Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates

Article abstract of DOI:10.1016/j.chempr.2019.10.009

If cycloaddition of CO₂ to epoxides is to become a viable non-redox CO₂ fixation path, it is crucial that researchers develop an active, stable, selective, metal-free, reusable, and cost-effective catalyst. To this end, we report here a new catalyst that is based on imidazolinium functionality and is synthesized from an unprecedented, one-pot reaction of the widely available monomers terephthalaldehyde and ammonium chloride. We show that this covalent organic polymer (COP)-222 exhibits quantitative conversion and selectivity for a range of substrates under ambient conditions and without the need for co-catalysts, metals, solvent, or pressure. COP-222 is recyclable and has been demonstrated to retain complete retention of activity for over 15 cycles. Moreover, it is scalable to at least a kilogram scale. We determined the reaction mechanism by using quantum mechanics (density functional theory), showing that it involves nucleophilic-attack-driven epoxide ring opening (ND-ERO). This contrasts with the commonly assumed mechanism involving the concerted addition of chemisorbed CO₂. To stop global warming, we must introduce a variety of CO₂ reuse pathways. Redox chemistry is not trivial; reduction of CO₂ back to methane requires up to 8 electrons per molecule, leading to heavy energy demand. Non-redox paths have low energy needs and could provide a quick relief. A promising non-redox CO₂ product, cyclic carbonate is a versatile building block for green plastics and solvents. Although studies date back as early as 1969, no industrially viable process has since been introduced, mainly because of the lack of an effective catalyst for direct addition of CO₂ to the epoxides. Conceptually, the ideal catalyst should (1) be free of metals; (2) be free of co-catalysts; (3) be free of high pressure requirements; (4) provide quantitative selectivity to cyclic carbonate (5) provide a wide substrate scope, including very hard substrates; (6) provide reusability; and (7) be inexpensive. The imidazolinium catalyst that we developed herein addresses all 7 qualities and offers rapid implementation for CO₂ reclamation. Yavuz and colleagues introduced a highly active catalyst for non-redox fixation of CO₂ into cyclic carbonates, a versatile product family with potential use in green polymers and solvents. The metal-free, heterogeneous imidazolinium network structure is easily made, scaled up, recycled, and inexpensive and provides quantitative selectivity and conversion yields over a wide substrate scope of epoxides.

Full text of DOI:10.1016/j.chempr.2019.10.009

Article
Catalytic Non-redox Carbon Dioxide Fixation
in Cyclic Carbonates
Saravanan Subramanian, Julius
Oppenheim, Doyun Kim, ...,
Byoungkook Kim, William A.
Goddard III, Cafer T. Yavuz
wag@caltech.edu (W.A.G.)
yavuz@kaist.ac.kr (C.T.Y.)
HIGHLIGHTS
A direct synthetic pathway to the
elusive imidazoline-based
network polymer
Quantitative and selective
cycloaddition of CO2 to epoxides
at ambient conditions
Wide substrate scope, low-cost,
heterogeneous, recyclable, and
metal-free catalyst
A comprehensive simulation of
speciation explaining the
cycloaddition mechanism
Yavuz and colleagues introduced a highly active catalyst for non-redox fixation of
CO2 into cyclic carbonates, a versatile product family with potential use in green
polymers and solvents. The metal-free, heterogeneous imidazolinium network
structure is easily made, scaled up, recycled, and inexpensive and provides
quantitative selectivity and conversion yields over a wide substrate scope of
epoxides.
Subramanian et al., Chem 5, 1–11
December 12, 2019 ª 2019 Elsevier Inc.
https://doi.org/10.1016/j.chempr.2019.10.009

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
Article
Catalytic Non-redox Carbon Dioxide
Fixation in Cyclic Carbonates
Saravanan Subramanian,1,7,8 Julius Oppenheim,2,7 Doyun Kim,3,7 Thien S. Nguyen,1 Wahyu M.H. Silo,3
1,3,5,6,9,
Byoungkook Kim,4 William A. Goddard III,2, and Cafer T. Yavuz
*
*
SUMMARY
The Bigger Picture
To stop global warming, we must
introduce a variety of CO2 reuse
pathways. Redox chemistry is not
trivial; reduction of CO2 back to
methane requires up to 8
If cycloaddition of CO2 to epoxides is to become a viable non-redox CO2 fixation
path, it is crucial that researchers develop an active, stable, selective, metal-free,
reusable, and cost-effective catalyst. To this end, we report here a new catalyst
that is based on imidazolinium functionality and is synthesized from an unprece-
dented, one-pot reaction of the widely available monomers terephthalaldehyde
and ammonium chloride. We show that this covalent organic polymer (COP)-222
exhibits quantitative conversion and selectivity for a range of substrates under
ambient conditions and without the need for co-catalysts, metals, solvent, or
pressure. COP-222 is recyclable and has been demonstrated to retain complete
retention of activity for over 15 cycles. Moreover, it is scalable to at least a kilo-
gram scale. We determined the reaction mechanism by using quantum me-
chanics (density functional theory), showing that it involves nucleophilic-
attack-driven epoxide ring opening (ND-ERO). This contrasts with the commonly
assumed mechanism involving the concerted addition of chemisorbed CO2.
electrons per molecule, leading to
heavy energy demand. Non-
redox paths have low energy
needs and could provide a quick
relief. A promising non-redox CO2
product, cyclic carbonate is a
versatile building block for green
plastics and solvents. Although
studies date back as early as 1969,
no industrially viable process has
since been introduced, mainly
because of the lack of an effective
catalyst for direct addition of CO2
to the epoxides. Conceptually,
the ideal catalyst should (1) be
free of metals; (2) be free of co-
catalysts; (3) be free of high
pressure requirements; (4)
INTRODUCTION
The rising atmospheric CO2 levels1 call for suitable conversion methods, those that
do not produce more CO2 than sequestering.2–5 It is rather difficult since CO2 is a
very stable molecule.6 Despite the high kinetic stability, CO2 is an inexpensive feed-
stock for production of valuable chemicals. Unfortunately, catalysts with dramatically
reduced activation barriers are essential to increase efficiency.7,8 Conventional ap-
proaches make use of high-energy redox processes to produce methanol, methane,
ethanol, or ethylene.9 In contrast, non-redox reactions of CO2 such as cyclic carbon-
ate formation present significant advantages. Specifically, non-redox reactions can
be conducted under ambient conditions, they are kinetically and thermally favor-
able, and the reaction rates can be tuned easily by designing suitable catalysts.
provide quantitative selectivity to
cyclic carbonate (5) provide a wide
substrate scope, including very
hard substrates; (6) provide
reusability; and (7) be
Indeed, the synthesis of cyclic carbonates through the cycloaddition of CO2 to ep-
oxides is a small scale but particularly effective way of chemically fixing CO2 since
cyclic carbonates are of widespread utility.10–12 For example, they are precursors
for synthesis of polycarbonates that are used as electrolytes, aprotic polar solvents,
and intermediates in fine chemical production, such as dialkyl carbonates, glycols,
carbamates, and pyrimidines.13–15 Because of their importance, many catalysts
have been developed for cyclic carbonate formation, including ionic liquids,16–18
metal complexes,15,19 functional polymers,20–22 ammonium and phosphonium
salts,23,24 metal-organic frameworks (MOFs),8,25,26 hydrogen-bonding cata-
lysts,7,27,28 and supported catalysts.29,30 Homogeneous, organometallic catalysts
have often proved effective with well-defined mechanisms and notable catalytic ac-
tivity,10,11 but the added cost of separations and purification of metallic species
inexpensive. The imidazolinium
catalyst that we developed herein
addresses all 7 qualities and offers
rapid implementation for CO2
reclamation.
Chem 5, 1–11, December 12, 2019 ª 2019 Elsevier Inc.
1

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
makes them unattractive. Heterogeneous catalysts are more advantageous in terms
of scalability, rapid catalyst recovery, and reusability. For example, Farha and co-
workers reported zirconium-based MOFs as catalysts and tetrabutylammonium bro-
mide (TBAB) as a co-catalyst for CO2 fixation into styrene oxide and achieved quan-
titative conversion under mild reaction conditions.31 More recently, Byun and Zhang
reported a facile route for imidazolium ionic liquids with heterogeneous property
that enabled the selective formation of cyclic carbonates (up to 99%) under ambient
reaction conditions.32 We also previously reported a pyridyl salicylimine polymeric
catalyst that showed quantitative conversion without the need for additives or co-
catalysts for conversion of even hard substrates such as styryl epoxide.33 See Table
S1 for a list of selected catalysts and their comparisons in catalytic activities.
Despite these remarkable advances, most literature reports demand the use of co-
catalysts, harsh reaction conditions (e.g., high pressure), and multi-step synthesis of
rare metal-based catalysts, leading to high energy requirements and high costs.34
To date, very few catalytic systems have shown considerable activity under ambient
conditions, and they remain far from being feasible for successful industrial applica-
tions. This necessitates the design of new catalysts that combine structural versatility
with suitable functionalities, amenability, and durability35 for cycloaddition reactions
of CO2 with epoxides.36
In order to develop a commercially attractive and environmentally benign metal-free
heterogeneous catalyst, we examined low-priced and non-toxic starting materials.
To achieve this challenging task, we limited the steps to synthesize the catalysts,
avoided the use of specialty chemicals, and focused on inert reaction conditions.
In a serendipitous discovery, we achieved a thermally robust heterogeneous catalyst
with inbuilt basic and quaternary ammonium centers that paved the way toward a
sustainable catalyst. We use a one-pot reaction of terephthalaldehyde and ammo-
nium chloride, to form an imidazolinium network polymer that shows excellent
chemical fixation of CO2 without the need for co-catalysts, a solvent, or extreme con-
ditions. The reactions proceed with great substrate tolerance and no obvious cata-
lyst deactivation was observed over 15 cycles. In order to understand the reaction
mechanisms underlying this dramatically improved catalyst, we carried out quantum
mechanics (QM) calculations (density functional theory [DFT]) to obtain the reaction
barriers for the various possible reaction pathways. These DFT studies showed the
key role played by the quaternary ammonium center and its chloride counter anion.
Indeed, our QM predicted rate constants lead to kinetics (including products and in-
termediate concentrations) in excellent agreement with experiments, validating the
new nucleophilic attack-driven epoxide ring opening (ND-ERO) mechanism.
1Graduate School of EEWS, Korea Advanced
Institute of Science and Technology (KAIST),
Daejeon 34141, Republic of Korea
2Materials and Process Simulation Center (MSC)
and Joint Center for Artificial Photosynthesis
(JCAP), California Institute of Technology,
Pasadena, CA 91125, USA
3Department of Chemical and Biomolecular
Engineering, Korea Advanced Institute of Science
and Technology (KAIST), Daejeon 34141,
Republic of Korea
4KAIST Analysis Center for Research
Advancement, Korea Advanced Institute of
Science and Technology (KAIST), Daejeon 34141,
Republic of Korea
RESULTS AND DISCUSSION
5Department of Chemistry, Korea Advanced
Institute of Science and Technology (KAIST),
Daejeon 34141, Republic of Korea
Developing catalysts for cyclic carbonate production that work at atmospheric pres-
sure conditions is crucial to provide a negative emissions technology (NET) for indus-
trial implementations. This requires a well-designed heterogeneous catalyst that is
robust, easily synthesized, scalable, cost effective, rapidly recovered, and recy-
clable. Based on our previous experience, we considered cyclic amine systems to
provide a solution since these structures have highly modular properties and can
be built stepwise, providing tunable structures. Consequently, we examined reac-
tions of aliphatic aldehydes with ammonia. In particular, we examined a commonly
reported procedure to form an aldehyde ammonia trimer.37,38 We speculated that
the one-pot reaction of terephthalaldehyde and ammonium chloride salt would
form a new porous material containing cyclic amines.
6Saudi Aramco, KAIST CO2 Management Center,
KAIST, Daejeon 34141, Republic of Korea
7These authors contributed equally
8Present address: Inorganic Materials and
Catalysis Division, Central Salt and Marine
Chemicals Research Institute, CSIR, G.B. Marg,
Bhavnagar, Gujarat 364 002, India
9Lead Contact
*Correspondence: wag@caltech.edu (W.A.G.),
yavuz@kaist.ac.kr (C.T.Y.)
https://doi.org/10.1016/j.chempr.2019.10.009
2
Chem 5, 1–11, December 12, 2019

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
Figure 1. Synthesis and Characterization of COP-222
(A) One-step, one-pot synthesis from commercially available substrates.
(B) SEM micrograph of the as-synthesized network polymer. The scale bar represents 100 mm.
(C) Theoretical and experimental 13C-NMR.
(D) 15N-NMR with 15N-enriched COP-222.
(E) XPS (N-1s) data.
(F) Theoretical and experimental FTIR spectra.
(G) CO2 (273 K) uptake isotherm, filled symbols show adsorption and empty symbols represent desorption.
Therefore, we investigated the reaction of terephthalaldehyde with ammonium chlo-
ride salt in the presence of dimethylformamide (DMF) solvent and heated the reac-
tion mixture to 150ꢀC for 24 h. We obtained a yellow precipitate (covalent organic
polymer [COP-222], the code is only for indexing purposes, no chemical information
is intended) that was collected and washed repeatedly with water, methanol, tetra-
hydrofuran (THF), and acetone (Figure 1A). The yellow solid was dried under vacuum
at 100ꢀC overnight and characterized with a series of analyses. The morphology of
the developed material was examined by SEM analysis (Figure 1B) and appears to
have rough surfaces with macropores. The solid-state cross-polarization magic-
angle spinning (CP-MAS) 13C NMR spectra shows an aromatic (aryl) signal at
129.61 ppm that was accompanied by two distinct peaks in the region of 43.02
and 68.89 ppm and one around 166.59 ppm. This shows clearly that the cyclic
ammonia trimer was not formed (Figure 1C). This prompted us to further elucidate
the structure by examining the 15N CP-MAS NMR (Figure 1D). In a symmetrical
trimer, one would expect a single peak for 15N. We observed two separate peaks
(83.72 and 124.18 ppm), revealing that there are two chemically dissimilar nitrogens.
Chem 5, 1–11, December 12, 2019
3

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
Quaternary ammonium nitrogens are usually observed at 20–80 ppm and imines at
130–300 ppm, while secondary amines show up at 10–90 ppm.39,40 These findings
led to us to postulate an imidazolinium construct (Figure 1). Revisiting the
13C-NMR, we then confirmed that the peaks at 43.02 and 68.89 ppm can be ascribed
to the carbon (d, e) and the peak at 166.59 ppm corresponds to -C=N- imine carbon
(a) in the imidazolyl ring (Figure 1A). These signals match well with the theoretical
13C-NMR spectra for an imidazolinium ring system (Figure 1C), and the observed
chemical shift patterns are in line with similar structures reported in the litera-
ture.41,42 Furthermore, X-ray photoelectron spectroscopy (XPS) showed strong
N-1s signals located at 399.78. We deconvoluted the N-1s spectrum for COP-222
into three peaks at 398.28, 399.78, and 400.78 corresponding to imine, pyrrolic,
and quaternary ammonium nitrogens, respectively. The XPS spectra for C, O, and
Cl are given in the Supplemental Information (Figure S1). The chemical connectivity
and the presence of the imidazoline ring were also confirmed by the stretching
bands observed at 1,618 cmÀ1 (-C=N-) and a broad peak around 3,400 cmÀ1
(-NH4+-) in Fourier transform infrared spectroscopy (FTIR) spectra (Figure 1F).
First, we studied the porosity of our COP-222 network polymer by the nitrogen
adsorption-desorption isotherms measured at 77 K leading to a Brunauer-Em-
mett-Teller (BET) surface area of 21 m2/g (Figure S2). This unexpected low surface
area might be due to the well-packed polymer through the hydrogen bonding,
which leads to a low diffusive volume. We then investigated the affinity of COP-
222 toward CO2 and carried out uptake measurements at 273, 298, and 323 K (Fig-
ures 1G and S2). The CO2 uptake of 1.08 mmol/g at 273 K is appreciable, and isos-
teric heat of adsorption (Qst) was found to be in the order of 27–30 kJ/mol, reflecting
an expected physisorptive nature. In order to confirm the chemical composition,
elemental analysis was performed showing that COP-222 possess a high nitrogen
content of 11.36%, which is close to the expected value of 12.13% (Figure S3). We
also analyzed the powder X-ray Diffraction (XRD) of COP-222 and found a character-
istic spectrum of an amorphous solid (Figure S3). Thermogravimetric analysis shows
that COP-222 is thermally stable up to 250ꢀC (Figure S4). Acidic and basic sites of
COP-222 were screened using temperature-programmed desorption (TPD) with
CO2 and NH3 gases. In CO2-TPD, the peaks at 107ꢀC and 209.9ꢀC represent the
basic sites that may be derived from imine and pyrrolic nitrogens (Figure S5A). A
broad NH3 desorption peak appeared at 102.6ꢀC in the NH3-TPD profile (Figure S5B)
representing quaternary ammonium acidic sites. Co-existence of both acidic and
basic sites confirm its potential as a cycloaddition catalyst.
The imidazolinium structure features two key functionalities for an effective catalyst
in cycloaddition of CO2 to epoxides:33 a quaternary amine and a basic amine. This is
why we investigated its utility in non-redox CO2 fixation. To test its activity, we chose
the reaction of epichlorohydrin since it is very reactive and is used in almost all re-
ports dealing with CO2 cycloaddition to epoxides (Figure 2).43 Industrially, however,
epichlorohydrin is not too important. In a typical experiment, we subjected epichlo-
rohydrin (5 mmol) to our catalysts (30 mg) under solvent-free conditions at 100ꢀC by
purging with atmospheric pressure CO2. Owing to the presence of imidazolinium
centers, COP-222 demonstrated excellent catalytic activity (>99% conversion)
compared with control structures (Figure 2B). Among those, COP-1 is a nitrogen-
rich porous COP44 while others are commercially available molecules: 2-imidazolidi-
none, ammonia acetaldehyde trimer, and imidazole. Despite being heterogeneous,
COP-222 accomplished quantitative conversion and stoichiometric selectivity. This
remarkable catalytic activity shows that the high N content combined with bifunc-
tional sites is critical. This also explains why other heterogeneous catalysts always
4
Chem 5, 1–11, December 12, 2019

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
Figure 2. Optimization of Catalyst Activity
(A) Cycloaddition of CO2 to epichlorohydrin was used to optimize catalytic activity.
(B) Screening of control structures for the cycloaddition reaction.
(C) Screening of catalyst loading.
(D) Screening of temperature.
(E) Conversion with respect to time.
(F) Recyclability of COP-222 for 15 cycles. Each cycle was set up using the recovered catalyst. Reaction conditions: catalyst, epichlorohydrin (5 mmol),
and CO2 (1 atm). Conversions were determined by using 1H NMR.
need a co-catalyst (in many cases TBAB) even though they have basic nitrogen sites.
Porosity seems to be less important as the reaction is rather slow and diffusion hin-
drance is not the bottle neck. This is also reflected in the need for an induction time
(Figure 2D), as is common for biphasic systems where intermediate species need to
build up to proceed further.
In order to optimize catalytic activity, we screened various reaction parameters such
as catalyst loading, temperature, and time. Decreasing catalyst loading to 20 mg did
not improve reaction yield (84%), and increasing it to 40 mg led to no significant
change (Figure 2C). At temperatures higher than 60ꢀC, COP-222 worked well and
Chem 5, 1–11, December 12, 2019
5

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
resulted in good to excellent yields (Figure 2D). The influence of reaction time in the
formation of cyclic carbonate is crucial, which increases with increasing time, reach-
ing a maximum at 24 h (Figure 2E). The steeper increase in yield after initial reactions
could be due to the enhanced solubility of CO2 in the cyclic carbonate-rich reaction
medium.33,45 Since rapid separation of the catalyst from the reaction mixture and re-
cyclability are the main reasons for developing heterogeneous catalytic systems, we
examined the cyclability of COP-222 using epichlorohydrin as a substrate under
optimized reaction conditions and found that the catalyst can be recycled at least
15 times without any loss in its activity (Figure 2F). We also monitored for potential
chloride leaching using ion chromatography and found no detectable presence in
the catalytic reactions. In addition, we used ammonium bromide and iodide in mak-
ing COP-222 variations. These derivatives did not improve the catalytic activity, nor
the synthesis procedure. Since chloride is found to be more active in cycloaddition
reactions, and also commonly available, we focused our attention to the chloride
version.
For widespread industrial use, the catalysts need to show effective cyclic carbonate for-
mation from challenging epoxides such as styryl or hexyl epoxides.33 To access synthet-
ically useful cyclic carbonates, we turned our attention to explore the scope of epoxide
substrates (Scheme 1). Simple aliphatic epoxides that are similar to the benchmark sub-
strate, epichlorohydrin, also yielded quantitative conversion and selectivity. Styrene ox-
ide and derivatives are commonly known as challenging substrates because of their low
reactivity.14 COP-222 was very effective, converting styryl substrates quantitatively to
the corresponding cyclic carbonates. We then investigated hard substrates such as hex-
yl- and phenoxy-substituted epoxides, where we also found high conversion yields and
selectivity. It is important to note that there are only a few previous literature reports
showing appreciable conversions of hexyl epoxide.
In order to determine the limits of our catalyst, we tested the cycloaddition of CO2 to
extremely hard substrates (Scheme 1). Cyclohexene oxide and resorcinol diglycidyl
ether are both sterically and electronically challenging. We found that COP-222 shows
great performance in converting both with quantitative selectivity and high conversion
yields. In the literature, cyclohexene oxide has been rarely studied, except a calcium-
polyethylene glycol catalyst46 and a squaramide organocatalyst.27 However, those
studies require either high pressure (20–30 bar) CO2 or a co-catalyst. Finally, we took
on a completely new epoxide, never before tested for cyclic carbonate formation.
The bis epoxide 1,3-butadiene diepoxide presents a unique challenge since the adja-
cent ring strains and their proximity could lead to unconventional products, including
extended ring formation. In repetitive tests, we achieved, as a single product, a five-
membered cyclic carbonate while the other epoxide is untouched. This new functional
asymmetry opens new directions for this commercial monomer. With all the substrate
scope and conversion yields, it should be noted that the catalytic performance of
COP-222 is outright superior to all previously reported heterogeneous catalysts8,34,47
and even better than most homogeneous systems.48–50
The mechanism of the cycloaddition of CO2 to epoxides is commonly believed to be a
multi-component ordeal, where epoxide is preferentially activated on a Lewis acid
site and CO2 is provided by a base through chemisorption.8,51 In order to validate
the accepted mechanism, we performed DFT calculations as described in the Supple-
mental Information. The calculation models of the catalyst, intermediate energy
profiles, and their predicted geometries in the stationary phase are presented in
the Supplemental Information. The DFT calculations find a ND-ERO mechanism (Fig-
ure 3). The ND-ERO catalytic cycle begins with (1) coordination of the epoxide to an
6
Chem 5, 1–11, December 12, 2019

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
Scheme 1. Cycloaddition Reaction of CO2 with Various Epoxides Catalyzed by COP-222
Conversion yields for the corresponding catalytic reactionsa are given in percentages.b The
selectivities are reported in parentheses.
aReaction conditions: substrate (5 mmol), COP-222 (30 mg), CO2 (1 atm) and temperature (100ꢀC).
bDetermined by using 1H-NMR.
cCarried out in glass lined autoclave reactor because of the volatility of the epoxide.
ammonium proton (step 1), (2) the nucleophilic chloride ion then opens the epoxide
ring via a SN2-type mechanism (step 2), (3) subsequently, the oxyanion is stabilized by
the ammonium proton until the oxyanion can attack the carbon of a nearby coordi-
nated carbon dioxide (step 3), and (4) the new oxyanion then undergoes another
SN2-type reaction, closing the 5-membered carbonate ring (step 4).52 In addition
to the above quite plausible pathway, we also assessed side reactions for potential
involvement. For example, we find that the chemical binding of CO2 to the imine ni-
trogen (Lewis base) is not favorable (Figure 3) requiring +6 kcal/mol energy. Another
possible side reaction is the release of 2-chloropropanol after the epoxide ring open-
ing. Once again, the energy differences were far lower than the adduct (À10.7 versus
À21.4 kcal/mol), leading to the completion of the catalytic cycle. We observed the
same after the CO2 addition to the alkoxide. The free carbonic acid intermediate pro-
duced half the enthalpic change compared to the coordinated version. Finally, we
calculated the energy requirement to release the HCl molecule from COP-222.
This is particularly important since the cyclability and product purity would otherwise
be adversely affected. The decomposition of COP-222 into neutral base and HCl was
energetically uphill, in the order of a staggering +19.9 kcal/mol. This confirms our
experimental observations of a stable catalyst.
Imidazoles are often reported to be activefor cycloaddition catalysis.16 For comparisons,
we calculated the neutral imidazole equivalent of COP-222 and found that the
Chem 5, 1–11, December 12, 2019
7

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
Figure 3. The Nucleophilic Attack-Driven Epoxide Ring Opening (ND-ERO) Reaction Mechanism
Reaction mechanism for the COP-222 catalyst derived from quantum mechanics, including free energy reaction barriers.
zwitterionic product is >25 kcal/mol uphill from the reactants with a barrier of
ꢁ40 kcal/mol (Figure S23, calculated with M06-2X/6-311G and implicit solvation). In
comparison, the initial catalytic step for COP-222 has a barrier of ꢁ20 kcal/mol and is
downhill by 21 kcal/mol. Similarly, non-aromatic imidazolium was found to be unstable
and form the aromatic imidazolium cation (Figure S24). The aromatic imidazolium, how-
ever, is also unable to activate an epoxide as there is no site for the epoxide oxygen to
coordinate to, rendering the chloride attack entropically unfavorable.
EXPERIMENTAL PROCEDURES
General Synthetic Procedure for COP-222
A 20 mL sample vial was charged with terephthaldehyde (7.45 mmol), ammonium
chloride (29.82 mmol), and 15 mL DMF. The sample vial was closed with the screw
cap and covered with a Teflon tape. The mixture was then allowed to heat at
150ꢀC (in the rate of 2ꢀC/min) for 48 h. The yellow precipitate was collected, filtered,
and then washed repeatedly with water, methanol, and acetone and chloroform. The
resulting powder was dried at 100ꢀC under vacuum overnight.
8
Chem 5, 1–11, December 12, 2019

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
General Procedure for the Synthesis of Cyclic Carbonates
In a typical reaction, a well-dried Schlenk tube equipped with a magnetic stirring bar
was charged with catalyst (30 mg) and epoxides (5 mmol). The Schlenk tube was
purged and pressurized with CO2 (1 atm), placed in a preheated oil bath, and stirred
for 24 h at 100ꢀC. Upon completion of the reaction, the tube was cooled to room
temperature. The reaction mixture was filtered off and washed thoroughly with chlo-
roform to ensure complete removal of the product. The collected filtrate was evap-
orated under reduced pressure and the samples were analyzed by 1H NMR spectros-
copy. The recovered catalyst was dried at 100ꢀC under vacuum and reused for
further cycles.
SUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.chempr.
2019.10.009.
ACKNOWLEDGMENTS
This work was supported by the Saudi Aramco-KAIST CO2 Management Center and
National Research Foundation of Korea (NRF) grants funded by the Korean govern-
ment (MSIP) (nos. NRF-2016R1A2B4011027, NRF-2017M3A7B4042140, and NRF-
2017M3A7B4042235). J.O. was supported by an Ernest H. Swift Summer Under-
graduate Research Fellowship at Caltech. The QM studies were supported by the
Joint Center for Artificial Photosynthesis, a DOE Energy Innovation Hub, supported
through the Office of Science of the U.S. Department of Energy under award no. DE-
SC0004993.
AUTHOR CONTRIBUTIONS
S.S. and D.K. synthesized and characterized the catalyst and carried out catalytic ex-
periments. J.O. and W.A.G. carried out DFT calculations and explained the mecha-
nism. T.S.N. studied the new butadiene oxide substrate. W.M.H.S. carried out the
cyclic experiments. B.K. performed all solid-state NMRs. C.T.Y. conceived the proj-
ect and wrote the manuscript with contributions from all authors.
DECLARATION OF INTERESTS
KAIST has filed a provisional patent application (KR #10-2019-0060197) related to
the catalysts reported in this manuscript.
Received: March 19, 2019
Revised: April 24, 2019
Accepted: October 11, 2019
Published: November 14, 2019
REFERENCES AND NOTES
1. Xu, Y., Ramanathan, V., and Victor, D.G. (2018).
Global warming will happen faster than we
think. Nature 564, 30–32.
4. Lan, Y.Q., Jiang, H.L., Li, S.L., and Xu, Q.
(2011). Mesoporous metal-organic
frameworks with size-tunable
6. Sholl, D.S., and Lively, R.P. (2016). Seven
chemical separations to change the world.
Nature 532, 435–437.
cages: selective CO2 uptake,
2. Sun, Z., Ma, T., Tao, H., Fan, Q., and Han, B.
(2017). Fundamentals and challenges of
electrochemical CO2 reduction using two-
dimensional materials. Chem 3, 560–587.
encapsulation of Ln3+ cations
for luminescence, and column-
chromatographic dye separation. Adv. Mater.
23, 5015–5020.
7. Liu, N., Xie, Y.F., Wang, C., Li, S.J., Wei, D., Li,
M., and Dai, B. (2018). Cooperative
multifunctional organocatalysts for ambient
conversion of carbon dioxide into cyclic
carbonates. ACS Catal. 8, 9945–9957.
3. Wang, C., Luo, X., Luo, H., Jiang, D.E., Li, H.,
and Dai, S. (2011). Tuning the basicity of ionic
liquids for equimolar CO2 capture. Angew.
Chem. Int. Ed. 50, 4918–4922.
5. Patel, H.A., Byun, J., and Yavuz, C.T. (2017).
Carbon dioxide capture adsorbents:
chemistry and methods. ChemSusChem 10,
1303–1317.
8. Beyzavi, M.H., Stephenson, C.J., Liu, Y.,
Karagiaridi, O., Hupp, J.T., and Farha, O.K.
(2015). Metal-organic framework-based
Chem 5, 1–11, December 12, 2019
9

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
catalysts: chemical fixation of CO2 with
epoxides leading to cyclic organic carbonates.
Front. Energy Res. 2, 63.
incorporating sterically confined N-
heterocyclic carbenes for simultaneous CO2
capture and conversion at ambient pressure.
Chem. Mater. 27, 6818–6826.
35. Gao, W.Y., Chen, Y., Niu, Y., Williams, K., Cash,
L., Perez, P.J., Wojtas, L., Cai, J., Chen, Y.S., and
Ma, S. (2014). Crystal engineering of an nbo
topology metal-organic framework for
chemical fixation of CO2 under ambient
conditions. Angew. Chem. Int. Ed. 53, 2615–
2619.
9. Gao, P., Li, S., Bu, X., Dang, S., Liu, Z., Wang, H.,
Zhong, L., Qiu, M., Yang, C., Cai, J., et al.
(2017). Direct conversion of CO2 into liquid
fuels with high selectivity over a bifunctional
catalyst. Nat. Chem. 9, 1019–1024.
23. Liu, S., Suematsu, N., Maruoka, K., and
Shirakawa, S. (2016). Design of bifunctional
quaternary phosphonium salt catalysts for CO2
fixation reaction with epoxides under mild
conditions. Green Chem. 18, 4611–4615.
36. Kamphuis, A.J., Picchioni, F., and Pescarmona,
P.P. (2019). CO2-fixation into cyclic and
polymeric carbonates: principles and
10. Martı´n, C., Fiorani, G., and Kleij, A.W. (2015).
Recent advances in the catalytic preparation of
cyclic organic carbonates. ACS Catal. 5, 1353–
1370.
24. Steinbauer, J., Longwitz, L., Frank, M., Epping,
J., Kragl, U., and Werner, T. (2017).
Immobilized bifunctional phosphonium salts as
recyclable organocatalysts in the cycloaddition
of CO2 and epoxides. Green Chem. 19, 4435–
4445.
applications. Green Chem. 21, 406–448.
37. Nielsen, A.T., Moore, D.W., Ogan, M.D., and
Atkins, R.L. (1979). Structure and chemistry of
the aldehyde ammonias. 3. Formaldehyde-
ammonia reaction. 1, 3, 5-Hexahydrotriazine.
J. Org. Chem. 44, 1678–1684.
11. Cokoja, M., Wilhelm, M.E., Anthofer, M.H.,
Herrmann, W.A., and Kuhn, F.E. (2015).
¨
Synthesis of cyclic carbonates from epoxides
and carbon dioxide by using organocatalysts.
ChemSusChem 8, 2436–2454.
25. Han, Q., Qi, B., Ren, W., He, C., Niu, J., and
Duan, C. (2015). Polyoxometalate-based
homochiral metal-organic frameworks for
tandem asymmetric transformation of cyclic
carbonates from olefins. Nat. Commun. 6,
10007.
38. Tuguldurova, V.P., Fateev, A.V., Malkov, V.S.,
Poleshchuk, O.K., and Vodyankina, O.V. (2017).
Acetaldehyde-ammonia interaction: a DFT
study of reaction mechanism and
`
12. Guo, W., Go´ mez, J.E., Cristo` fol, A., Xie, J., and
Kleij, A.W. (2018). Catalytic transformations of
functionalized cyclic organic carbonates.
Angew. Chem. Int. Ed. 57, 13735–13747.
product identification. J. Phys. Chem. A 121,
3136–3141.
26. Chen, D.P., Luo, R., Li, M., Wen, M., Li, Y., Chen,
C., and Zhang, N. (2017). Salen(Co(III))
imprisoned within pores of a metal-organic
framework by post-synthetic modification and
its asymmetric catalysis for CO2 fixation at
room temperature. Chem. Commun. (Camb.)
53, 10930–10933.
13. Shaikh, R.R., Pornpraprom, S., and D’Elia, V.
(2018). Catalytic strategies for the
ꢀ
39. Marek, R., Brus, J., Tousek, J., Kova´cs, L., and
Hockova´, D. (2002). N7-and N9-substituted
purine derivatives: a 15N NMR study. Magn.
Reson. Chem. 40, 353–360.
cycloaddition of pure, diluted, and waste CO2
to epoxides under ambient conditions. ACS
Catal. 8, 419–450.
40. Poll, E.M., Samba, S., Dieter Fischer, R.,
Olbrich, F., Davies, N.A., Avalle, P., Apperley,
D.C., and Harris, R.K. (2000). Simple quaternary
ammonium ions R4N+ ,(R = nPr, nBu, nPen) as
versatile structure directors for the synthesis of
zeolite-like, heterobimetallic cyanide
14. Lu, X.B., and Darensbourg, D.J. (2012). Cobalt
catalysts for the coupling of CO2 and epoxides
to provide polycarbonates and cyclic
27. Sopena, S., Martin, E., Escudero-Ada´n, E.C.,
˜
and Kleij, A.W. (2017). Pushing the limits with
Squaramide-based organocatalysts in cyclic
carbonate synthesis. ACS Catal. 7, 3532–3539.
carbonates. Chem. Soc. Rev. 41, 1462–1484.
15. Comerford, J.W., Ingram, I.D.V., North, M., and
Wu, X. (2015). Sustainable metal-based
catalysts for the synthesis of cyclic carbonates
containing five-membered rings. Green Chem.
17, 1966–1987.
28. Alves, M., Grignard, B., Mereau, R., Jerome, C.,
Tassaing, T., and Detrembleur, C. (2017).
Organocatalyzed coupling of carbon dioxide
with epoxides for the synthesis of cyclic
frameworks. J. Solid State Chem. 152, 286–301.
41. Murai, K., Fukushima, S., Hayashi, S., Takahara,
Y., and Fujioka, H. (2010). C3-symmetric chiral
Trisimidazoline: design and application to
organocatalyst. Org. Lett. 12, 964–966.
carbonates: catalyst design and mechanistic
studies. Catal. Sci. Technol. 7, 2651–2684.
16. Girard, A.L., Simon, N., Zanatta, M., Marmitt, S.,
Gonc¸alves, P., and Dupont, J. (2014). Insights
on recyclable catalytic system composed of
task-specific ionic liquids for the chemical
fixation of carbon dioxide. Green Chem. 16,
2815–2825.
29. Kohrt, C., and Werner, T. (2015). Recyclable
bifunctional polystyrene and silica gel-
supported organocatalyst for the coupling of
CO2 with epoxides. ChemSusChem 8, 2031–
2034.
42. Akalay, D., Durner, G., Bats, J.W., Bolte, M.,
¨
and Gobel, M.W. (2007). Synthesis of C -
¨
2
symmetric bisamidines: a new type of chiral
metal-free lewis acid analogue interacting with
carbonyl groups. J. Org. Chem. 72, 5618–5624.
17. Cheng, W., Su, Q., Wang, J., Sun, J., and Ng, F.
(2013). Ionic liquids: the synergistic catalytic
effect in the synthesis of cyclic carbonates.
Catalysts 3, 878–901.
30. Bhanja, P., Modak, A., and Bhaumik, A. (2018).
Supported porous nanomaterials as efficient
heterogeneous catalysts for CO2 fixation
reactions. Chemistry 24, 7278–7297.
43. Liang, J., Chen, R.P., Wang, X.Y., Liu, T.T.,
Wang, X.S., Huang, Y.B., and Cao, R. (2017).
Postsynthetic ionization of an imidazole-
containing metal-organic framework for the
cycloaddition of carbon dioxide and epoxides.
Chem. Sci. 8, 1570–1575.
18. He, Q., O’Brien, J.W., Kitselman, K.A.,
Tompkins, L.E., Curtis, G.C.T., and Kerton, F.M.
(2014). Synthesis of cyclic carbonates from CO2
and epoxides using ionic liquids and related
catalysts including choline chloride-metal
halide mixtures. Catal. Sci. Technol. 4, 1513–
1528.
31. Lyu, J., Zhang, X., Otake, K.I., Wang, X., Li, P., Li,
Z., Chen, Z., Zhang, Y., Wasson, M.C., Yang, Y.,
et al. (2019). Topology and porosity control of
metal-organic frameworks through linker
44. Patel, H.A., Karadas, F., Canlier, A., Park, J.,
Deniz, E., Jung, Y., Atilhan, M., and Yavuz, C.T.
(2012). High capacity carbon dioxide
functionalization. Chem. Sci. 10, 1186–1192.
adsorption by inexpensive covalent organic
polymers. J. Mater. Chem. 22, 8431–8437.
32. Byun, J., and Zhang, K.A.I. (2018). Controllable
homogeneity/heterogeneity switch of
imidazolium ionic liquids for CO2 utilization.
ChemCatChem 10, 4610–4616.
19. Castro-Osma, J.A., Lamb, K.J., and North, M.
(2016). Cr(salophen) complex catalyzed cyclic
carbonate synthesis at ambient temperature
and pressure. ACS Catal. 6, 5012–5025.
45. Subramanian, S., Patel, H.A., Song, Y., and
Yavuz, C.T. (2017). Sustainable nanoporous
benzoxazole networks as metal-free catalysts
for one-pot oxidative self-coupling of amines
by air oxygen. Adv. Sustainable Syst. 1,
1700089.
33. Subramanian, S., Park, J., Byun, J., Jung, Y., and
Yavuz, C.T. (2018). Highly efficient catalytic
cyclic carbonate formation by pyridyl
Salicylimines. ACS Appl. Mater. Interfaces 10,
9478–9484.
20. Xie, Y., Wang, T.T., Liu, X.H., Zou, K., and Deng,
W.Q. (2013). Capture and conversion of CO2 at
ambient conditions by a conjugated
microporous polymer. Nat. Commun. 4, 1960.
46. Steinbauer, J., and Werner, T. (2017).
Poly(ethylene glycol)s as ligands in calcium-
catalyzed cyclic carbonate synthesis.
ChemSusChem 10, 3025–3029.
34. Beyzavi, M.H., Klet, R.C., Tussupbayev, S.,
Borycz, J., Vermeulen, N.A., Cramer, C.J.,
Stoddart, J.F., Hupp, J.T., and Farha, O.K.
(2014). A hafnium-based metal-organic
framework as an efficient and multifunctional
catalyst for facile CO2 fixation and
21. Buyukcakir, O., Je, S.H., Talapaneni, S.N., Kim,
D., and Coskun, A. (2017). Charged covalent
triazine frameworks for CO2 capture and
conversion. ACS Appl. Mater. Interfaces 9,
7209–7216.
47. Liang, L., Liu, C., Jiang, F., Chen, Q., Zhang, L.,
Xue, H., Jiang, H.L., Qian, J., Yuan, D., and
Hong, M. (2017). Carbon dioxide capture and
conversion by an acid-base resistant metal-
organic framework. Nat. Commun. 8, 1233.
22. Talapaneni, S.N., Buyukcakir, O., Je, S.H.,
Srinivasan, S., Seo, Y., Polychronopoulou, K.,
and Coskun, A. (2015). Nanoporous polymers
regioselective and enantioretentive epoxide
activation. J. Am. Chem. Soc. 136, 15861–
15864.
10
Chem 5, 1–11, December 12, 2019

Please cite this article in press as: Subramanian et al., Catalytic Non-redox Carbon Dioxide Fixation in Cyclic Carbonates, Chem (2019), https://
doi.org/10.1016/j.chempr.2019.10.009
48. Whiteoak, C.J., Nova, A., Maseras, F., and Kleij,
A.W. (2012). Merging sustainability with
organocatalysis in the formation of organic
carbonates by using CO2 as a feedstock.
ChemSusChem 5, 2032–2038.
and cocatalyst-free conditions. J. CO2 Util. 16,
384–390.
Catalytic mechanisms proposed by QM/MM
calculations. J. Phys. Chem. C 122, 503–514.
50. Desens, W., and Werner, T. (2016). Convergent
activation concept for CO2 fixation in
52. Roshan, K.R., Kim, B.M., Kathalikkattil, A.C.,
Tharun, J., Won, Y.S., and Park, D.W. (2014).
The unprecedented catalytic activity of
alkanolamine CO2 scrubbers in the
cycloaddition of CO2 and oxiranes: a DFT
endorsed study. Chem. Commun. (Camb.) 50,
13664–13667.
carbonates. Adv. Synth. Catal. 358, 622–630.
49. Liu, M., Li, X., Liang, L., and Sun, J. (2016).
Protonated triethanolamine as multi-
51. Xu, K., Moeljadi, A.M.P., Mai, B.K., and Hirao,
H. (2018). How does CO2 react with styrene
oxide in Co-MOF-74 and Mg-MOF-74?
hydrogen bond donors catalyst for efficient
cycloaddition of CO2 to epoxides under mild
Chem 5, 1–11, December 12, 2019
11