system of AA/KI exhibited pronounced
catalytic activities for PC synthesis (entries 4, 6–15) under
0.02
0.04
0.08
0.08
0.08
0.08
4 : 1
2 : 1
1 : 1
1 : 2
1 : 4
—
mild, solvent free conditions of 120 °C and 1 MPa CO in 3 h,
2
with only 0.2 mol% catalyst. Comparative studies indicated
that the basic AAs were the most effective in interacting with
the KI to give the desired product, yielding 70–90% PC (entries
a
Reaction conditions: PO (42.8 mmol), catalyst (0.08 mmol = 0.2 mol%),
20 °C, 1 MPa, 3 h. Determined by GC (biphenyl as internal standard).
1
3–15). While glycine and the acidic AAs exhibited similar cata-
lytic activity, giving PC yields of almost 25–30% (entries 4, 11,
2), the AAs with polar uncharged side chains provided slightly
b
1
1
higher yields of 50–60% (entries 9 and 10). The histidine/KI
system (entry 15) was found to be the most efficient among all
the employed AA/KI combinations, achieving a PC yield of
amount of 1 mol% using L-tryptophan/KI as the most active
catalyst. Hence we assume that, the anomalies in the results
between Yang et al. and our current work rely on the very low
catalytic amount of 0.2 mol% employed in this study.
9
9%, with a TON of 535 in 3 h. The other basic AAs, lysine and
arginine (entries 13 and 14, respectively) yielded moderate PO
conversions of 82% and 75% under the same conditions.
Being most active, we employed histidine as the model cata-
lyst for conducting further studies. Fig. 1 shows the depen-
dence of the cycloaddition reaction of PO and CO2 on the
amount of histidine/KI added. The catalytic activity can be
seen to increase gradually with increasing catalyst quantity,
reaching a plateau at approximately 0.2 mol% (which corres-
ponds to 0.08 mmol each of histidine and KI), with the molar
ratio of histidine to KI maintained at 1 : 1. Since further
increase in mol% of the catalyst did not bring about any sig-
nificant change, 0.2 mol% was deduced to be the optimal
amount of catalyst. The influence of the molar ratio of histi-
dine (co-catalyst) to KI (catalyst) was also studied, and the
results are summarized in Table 2.
With 0.2 mol% of the catalyst considered to be the
optimum (from Fig. 1), different ratios of histidine to KI were
chosen. At a very low KI amount (histidine : KI ratio of 4 : 1
(
1
entry 2)), the catalytic activity was found to be no more than
1% (PC yield); however, when the ratio became 2 : 1 (entry 3),
the PC yield became 51% (entry 2). The best activity was
exhibited at a molar ratio of 1 : 1 (entry 4) and a comparatively
similar activity was obtained with even half amount of histi-
dine also (87%, entry 5) displaying the pre-eminent role of KI
in the catalysis under the employed reaction conditions of
1
20 °C, 1.2 MPa CO pressure, and 3 h.
2
Reaction temperature was found to have a strong influence
on the catalytic activity of the histidine/KI system, as shown in
Fig. 2. With an almost negligible yield of PC observed at 80 °C,
the catalytic activity then increased exponentially from 100 °C
onwards, reaching a maximum at 120 °C with 99% PO conver-
sion and over 99% PC selectivity. These results were attributed
to the increased homogeneity of the catalyst system in the reac-
tion mixture at elevated temperatures, thereby rendering
9
h
Quite coincidentally, Yang et al. reported the capability of
similar catalytic system in CO –epoxide cycloaddition which
2
appeared during the preparation of this manuscript. The work
reports the achievement of excellent conversions at moderate
reaction conditions of 2 MPa, 120 °C in 1 h with a catalyst
Fig. 2 Effect of temperature in the histidine/KI catalyzed cycloaddition.
Reaction conditions: PO = 42.8 mmol, cat. amt = 0.2 mol%, 1 MPa, 3 h
selectivity of PC > 99%.
Fig. 1 Dependence of PC yield on catalyst amount. Reaction
conditions: PO = 42.8 mmol, 120 °C, 1 MPa, 3 h. Selectivity of PC > 99%.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 2023–2031 | 2025

View Article Online
Paper
Dalton Transactions
Fig. 4 Time variant study of PC synthesis using histidine/MX catalyst
systems. Reaction conditions: PO = 42.8 mmol, cat. amt = 0.2 mol%,
120 °C, 1 MPa. Selectivity of PC > 99%.
2
Fig. 3 PC yield – CO pressure dependence with histidine/KI catalyst
system. Reaction conditions: PO = 42.8 mmol, cat. amt = 0.2 mol%,
20 °C. 3 h. Selectivity of PC > 99%.
1
attributed to the difference in interaction of the metal ion with
the epoxide active site caused by higher Lewis acidity and ionic
potential of lithium and the increased solubility of Li salts in
increased mobility to the active species, which would in turn
furnish more effective collisions with the substrate molecules,
overcoming the energy barrier of the reaction. The histidine/KI
system maintained its selectivity towards PC at 140 °C also,
without compromising the conversion rate, demonstrating the
stability of the system.
1
1f
the reaction mixture at elevated temperatures.
Mechanism
With the objective of exploring the mechanism by which the
histidine/KI system affected the coupling of PO with CO , we
2
examined the co-catalytic behavior of histidine with different
MX (Table 3). Even though the effect of different alkali metals
such as Li, Na and K was not much predominant (see Fig. 4),
the particular halide ion employed was seen to have a signifi-
The pressure of the applied CO was also found to play a
critical role in the histidine/KI-catalyzed coupling reaction of
2
CO
2
with epoxides. Fig. 3 shows the increase in catalytic
activity (PC yield) achieved with increasing pressures of CO2,
even in the low pressure region of 0.1 to 1.2 MPa. This obser-
−
cant effect on the histidine catalysis system. The Cl ion
(
entries 1, 4, 7) tended to give the lowest catalytic activity, with
2
vation demonstrates that the supercritical state of CO is not a
−
I providing the highest PC yields. Although small differences
were observed on the employment of different alkali halides,
the catalytic activity of halide anions followed the order of
their increased nucleophilicity and leaving ability, with better
leaving groups providing the higher activity. This indicated the
importance of nucleophilic attack on the epoxide C atom in
the progress of the reaction. I− being the most labile nucleo-
phile would have initiated the nucleophilic attack more effec-
necessity for this particular system, even though it is essential
for achieving high PO conversion with either histidine or KI
alone. There are no previous reports of the use of KI or AAs
alone being able to catalyze the coupling of PO with CO to
2
this appreciable extent at pressures below 2 MPa. Hence, the
excellent PO conversion achieved in this study demonstrates
the importance of the synergism between the histidine and KI
in the catalytic system. The catalytic efficiency was retained in
the high pressure region of 2–3 MPa, with negligible differ-
ences in PC yield; however, further increases in the pressure
tended to lower the PC yield, as has been previously repor-
−
−
tively than the Br and Cl , as evidenced by the cycloaddition
results (entries 1–7, Table 3). The TON obtained with histi-
dine/MI reached 535 (at 1 MPa, 120 °C, 3 h), which was
5
f,8e
ted.
2
At higher pressures of CO , the equilibrium between
the gaseous and liquid phases must have affected such that it
Table 3 Co-catalytic performance of histidine with different alkali
may retard the interaction between PO and the catalyst by the
a
metal halides on CO
2
coupling with PO to yield PC
8
j,k
dilution effect, resulting in a lower conversion.
b
Entry
Catalyst
PC yield (%)
TON
The dependence of PC yield and selectivity on the reaction
time was also evaluated (Fig. 4), with the catalytic activity
found to be optimum at 3 h. The increased solubility of the
catalytic system in the more polar PC that was formed in the
initial stages further accelerated the interactions of the remain-
ing PO molecules with the catalyst. However, the catalysis rate
of the histidine/LiI system was found to be higher than that of
the histidine/KI and histidine/NaI as shown in Fig. 4, which is
1
2
3
4
5
6
7
LiCl
LiBr
NaCl
NaBr
KCl
KBr
KI
17
42
13
46
11
59
98
73
180
56
197
47
252
535
a
b
Reaction conditions: PO (42.8 mmol), 120 °C, 1 MPa, 3 h. Histidine-
indicative of the role of the alkali metal ion. This could be catalyst (0.2 mol% each).
2026 | Dalton Trans., 2014, 43, 2023–2031
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
comparable to the TON values achieved with the organic base yield of 73% (Table 1, entry 18), while the –COOH terminal
N,N-dimethylaminopyridine
(DMAP)/MX-catalyzed
cyclo- protected histidine methyl ester hydrochloride (Scheme 2f)
addition of epoxides with CO
2
(at 2 MPa, 130 °C, 3 h) reported yielded 68% PC (Table 1, entry 19), thereby displaying the sig-
1
1f
+
by Ramidi et al.
The co-operative role of hydrogen bonding groups in the –COO of the AAs with KI and the epoxide substrate.
halide-mediated catalysis of PO cycloaddition with CO for PC
synthesis has been established experimentally and theoreti-
cally by a number of groups, including ours.
ionic moiety is the common characteristic of all amino acids.
3
nificance of interaction between the zwitterionic –NH and
−
2
5c,k,8a–h
The zwitter-
DFT assisted mechanistic exploration
Since all the AAs were active, it was meaningful to consider the To gather more intuitive evidence for the proposed mechanis-
NH3 and –COO− to be responsible for the synergistic effect tic features, we conducted theoretical studies using density
with the halide nucleophile. It has been reported that –COOH functional theory (DFT) calculations. All the quantum mechan-
groups function with the halide nucleophiles in the most ical calculations were performed with the Gaussian 09 set package
effective way, initiating the catalysis by means of activating the with the B3LYP correlation functional. The basis sets employed
+
5
k,8e,f
epoxide O atom.
However, the similar activity of the acidic were the 6-311++G(d,p) for non-iodine atoms, and the
) required for
AAs, aspartic acid/KI and glutamic acid/KI (Table 1, entries 11 LANL2DZ dp for iodine. The activation energy (E
a
and 12, respectively) with the neutral AAs such as glycine/KI a non-catalyzed cycloaddition of PO with CO to produce PC
2
and alanine/KI (Table 1, entries 4 and 6) hinted that the roles has been reported previously to be around 55–56 kcal
of extra –COOH groups were negligible. At the same time, the mol−
1 8g,12a,b
.
This energy barrier is too high for the reaction to
basic AAs such as histidine, lysine and arginine had a pro- spontaneously proceed, necessitating the addition of a catalyst.
nounced effect on cyclic carbonate synthesis. As the ability of However, since glycine–KI system was found to activate the
organic bases such as DMAP, triphenylphosphine, pyridine, cycloaddition efficiently, we thought it worthwhile to compare
trialkylamines and many others in promoting the MX catalysis the potential energy surfaces (PES) of CO –PO cycloaddition
2
1
1a–f
of CO
2
–PO coupling has been already proposed,
we sur- catalyzed by KI alone and in combination with glycine (Fig. 5).
mised that the basic AAs/KI might be following a mechanism Glycine was chosen as the representative amino acid co-catalyst
similar to that exemplified by the organic base/KI system. We owing to its simplicity and because of the characteristic zwitter-
speculated that the epoxide ring would be activated by the ionic moiety it possesses.
zwitterionic ends of the AAs, whereas the additional basic
Since cellulose/KI represents a catalytic system similar to
moiety (of the basic AAs) would activate the CO2 molecule, ours, we considered it more sensible to follow the molecular
8
g
thereby lowering the energy barrier for the cycloaddition (see level calculations as of Han et al. (paths 4, 11 and 12), and
Scheme 3). The results obtained using protected histidine the catalyst interaction with the epoxide substrate was
1
2a,b
derivatives (Table 1, entries 18, 19) furnished additional evi- assumed to be the rate determining step.
The total energy
dence for this. The histidine derivative with the NH -terminal of the system including the catalyst and PO was preset to zero
2
protected with Boc-anhydride (Scheme 2f) gave a lower PC in both cases. Path A depicts the reaction scheme catalyzed by
KI alone, while Path B shows the co-catalytic effect of glycine
+
with KI. In Path A, the epoxide interacts with the K ion,
2
forming intermediate 1 (Int. 1). The O–CH bond of the PO
subsequently breaks, leading to the first transition state (TS-1),
−
1
which requires 38.06 kcal mol of E
a
. However, TS-1 drops to
Int. 2, in which the I− ion is covalently attached to the β-C of
the PO (forming a C–I bond), which is only 17.9 kcal mol
lower than TS-1. Formation of Int. 3 (−36.1 kcal mol ) occurs
2
without TS when the CO molecule contacts the K ion. The
−
1
−1
+
C–I bond disappears in the next transition state (TS-2), followed
by the ring closure, generating the final intermediate (Int. 4)
stabilized at −52.60 kcal mol− which ultimately renders the
product PC. All the DFT figures are included in the ESI.†
1
+
−
In Path B, the K interacts with the –COO entity of the
−
1
zwitterionic AA, forming Int. 1 at −29.65 kcal mol . The
stereochemistry of the PO–glycine–KI system rearranges from
Int. 1, with the K ion simultaneously interacting with the
+
−
epoxide oxygen and the –COO of glycine, weakening the β-C–
O bond of PO, and forming TS-1. The E required for TS-1 for-
a
mation was calculated to be 36.53 kcal mol− , with the I ion
now being in a suitable position for initiating nucleophilic
attack on the β-C atom of the PO. Interestingly, one of the H
1
−
Scheme 3 The probable mechanism of histidine/KI catalyzed cyclo-
addition of CO with propylene oxide to yield the propylene carbonate.
2
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 2023–2031 | 2027

View Article Online
Paper
Dalton Transactions
Fig. 5 Potential energy surfaces (PES) from DFT for: (a) Path A: KI catalyzed (without glycine) cycloaddition of PO with CO
2
; (b) Path B: KI–glycine
2
catalyzed cycloaddition of PO with CO (above). Possible pathway through which the cycloaddition proceeds in Path A and Path B (below).
+
atoms of the –NH3 moiety of glycine was found to be dis-
located between the N atom (of glycine) and the epoxide O.
The electron density of this H atom was calculated to be closer
to the epoxide O when TS-1 breaks down to Int. 2 (−37.71 kcal
−
1
mol ), with an apparent seven membered ring formed with
the epoxide O (Fig. 6), the exchanged H, and the N and C
+
termini of glycine and the K ion, stabilizing the intermediate
by 44.59 kcal mol− compared to TS-1 which lies lower in
the potential well. This could be paralleled to the seven mem-
bered ring formed by the vicinal hydroxyl groups of cellulose
1
8
a
2
with the epoxide. At this point, the CO molecule approaches
Int. 2, whereby the epoxide O commences nucleophilic attack
on the CO , forming an alkyl carbonate anion. The formation
2
of this intermediate (Int. 3) occurs without a TS, and is stabil-
ized by an even lower energy (−55.12 kcal mol−1). Sub-
Fig. 6 The seven membered ring which is supposed to stabilize the
sequently, the halide ion leaves TS-2, followed by ring closure ring opening complex, formed as intermediate 2 in Path B.
2028 | Dalton Trans., 2014, 43, 2023–2031
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
of the alkyl carbonate to generate the stable intermediate PC – Table 4 Catalytic performance of histidine/KI system on the cyclo-
a
addition of various epoxides with CO
2
glycine–KI adduct at an even reduced energy of −65.51 kcal
mol leading to the product PC leaving behind the catalyst
−
1
b
Yield [Selectivity]
(%)
system.
Entry Epoxide
1
Product
As mentioned earlier, for both pathways, the rate determin-
ing step was deduced to be the ring opening of the PO (TS-1).
In the KI-catalyzed reaction (Path A), only a narrow energy gap
exists between TS-1 and Int. 2, which makes the system vulner-
able to undergo the reverse reaction, and makes the inter-
86 [98]
2
action with CO less probable. The cation–anion mediated
2
99 [99]
catalysis has been previously reported to follow reversible reac-
1
2a,c
tion kinetics.
Hence, there is an increased possibility of
Int. 2 reverting to TS-1, which would then likely convert back
to Int. 1, significantly hindering the reaction. However, in Path
B, which is catalyzed by the glycine/KI system, even though the
TS-1 activation energy (36.83 kcal mol−1) is comparable to that
3
4
98 [99]
97 [98]
−
1
of the KI-catalyzed reaction (38.03 kcal mol ), Int. 2
1
−1
(
−37.71 kcal mol− ) is highly stabilized by 44.59 kcal mol
compared to TS-1, eliminating the possibility of reversibility.
The aforementioned justified the experimental observation of
better activity of glycine–KI system over KI.
By extrapolating the above ab initio calculations, the un-
ambiguous higher activity displayed by basic AA/KI system in
the cyclic carbonate yield could be surmised as follows. The
additional amine groups of the basic AAs (histidine/lysine/argi-
5
10 [96]
nine) would generate a nucleophilic attack to the CO mole-
2
cule forming a carbamate salt reversibly as suggested by many
a
8
b,i
Reaction conditions: Epoxide = 42.8 mmol, histidine/KI = 0.2 mol%,
20 °C, 1 MPa, 3 h. Calculated from GC.
previous studies.
This would be followed by nucleophilic
b
1
attack of the polarized O atom of PO on the carbamate salt,
producing the alkyl carbonate anion, which would sub-
sequently follow the intramolecular closure to generate the
desired product (Scheme 3). To summarize, the extra edge in
Conclusion
activity of basic amino acid co-catalysts could be attributed to Naturally occurring AAs were found to act as environmentally
the generation of stronger interactions made possible by the benign and readily available co-catalysts for the MX catalysis of
additional basic moieties with the CO molecule during the CO
cycloaddition with epoxides. The multi-functional AA/MX
2
2
cycloaddition.
catalyst system was found to efficiently yield the corresponding
This theoretical analysis explains the superior activity of cyclic carbonates in excellent yields, with almost complete
basic AAs in combination with KI for PC synthesis. However, selectivity under mild solvent free conditions of 1 MPa CO
2
the superior activity of histidine over lysine and arginine still and 120 °C in a short time of 3 h with a very low catalytic
needs to be addressed. We attribute these variations mainly to amount of 0.2 mol%. Among the different AAs employed,
conjugation of the imidazole ring of histidine, which would be those with basic moieties, histidine in particular, were
stabilizing the carbamate salt more than what would occur observed to have significant co-catalytic ability with MX salts.
with the primary amine-containing side chains of arginine The reason for the high catalytic activity of the AA/KI system
and lysine. In addition, differences in the solubilities of these was attributed to the probable synergism between the halide
AAs in the reaction mixture may also have an effect.
anion and the hydrogen bonding groups of the AA. To further
The versatility of the histidine/KI system for the cyclo- corroborate the proposed mechanism, a DFT investigation into
addition of other epoxides was also examined. Almost all the AA co-catalysis was carried out. To the best of our knowl-
the tested epoxides, except the usually challenging cyclohexene edge, this is the first time such a theoretical methodology has
oxide, were efficiently converted into their corresponding been applied to this system. The calculations indicated the for-
cyclic carbonates in the presence of a low amount of mation of a seven membered intermediate in the AA co-cata-
2
catalyst (0.2 mol%, Table 4). The low conversion rate of lyzed cycloaddition of PO with CO , which stabilized the
cyclohexene oxide could be due to the high steric hindrance otherwise reversible rate determining step of the MX-catalyzed
caused by the crowded cyclohexene ring. However, the high cycloaddition. The exploitation of the potential of bio-renew-
selectivity exemplified by the amino acid/KI catalyst irrespec- able natural materials for the synthesis of industrially valuable
tive of the type of amino acid or the epoxide employed is materials is a huge step towards the development of a sustain-
noteworthy.
able world. We believe that the exploration of the catalytic
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 2023–2031 | 2029

View Article Online
Paper
Dalton Transactions
capability of AAs will provide significant advances in this
emerging field of eco-friendly chemical fixation of CO2.
227, 537; (d) J. L. Jiang, F. X. Gao, R. M. Hua and X. Q. Qiu,
J. Org. Chem., 2005, 70, 381.
4
5
(a) M. Tu and R. J. Davis, J. Catal., 2001, 85, 199;
(
1
b) E. J. Doskocil, Microporous Mesoporous Mater., 2004, 76,
77; (c) E. J. Doskocil, J. Phys. Chem. B, 2005, 109, 2315.
Characterization
(a) J. Sun, W. Cheng, W. Fan, Y. Wang, Z. Meng and
S. Zhang, Catal. Today, 2009, 148, 361; (b) L. Dai, L. Chen,
S. F. Yin, W. H. Li, Y. Y. Zhang, S. L. Luo and C. T. Au,
Catal. Lett., 2010, 137, 74; (c) T. Takahashi, T. Watahiki,
S. Kitazume, H. Yasuda and T. Sakakura, Chem. Commun.,
A gas chromatograph (HP 6890 A) equipped with an HP-1
capillary column was used to determine the concentration of
the various cyclic carbonates formed. The peak of the cyclic
carbonate was identified with the authentic sample.
2006, 1664; (d) J. Q. Wang, X. D. Yue, F. Cai and L. N. He,
Catal. Commun., 2007, 8, 167; (e) X. Zhang, D. Wang,
N. Zhao, A. S. N. Al-Arifi, T. Aouak, Z. A. Al-Othman, W. Wei
and Y. Sun, Catal. Commun., 2009, 11, 43; (f) L. Han,
S. W. Park and D. W. Park, Energy Environ. Sci., 2009, 2,
Materials used
All the α-amino acids, alkali metal halides and the epoxides
were purchased from Sigma Aldrich with 99.5% purity. N -Boc-
α
L-histidine and L-histidine methyl ester dihydrochloride were
purchased from TCI chemicals Korea holding a purity of
1286; (g) X. Zhang, N. Zhao, W. Wei and Y. Sun, Catal.
Today, 2006, 115, 102; (h) S. Udayakumar, M. Lee, H. Shim,
S. Park and D. Park, Catal. Commun., 2009, 10, 659;
9
2
9.5%. CO was obtained from MS Gas Corporation, Korea.
(
i) S. Udayakumar, S. W. Park, D. W. Park and B. S. Choi,
Catal. Commun., 2008, 9, 1563; ( j) Y. Zhao, J. S. Tian,
X. H. Qi, Z. N. Han, Y. Y. Zhuang and L. N. He, J. Mol.
Catal. A: Chem., 2007, 271, 284; (k) L. Han, H. J. Choi,
S. J. Choi, B. Liu and D. W. Park, Green Chem., 2011, 13,
Coupling reaction
All the reactions were carried out in a 25 mL stainless-steel
batch reactor with a magnetic stirrer at 800 rpm. In a typical
batch reaction process, a pre-decided amount of the catalyst
was charged into the reactor containing 42.8 mmol of PO. The
reaction was carried out under a preset pressure of carbon
dioxide at different temperatures. After the completion of the
reaction, the reactor was cooled to zero degree and the pro-
ducts were identified by a gas chromatograph (Agilent HP 6890
A) equipped with a capillary column (HP-5, 30 m × 0.25 μm)
using a flame ionized detector. The product yield was deter-
mined using an internal standard method with biphenyl
1
023–1028; (l) H. L. Shim, S. Udayakumar, J. I. Yu, I. Kim
and D. W. Park, Catal. Today, 2009, 148, 350;
m) M. M. Dharman, H. J. Choi, D. W. Kim and D. W. Park,
(
Catal. Today, 2011, 164, 544.
6
(a) H. Y. Ju, J. Y. Ahn, M. D. Manju, K. H. Kim and
D. W. Park, Korean J. Chem. Eng., 2008, 25, 471; (b) J. Sun,
S. Zhang, W. Cheng and J. Ren, Tetrahedron Lett., 2008, 49,
3588; (c) H. Kawanami, A. Sasaki, K. Matsui and
Y. Ikushima, Chem. Commun., 2003, 896; (d) E. H. Lee,
J. Y. Ahn, M. M. Dharman, D. W. Park, S. W. Park and
I. Kim, Catal. Today, 2008, 131, 130; (e) J. Sun, S. Ichiro,
S. Fujita and M. Arai, J. Organomet. Chem., 2005, 690, 3490;
(0.05 g) as the standard.
(
f) J. Peng and Y. Deng, New J. Chem., 2001, 25, 639.
7
(a) C. M. Miralda, E. E. Macias, M. Zhu, P. Ratnasamy and
M. A. Carreon, ACS Catal., 2012, 2, 180; (b) J. Song,
Z. Zhang, S. Hu, T. Wu, T. Jiang and B. Han, Green Chem.,
Acknowledgements
This work was supported by the National Research Foun-
dation of Korea (NRF 2012-001507) and Global Frontier
Program. Y. Won is also thankful to NRF (grant number:
2009, 11, 1031; (c) H. Deng, C. J. Doonan, H. Furukawa,
R. B. Ferreira, J. Towne, C. B. Knobler, B. Wang and
O. M. Yaghi, Science, 2010, 327, 846; (d) W. Kleist, F. Jutz,
M. Maciejewski and A. Baiker, Eur. J. Inorg. Chem., 2009,
2
013R1A1A2058739).
2009, 3552; (e) Y. J. Kim and D. W. Park, J. Nanosci. Nano-
technol., 2013, 13, 2307; (f) M. Zhu, D. Srinivas,
S. Bhogeswararao, P. Ratnasamy and M. A. Carreon, Catal.
Commun., 2013, 32, 36; (g) A. C. Kathalikkattil and
D. W. Park, J. Nanosci. Nanotechnol., 2013, 13, 2230.
Notes and references
1
(a) M. Aresta and A. Dibenedetto, Dalton Trans., 2007, 2975;
b) M. North, R. Pasquale and C. Young, Green Chem., 2010,
(
1
1
2, 1514; (c) M. Yoshida and M. Ihara, Chem.–Eur. J., 2001,
0, 2886; (d) T. Sakakura, J. C. Choi and H. Yasuda, Chem.
8 (a) S. Liang, H. Liu, T. Jiang, J. Song, G. Yang and B. Han,
Chem. Commun., 2011, 47, 2131; (b) J. Sun, J. Wang,
W. Cheng, J. Zhang, X. Li, S. Zhang and Y. She, Green
Chem., 2012, 14, 654; (c) L. Song, Z. F. Zhang, B. Han,
S. Q. Hu, W. J. Li and Y. Xie, Green Chem., 2008, 10, 1337;
(d) J. Shi, J. Song, J. Ma, Z. Zhang, H. Fan and B. Han,
Pure Appl. Chem., 2013, 85, 1633; (e) K. R. Roshan,
G. Mathai, J. Kim, J. Tharun, G. A. Park and D. W. Park,
Rev., 2007, 107, 2365.
2
3
(a) T. Sakakura and K. Kohno, Chem. Commun., 2009, 1312.
(a) W. L. Dai, S. L. Luo, S. F. Yin and C. T. Au, Appl. Catal..
A, 2009, 366, 2; (b) J. Meléndez, M. North and
P. Villuendas, Chem. Commun., 2009, 18, 2577; (c) X. B. Lu,
Y. J. Zhang, K. Jin, L. M. Luo and H. Wang, J. Catal., 2004,
2030 | Dalton Trans., 2014, 43, 2023–2031
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Green Chem., 2012, 14, 2933; (f) J. Tharun, G. Mathai,
A. C. Kathalikkattil, R. Roshan, J. Y. Kwak and D. W. Park,
Green Chem., 2013, 15, 1673; (g) J. Ma, J. Liu, Z. Zhang and
S. J. Miller, Tetrahedron, 2002, 58, 2481; (h) Z. Yang, J. Sun,
W. Cheng, J. Wang, Q. Li and S. J. Zhang, Catal. Commun.,
2014, 44, 3.
B. Han, Green Chem., 2012, 14, 2410; (h) T. Jose, Y. Hwang, 10 (a) K. Kossev, N. Koseva and K. Troev, J. Mol. Catal. A:
R. Roshan, S. Ahn, A. C. Kathalikkattil and D. W. Park,
Catal. Sci. Technol., 2012, 2, 1674; (i) K. R. Roshan,
J. Tharun, A. C. Kathalikkattil, D. W. Kim, B. Kim and
Chem., 2003, 194, 29; (b) F. Li, C. Xia, L. Xu, W. Sun and
G. Chen, Chem. Commun., 2003, 2042; (c) J. Sun, S.-I. Fujita,
F. Zhao and M. Arai, Green Chem., 2004, 6, 613–616.
D. W. Park, Appl. Catal., A, 2013, 467, 17; ( j) J. J. Peng and 11 (a) Y. Kishimoto and I. Ogawa, Ind. Eng. Chem. Res., 2004,
Y. Q. Deng, New J. Chem., 2001, 25, 639; (k) L. F. Xiao,
F. W. Li, J. Peng and C. Xia, J. Mol. Catal. A: Chem., 2006,
43, 8155; (b) N. Kihara and T. Endo, Macromolecules, 1992,
25, 4824; (c) J. W. Huang and M. Shi, J. Org. Chem., 2003,
68, 6705; (d) N. Kihara, N. Hara and T. Endo, J. Org. Chem.,
1993, 58, 6198; (e) B. Barkakaty, K. Morino, A. Sudo and
T. Endo, Green Chem., 2010, 12, 42; (f) P. Ramidi,
P. Munshi, Y. Gartia, S. Pulla, A. S. Biris, A. Paul and
A. Ghosh, Chem. Phys. Lett., 2011, 512, 273.
2
53, 265.
9
(a) C. Qi, H. Jiang, Z. Wang, B. Zou and S. Yang, Synlett,
2
1
007, 2, 255; (b) C. Qi and H. Jiang, Sci. China, 2010, 53,
566; (c) C. Qi, J. Ye, W. Zeng and H. Jiang, Adv. Synth.
Catal., 2010, 352, 1925; (d) F. Wu, X. Y. Dong, L. N. He and
C. X. Miao, Lett. Org. Chem., 2010, 7, 73; (e) Q. Gong, 12 (a) Y. Ren, C. H. Guo, J. F. Jia and H. S. Wu, J. Phys.
H. Luo, J. Cao, H. Zhang, W. Wang and X. Zhou,
Aust. J. Chem., 2012, 65, 381; (f) J. Tharun, G. Mathai,
R. Roshan, A. C. Kathalikkattil, B. M. Kim and D. W. Park,
Phys. Chem. Chem. Phys., 2013, 15, 9023; (g) E. R. Jarvo and
Chem. A, 2011, 115, 2258; (b) H. Sun and D. Zhang,
J. Phys. Chem. A, 2007, 111, 8036; (c) S. Udayakumar,
S. W. Park, D. W. Park and B. S. Choi, Catal. Commun.,
2008, 9, 1563.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 2023–2031 | 2031