Succinimide-KI: An efficient binary catalyst system for mild, solvent-free cycloaddition of CO2 to epoxides

Article abstract of DOI:10.1016/j.mcat.2019.03.012

In this study, we have for the first time demonstrated that succinimide (SI)-KI is an efficient catalyst system to synthesize propylene carbonate (PC) from carbon dioxide (CO₂) and propylene oxide (PO) at a solvent-free condition. A synergistic effect was observed between SI and KI that largely increased the reaction yield. It enabled the cycloaddition reaction to take place at 70 °C under a low pressure (0.4 MPa) for only 4 h, with a PC yield as high as 97.5%. The overall condition is milder than the reaction catalyzed by most of the other KI involved co-catalyst systems reported in literatures. The excellent catalytic ability was explained by the increased KI solubility in PO due to the presence of SI and the weak acidity of NH in SI which can be enhanced by KI. A reaction mechanism was proposed based on a reaction kinetics study. SI-KI was applicable to cycloaddition of CO₂ with other epoxides. It may offer an inexpensive, environmentally-friendly route to synthesis of propylene carbonate from CO₂.

Full text of DOI:10.1016/j.mcat.2019.03.012

Molecular Catalysis 469 (2019) 111–117
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Succinimide-KI: An efficient binary catalyst system for mild, solvent-free
cycloaddition of CO to epoxides
2
a
a,⁎
a
b
c
a
Qingshuo Li , Haibo Chang , Runming Li , Hongxia Wang , Jichun Liu , Shanhu Liu ,
a
b,⁎
Congzhen Qiao , Tong Lin
a
Institute of Functional Polymer Composites, College of Chemistry and Chemical Engineering, Henan University, Kaifeng, 475004, Henan, China
Institute for Frontier Materials, Deakin University, Geelong, VIC, 3216, Australia
School of Chemical Engineering and Pharmaceutics, Henan University of Science and Technology, Luoyang, 471023, Henan, China
b
c
A R T I C L E I N F O
A B S T R A C T
Keywords:
In this study, we have for the first time demonstrated that succinimide (SI)-KI is an efficient catalyst system to
Carbon dioxide
Succinimide
Epoxide
Cycloaddition reaction
Synergetic effect
synthesize propylene carbonate (PC) from carbon dioxide (CO ) and propylene oxide (PO) at a solvent-free
condition. A synergistic effect was observed between SI and KI that largely increased the reaction yield. It
2
enabled the cycloaddition reaction to take place at 70 °C under a low pressure (0.4 MPa) for only 4 h, with a PC
yield as high as 97.5%. The overall condition is milder than the reaction catalyzed by most of the other KI
involved co-catalyst systems reported in literatures. The excellent catalytic ability was explained by the in-
creased KI solubility in PO due to the presence of SI and the weak acidity of NH in SI which can be enhanced by
KI. A reaction mechanism was proposed based on a reaction kinetics study. SI-KI was applicable to cycloaddition
of CO
2
with other epoxides. It may offer an inexpensive, environmentally-friendly route to synthesis of propylene
carbonate from CO
2
.
1. Introduction
simple, convenient and effective.
Four co-catalyst systems have been reported. The first co-catalyst
system, which is also the most studied one, is hydrogen-bond donor
(HBD), such as β-cyclodextrin [21], cellulose [22], lignin [23], C60
fullerene [24], pentaerythritol [25], sugarcane bagasse [26], tannic
acid [27], formic acid [28], boron phosphate [29] and polyvinyl al-
cohol [30]. HBD can form hydrogen bonds with the oxygen atom of
epoxide, activating the epoxide molecule and therefore accelerating the
reaction. When HBD contains a basic moiety such as triethanolamine
[31], amino alcohols [32], polydopamine [33], amino acid [34,35] and
2
Turning greenhouse gas (CO ) into usable materials has been a hot
topic in the fields of environment, chemistry and materials science
1–5]. A promising method is to produce five-membered cyclic carbo-
nates through cycloaddition of CO to epoxides [3–5]. Five-membered
[
2
cyclic carbonates have been widely used as aprotic polar solvents for
chemical engineering, electrolytes for lithium-ion batteries, and starting
materials for pharmacy, preparation of polycarbonates and poly-
urethane and other new material synthesis [6–8].
CO
high oxidation state and inherent thermodynamic stability of CO
Metal oxides [9], alkali metal halides [5,10], ionic liquids (ILs)
11–14], transition metal complexes [15,16], metal-organic frame-
works (MOFs) [17,18] and N-succinimides [19,20] have been reported
as catalysts to lower the energy barrier and promote the CO -epoxide
cycloaddition reaction. Among them, alkali metal salts have received
particular attention owing to their abundance, non-toxicity and low
price [21–31]. However, alkali-metal salts alone usually have a low
cycloaddition catalytic activity. To improve catalysis activity, the alkali
metal catalysts are either loaded on a porous carrier or combined with a
co-catalyst. Co-catalysis was extensively studied because the process is
2
-epoxide cycloaddition is an energy-intense reaction due to the
2
wool powder [36], the co-catalyst activate the CO molecule in the
2
.
meanwhile, showing higher activity. The second co-catalyst system is
ligand compounds [37–40], including crown-ethers [37] cucurbit[6]
uril [38], polydibenzo-18-crown-6 [39] and lecithin [40] which can
coordinate with alkali metal cations to enhance their solubility. The
ligand compounds can also reduce the interaction between metal cation
and halide ion facilitating nucleophilic attack of halide ions to the ep-
oxide. The third co-catalyst system is Lewis acids (e.g., metalorganic
framework-5 [41], titanocene dichloride [42], titanate nanotube [43]).
[
2
2
Lewis acids are able to enhance the cycloaddition of CO to epoxide
through activation of epoxy rings. The last co-catalyst system is mod-
ified polyethylene glycol such as α, ω-hydroxyl telechelic polyethylene
⁎
Corresponding authors.
E-mail addresses: changhaibo@henu.edu.cn (H. Chang), tong.lin@deakin.edu.au (T. Lin).
https://doi.org/10.1016/j.mcat.2019.03.012
Received 9 November 2018; Received in revised form 27 February 2019; Accepted 9 March 2019
2468-8231/ © 2019 Elsevier B.V. All rights reserved.

Q. Li, et al.
Molecular Catalysis 469 (2019) 111–117
glycol (PEG400) [44] and tetraethylene glycol [45] which can chelate
the potassium cation of KBr or KI, respectively, and meanwhile activate
epoxide oxygen atom through a hydrogen bond interaction. With these
catalysts, the reaction was reported to undertaken at a mild condition
2.3. Solubility of the catalysts in PO
The solubility of KI in PO was measured by weight method at room
temperature. In brief, 830 mg KI (5 mmol) was added into PO (7 ml,
100 mmol). The mixture was then ultra-sonicated for 5 min and sealed
in desiccator for three days to allow sufficient dissolving of KI in PO.
The mixture was centrifuged and 5 ml supernatant liquor was removed.
PO in the supernatant and the rest of the mixture were volatilized and
dried in vacuum at room temperature until the total mass of the two
parts of KI was equal to the initial mass of KI (830 mg). The similar
experiments were carried out to measure the solubility of SI and KI + SI
in PO. The experiments were repeated for three times to ensure accu-
racy and repeatability.
(
2
e.g. 40–60 °C and atmospheric pressure). However, most of the CO -
epoxide cycloaddition catalyzed by the co-catalyst systems containing
alkali metal salts have to be carried out at a harsh reaction condition
(
2
e.g. > 100 °C, > 1 MPa) in the presence of organic solvents. CO -ep-
oxide catalysis cycloaddition under a mild, solvent-free condition has
much less reported.
Recently, N-succinimides was reported to show catalytic activity for
2
epoxide-CO cycloaddition and the catalysis mechanism was highly
dependent on the molecular structure of N-succinimide and catalytic
condition. Jamison et al [19] used N-Bromosuccinimide (NBS) and
benzoyl peroxide as co-catalyst for cycloaddition of CO
reaction was carried out in N, N-dimethylformamide which functioned
as both solvent and CO nucleophilic activation agent. He et al [20]
combined N-iodosuccinimide (NIS) with DBU as a co-catalyst for cy-
cloaddition of CO to epoxide. This reaction was carried out under a
2
to epoxide. The
2
.4. Kinetic experiments
2
The experimental procedure was similar to the above process in the
above-mentioned steel reactor. The difference is that KI, SI and PO were
firstly dissolved in PC (90 mmol) and the cycloaddition was carried out
under the designated temperature and 0.4 MPa (CO was continuously
2
supplied to the reactor during the reaction process.). After the reaction
was completed, the autoclave was cooled in an ice water bath and the
2
mild, free-solvent condition. However, when NIS was replaced with SI,
the SI-DBU showed lower catalysis activity than NIS-DBU. As far as we
knew, using SI and alkali metal as co-catalyst for CO
cloaddition has not been reported in research literature.
2
-epoxide cy-
excess CO1
2
was vented slowly. The conversion was determined by
as the solvent
In this study, we have for the first time demonstrated that SI-po-
tassium iodide (KI) is a very efficient catalyst system for cycloaddition
means of H NMR spectroscopy using CDCl
3
2
of CO to epoxide without using any solvent. A synergistic effect was
2
.5. Reusability test
observed between the two catalyst components. SI-KI enables the re-
action to take place at a mild condition (e.g. 70 °C and 0.4 MPa pres-
sure) with a reaction yield as high as 97.5%. A reaction mechanism was
proposed based on a reaction kinetics study. Our SI-KI system differs to
2
0 mmol PO, 1 mmol SI, and 1 mmol KI were added into the reactor.
The reaction was carried out at 70 °C and 0.4 MPa CO initial pressure
for 4 h. Then, the reactor was cooled in ice water bath, and the excess
CO was vented slowly. A small amount of the product was taken out
for H-NMR analysis. The residual PO was removed out under reduced
pressure and another 20 mmol PO was subsequently added each time
for the next cycle of reaction without separation of PC and catalyst from
the reaction system. For comparison, another series of the simulated
experiments were carried out in the similar condition. In this case, a
desired amount of PC equivalent to that was produced in the actual
reaction during the reusability test was added.
2
2
the previous co-catalysts for CO -epoxide cycloaddition in not only the
2
1
initial material but also the intermediate originated from them. By re-
placing SI with 5 cyclic imides to perform the same reaction, we further
showed that cyclic imide-KI also showed catalytic activity. SI and KI are
both low cost chemicals with low toxicity. SI-KI may form a new cat-
alyst system for effective synthesis of cyclic carbonates from CO and
2
epoxides.
2
. Experimental section
.1. Chemicals
CO with a purity of 99.99% was obtained commercially from a
2.6. Characterizations
2
1H and 13C NMR spectra were measured on a Bruker DPX-400
2
spectrometer and LC–MS spectra were measured on an AmaZon SL
spectrometer. The SEM-EDS analyses were conducted using a Gemini
500 Scanning Electron Microscope (Carl Zeiss, Germany) combined
with Energy-Dispersive X-ray spectrometer (Oxford Instruments,
Abingdon, Oxfordshire, United Kingdom).
local plant. All the epoxides, potassium halides, amides and imides are
of analytical grade, which were purchased from the Aladdin and used as
received. Deuterochloroform (CDCl
Aldrich and used as received.
3
) was purchased from Sigma-
3
. Results and discussion
.1. SI-KI catalytic cycloaddition of CO
Scheme 1 shows the cycloaddition reaction between CO
2
.2. Cycloaddition of CO
2
to epoxides
3
2
to PO
Epoxide-CO cycloaddition reactions were carried out in a 100 ml
2
stainless steel reactor equipped with a magnetic stirrer. In the typical
procedure, the desired amount of potassium halide and cyclic imide
2
and PO.
No solvent was involved in the cycloaddition reaction. The PC yield was
obtained by experiment methods such as distillation. Table 1 shows the
effect of SI-KI content on PC yield. When SI:KI was kept at 1:1 (mol/
mol) meanwhile the contents of SI and KI (based on PO) were
was placed into the reactor and purged with CO
2
for three times. Then
20 mmol PO was added into the reactor and the reactor was sealed. CO
2
was introduced into the reactor until the pressure reached the desired
value. Subsequently, the reactor was heated to the reaction tempera-
ture. After a certain reaction time, the autoclave was cooled in an ice
2
water bath, and the excess CO was vented slowly. Isolated yields were
obtained by distillation or silica gel column chromatography using a
mixture consisting of petroleum ether and ethyl acetate as an eluent.
1
The chemical structures of the products were confirmed by H NMR,
1
3
C NMR and LC–MS. For comparison, two amides (i.e., butyrolactam
BL) and N-ethylacetamide (NEA)) were used separately as a co-catalyst
with KI to conduct the reaction at the same condition.
(
2
Scheme 1. Cycloaddition reaction between CO and PO.
112

Q. Li, et al.
Molecular Catalysis 469 (2019) 111–117
Table 1
a
Influence of catalyst content on the formation of PC .
Entry
SI (mol %)
KI (mol %)
Yield (%)b
TONc
1
2
3
4
5
6
7
8
2.5
4
5
6
4
6
5
4
2.5
4
5
6
6
4
4
5
47.2
87.4
97.5
97.9
88.0
90.7
88.3
87.9
18.9
21.9
19.5
16.3
14.7
22.7
22.1
17.6
2
a) Reaction conditions: PO (20 mmol), CO (initial pressure 0.4 MPa), 70 °C,
and 4 h. b) PC yield. c) TON: mole of synthesized PC per mole of KI.
Scheme 2. The interaction between KI and SI.
simultaneously increased from 2.5 mol % to 5.0 mol% (entries 1–3), the
PC yield increased. At the SI: KI: PO molar ratio of 5: 5: 100, the PC
yield reached a plateau value, being 97.5%. Higher SI and KI contents,
e.g. SI: KI: PO ratio at 6: 6: 100 (mol: mol: mol) (entry 4), led to very
little increase in PC yield. In addition, decreasing one of the compo-
nents (either SI or KI) meanwhile keeping the other component at 5 mol
were performed with and without the presence of the catalyst compo-
nent. For the reaction just catalysed by KI, it had a very low PC yield,
around 2.8%. When SI alone was used to catalyse the reaction, only
trace amount of PC was prepared. These results indicate that SI and KI
%
(relative to PO) also led to decrease in PC yield (entries 7 and 8).
2
form an efficient catalytic system for the cycloaddition of CO to PO
Therefore, SI: KI: PO = 5: 5: 100 (mol: mol: mol) was found to be the
optimal reaction condition, which was used in the further experiments
unless specification.
The effects of reaction temperature, CO pressure and reaction time
2
on PC yield were examined. Fig. 1a shows the effect of reaction pressure
and a catalytic synergistic effect exists between the two catalyst com-
ponents.
In structure, the carbonyl group and oxygen ion in SI possess three
resonance structures (Scheme 2). It was reported that when the car-
bonyl group has an interaction with the potassium cation of KI, activity
of iodine ion can be enhanced [38,40].
on PC yield. With increasing the pressure from 0.1 MPa to 0.4 MPa, the
PC yield increased from 24.0%–97.5%. Further increasing the CO
2
To examine the possible interaction between KI and SI, we mea-
1
pressure conversely lead to a slight decrease in PC yield. A similar trend
was also reported in other catalytic systems [21,22,26,31,38,42], and it
sured the H-NMR spectra of SI in the presence of KI. As shown in Fig. 2,
the NH proton of SI moved up-field to 8.29 ppm in the SI- KI system
when compared with the pure SI (NH proton chemical shift at
8.67 ppm), which may stem from the shield effect of iodine ion.
It was also reported that hydrogen bonding can promote the cy-
was attributed to that increasing CO
CO concentration in the reaction system.
Fig. 1b shows the influence of reaction temperature on PC yield.
2
pressure led to increase of initial
2
Temperature was found to play an important role in the reaction pro-
cess. When the temperature was increased from 40 °C to 70 °C, the PC
yield increased from 8.1%–97.5%. With further increasing the tem-
perature to 80 °C, the PC yield was almost unchanged.
The dependence of the PC yield on reaction time was also in-
vestigated as shown in Fig. 1c. The PC yield increased from
cloaddition of epoxides with CO
hydrogen bonding led to shift of H atom signal to the downfield in
NMR spectra [21,44–46]. Indeed, the H-NMR spectra shows that SI-KI
in the presence of PO led to the NH proton of SI moving downfield to
9.09 ppm (Fig. 2), indicating the formation of hydrogen bonds in the
system.
2
[21–30,46–48] and the formation of
1
H
1
1
6.3%–97.5% when reaction time changed from 1 h to 4 h. The yield
To probe to role of catalysts, we measured the solubility of KI and SI
in the reaction system (Table S2). KI alone had a very small solubility in
PO, being 2.38 mg in 1.4 ml PO (20 mmol). However, the presence of
99.1 mg SI (1 mmol) led to 376.5% increase in KI solubility, i.e. 8.96 mg
in 1.4 ml PO (Supporting information). Considering the low PC yield for
the reaction catalyzed only by KI and much improved PC yield for the
reaction in the presence of SI, both KI and SI play critical roles in cat-
alysis of the reaction.
showed almost no change when further prolonging the reaction time.
Based on the reaction results, the reaction condition of 70 °C, 0.4 MPa
and 4 h was chosen as an optimal condition for further experiments.
We also summarized the literature reports about PO-CO
dition catalysed by KI and co-catalyst (see Table S1). Our SI-KI cata-
lysed cycloaddition is undertaken at 70 °C under a low initial CO
pressure (0.4 MPa) for 4 h. It is relatively milder than the reaction
catalyzed by other KI involved co-catalyst systems reported in litera-
tures, except KI-tetraethylene glycol. However, the KI-tetraethylene
glycol involved reaction consumed much larger amount of the catalyst
than KI-SI, and it took longer reaction time as well.
2
cycload-
2
3.2. Effect of halide and cyclic imide types on catalytic activity
The effect of halide type on catalysis activity was examined. In the
same condition, KBr and KCl showed almost no catalytic activity for the
To probe the role of SI and KI in the reaction, control experiments
Fig. 1. Effect of the reaction conditions on the PC yield (20 mmol PO, 5 mol% SI and 5 mol% KI relative to PO, respectively, were used.). (a) Effect of CO
2
pressure on
the yield of PC (70 °C, 4 h); (b) Effect of reaction temperature on the yield of PC (0.4 MPa, 4 h); (c) Effect of reaction time on the yield of PC (70 °C, 0.4 Mpa).
113

Q. Li, et al.
Molecular Catalysis 469 (2019) 111–117
Table 2
The PC yield of the catalytic cycloaddition reactions a.
b
Yield (%) TONc
Entry Co-catalyst
pK
a
1
2
None
¯¯¯
2.8
97.5
0.56
19.5
9.66[[49]49a]
(SI)
(14.6[[49]49b])
3
4
9.46[[49]49c]
11.4
4.2
2.28
0.84
(MI)
8.3[[49]49a]
(13.4[[49]49d])
(PI)
5
11.97 ± 0.02
[49]49e]
54.9
10.98
[
(CHDC)
6
7
6.0[[49]49f]
¯¯¯
10.8
4.0
2.16
0.8
(NHS)
Fig. 2. The 1H-NMR spectra of SI, SI + KI and SI + KI + PO.
PO-CO2 cycloaddition. When separately combined with SI, they also
showed very low PC yield, 3.0% for SI-KBr catalysed reaction and 2.9%
for the SI-KCl system. This could be attributable to the difference in
nucleophilicity and leaving ability of the halide. I- has stronger nu-
(NMS)
8
9
11.26[49 g]
(24.2[[49]49b])
46.9
28.6
9.38
5.72
(
BL)
cleophilicity
21,23,26,27,31,33,35].
To examine the role of SI on the CO
and
leaving
ability
than
Cl-
and
Br-
14.6 ± 0.2[49 h]
[
(
NEA)
2
-PO cycloaddition, we sepa-
rately combined 5 other cyclic imides, i.e., maleimide (MI), phthali-
mide (PI), 1, 2-cyclohexanedicarboximide (CHDC), N-hydro-
xysuccinimide (NHS) and N-methylsuccinimide (NMS) with KI for
catalytic cycloaddition (Table 2, entries 3–7). The reasons for choosing
these chemicals are because MI and PI have a similar NH group as SI,
while they differ from SI in that MI has a carbon-carbon double bond π-
conjugated system whereas PI has a large π-conjugated system than MI
because of the benzene ring in the molecule. CHDC had a similar mo-
lecular structure to PI except for that the molecule contain a hexane
ring rather than a benzene ring. NHS and NMS belonged to N-sub-
stituted SI derivatives. For NHS, the NH proton is substituted by a hy-
droxyl group. However, NMS has no active proton in the molecule
because of the substitution of the NH proton by methyl group.
a) Reaction conditions: PO (20 mmol), co-cat. (5 mol %), KI (5 mol %), CO
2
(initial pressure 0.4 MPa), 70 °C, and 4 h. b) pKa value of the cyclic imides or
amides in water or dimethyl sulfoxide (brackets)[49]. c) TON: mole of syn-
thesized PC per mole of KI.
2
3.3. SI-KI catalytic cycloaddition of CO to other epoxides
Using SI-KI as catalyst, we also tested the cycloaddition of CO
other epoxides. All the reactions were carried out at 70 °C and initial
CO pressure of 0.4 MPa. The results are listed in Table 3. It was in-
2
to
2
teresting to note that SI-KI catalyst was applicable for all mono-sub-
stituted terminal epoxides. For some reactions, a reaction time slightly
longer than 4 h could be beneficial to get a high reaction yield. Styrene
oxide needs 19 h to attain a high yield (entry 6), which was attributed
to its low reactivity of its β-carbon centre. For cyclohexene oxide (entry
Table 2 lists the PC yield of the cyclic imide-KI catalytic cycload-
dition. In comparison with SI-KI, the other cyclic imides with KI led to a
much lower PC yield (Table 2, entries 3–7). The one without NH proton,
i.e. NMS, had the lowest PC yield among the reaction systems. For
comparison, butyrolactam (BL) and N-ethylacetamide (NEA) were also
7), only 9.6% cyclic carbonate was obtained even if the reaction time
was prolonged to 96 h, presumably due to the high steric hindrance
caused by its ring structure. In the literature [53], the cyclic carbonate
was reported to have cis- and trans- isomers. We also examined the
cyclic carbonate isomers using 1H NMR and C NMR (Supporting In-
formation) and found the produced cyclic carbonate exclusively had a
cis isomer.
2
used with KI to catalyse CO -PO cycloaddition (entries 8 and 9), as
listed in Table 2. The catalytic activity for BL-KI was slightly lower than
that of CHDC–KI, whereas NEA-KI had a lower catalytic activity than
BL-KI.
13
To understand the about structure effect, pKa of the "NH"-type
proton was examined. As listed in Table 2, pKa shows a certain effect on
PC yield. A large or small pKa value resulted in low PC yield. This can
be explained by the ability to form hydrogen bond with the oxygen
atom of epoxide. When acidity of an imide was too strong (low pKa
value), strong hydrogen bond was formed, which baffled the insertion
of carbon dioxide to hydrogen bond [50]. However, when the acidity
was too weak (high pKa value), the weak hydrogen bond failed to ac-
tivate the ring opening of epoxide [51,52].
It should be pointed out that the amides had higher pKa value than
the cyclic imides. The weak acidity could hinder them to effectively
form hydrogen bond with epoxide. NHS has lower pKa value therefore
stronger acidity than PI. However, NHS-KI has much higher PC yield
3.4. Reusability test
The reusability of SI-KI catalyst was examined using the PO-CO2
reaction system. To avoid possible catalyst loss during the separation,
SI-KI wasn’t separated after each cycle. After each reaction cycle, the
residue PO was removed simply by distillation in vacuum, and fresh PO
and CO2 were subsequently introduced into the reactor to perform
another cycle of reaction. As shown in Fig. 3, PC yield decreases gra-
dually with increasing the reaction cycles. To find out the possible re-
actions, we conducted a simulated experiment by using fresh SI-KI with
the amount of PC that was equivalent to the actually produced during
the reusability test added. It was found that PC yield was declined with
increasing PC content in reaction system. Moreover, PC yield in
(
10.8%) than PI-KI. This might be due to the high steric hindrance from
the PI benzene ring and low solubility of PI in PO and PC.
114

Q. Li, et al.
Molecular Catalysis 469 (2019) 111–117
Table 3
Cycloaddition of CO
to other epoxides a
Epoxides
.
2
Entry
1
Products
Time (h)
4
Yield (%)b
97.5
TON c
19.5
1
a
2
a
2
4
98.1
19.6
1
b
2b
3
4
95.5
19.1
1
c
e
2
c
4
5
4
6
96.1
94.3
19.2
18.9
1
d
2
d
1
2
e
6
7
19
96
97.0
9.6
19.4
1.9
1
f
2
f
1
g
2
g
a)Reaction conditions: PO (20 mmol), SI (5 mol%), KI (5 mol%), CO
of KI.
2
(initial pressure 0.4 MPa), 70 °C. b) Isolated PC Yield. c) TON: mole of synthesized PC per mole
2 2
where [PO], [CO ], [KI] and [SI] are the PO, CO , KI and SI con-
centrations, respectively. a and b are the orders of reaction with respect
to KI and SI, respectively. t is reaction time. Here, we assumed that the
concentrations of KI and SI were constant during the reaction and CO
2
was in large excess due to the semi-batch operation. Therefore, Eq. (1)
can be re-written as Eq. (2),
a
b
Rate = kobs1[PO], kobs1 = k [CO2][KI] [SI]
(2)
where and kobs1 was the observed pseudo-first order rate constant for
PO conversion. kobs1 can be obtained from the slope of a linear plot of
the natural logarithm of the changing sample concentration with time
(
Eqs. (3) and (4))
−
d[PO]
dt
Rate =
(3)
(4)
−
ln[PO] = kobs1t
Furthermore, using the Arrhenius equation, the activation energy
for the cycloaddition reaction can be determined from the relationship
Fig. 3. Recyclability test results and the reaction in the presence of equivalent
amount of PC (simulated run). Reaction conditions: PO (20 mmol), SI (5 mol%
PO), KI (5 mol% PO), CO
2
(initial pressure 0.4 MPa), 70 °C, and 4 h.
between kobs1 and temperature, (Eq. (5))
kobs1 = A exp(−Ea/RT)
(5)
6)
repeated cycles was almost the same to that of the reusability test, in-
dicating that the decrease in PC yield in repeated runs was derived from
the dilute effect of PC to PO. In addition, SI-KI system had a good
stability for multiple cycles of reaction.
ln kobs1 = ln A − Ea/RT
(
where R is the universal gas constant (8.314 J mol K−1), and T is the
−1
a
absolute temperature (K), A and E are the pre-exponential factor
−1
(
min ) and the activation energy (kJ/mol), respectively. The natural
3
.5. Reaction kinetics of catalytic CO
2
-PO cycloaddition
logarithm of PO concentration change with time at different tempera-
tures is shown in Fig. S2. Nearly linear relationships were obtained.
Based on the ln [PO] result, kobs1 was calculated using Eq. (5). Fig. 4a
shows the activation energy at different temperatures. By fitting the
data, pseudo-first order rate constant (ln kobs) against the reciprocal
absolute temperature (K/T). Based on the data and Eq. (6), the acti-
vation energy was calculated to be 22.7 kJ/mol.
To explore the reaction mechanism of cyclic carbonate synthesis, a
kinetic study was carried out using PC as solvent. Based on the previous
reports [54], there was a first-order rate dependence of both epoxides
and CO
2
. Thus, the rate equation for the PO-CO
2
coupling reaction
catalysed by SI-KI can be described as Eq. (1),
For comparison, we also measured the activation energy for using KI
alone for catalysis of the reaction (see the data in Fig. S3 & 4a), which
a
b
Rate = k [PO][CO2][KI] [SI]
(1)
1
15

Q. Li, et al.
Molecular Catalysis 469 (2019) 111–117
Fig. 4. (a) Arrhenius plot for the determination of activation energy using SI-KI co-catalyst system (■) and KI alone (●), (b) fitting curve of the natural logarithm of
the observed pseudo-first order rate constant (lnkobs) against the natural logarithm of the catalyst concentration (ln [KI] and ln [SI]).
2
Scheme 3. Proposed mechanism for SI-KI catalysed cycloaddition of CO to epoxide.
was 43.0 kJ/mol. Han et al [55] calculated the energy barrier of the
cycloaddition catalysed by KI alone using a density functional theory
four different concentrations (n) of KI, respectively. KI and SI were
completely dissolved in the mixture of PC and PO before reacting. The
kinetic data were fitted by the natural logarithm of the observed
pseudo-first order rate constant (ln kobs2) against the natural logarithm
of the KI concentrations (ln [KI]) (Eq. (9)) (Fig. S4 & 4b). The loga-
rithmic plot revealed that a is close to 1.0, indicating a first-order de-
pendence on KI.
(
DFT) method. In gas phase, activation energy calculated was
36–41 kcal/mol, which is comparable to our experimental results.
These results that activation energy for the SI-KI involved reaction
was 20.3 kJ/mol lower than that of the KI involved one suggest that the
presence of SI largely reduces the energy barrier to carry out the re-
action.
Apart from the energy barrier, reaction order was estimated. We
assumed that the steady state conditions at the beginning of the reac-
The same method was used to estimate the reaction order of SI (see
details in Fig. S5 & 4b). The kinetic data were fitted by the natural
logarithm of the observed pseudo first-order rate constant (ln kobs3
)
tion, CO
2
and PO concentrations, were more or less constant. Eq. (1)
against the natural logarithm of the SI concentrations (ln [SI]) (Eq.
(10)).
can be rewritten as Eq. (7) or Eq. (8) as needed,
a
b
Rate = kobs2[KI] , kobs2 = k [CO2][PO][SI]
(7)
(8)
ln Rate = ln kobs3 + b ln[SI]
(10)
b
a
Rate = kobs3[SI] , kobs3 = k [CO2][PO][KI]
The kinetic data from the double logarithmic plot revealed that b is
close to 1.0 indicating a first order dependence on SI. This means that
one molecular SI and one molecular KI were actively involved in the
mechanistic cycle.
in which kobs2 and kobs3 represents the rate constant, respectively. By
taking the natural logarithm of Eq. (7), Eq. (9) was obtained.
ln Rate = ln kobs2 + a ln[KI]
(9)
To obtain the reaction order of KI, four reactions were performed at
116

Q. Li, et al.
Molecular Catalysis 469 (2019) 111–117
3.6. Possible mechanism
Chem. 17 (2015) 108–122.
[
13] F.W. Li, L.F. Xiao, C.G. Xia, B. Hu, Tetrahedron Lett. 45 (2004) 8307–8310.
[
14] S.J. Zhang, Y.H. Chen, F.W. Li, X.M. Lu, W.B. Dai, R. Mori, Catal. Today 115 (2006)
It has been reported that a synergistic effect between various HBDs
6
1–69.
(
or Lewis acids) and nucleophile group in the catalyst system can
promote the cycloaddition of carbon dioxide and epoxides
21–30,56–59]. In our case, hydrogen proton in cyclic imide exhibits
[15] F.W. Li, C.G. Xia, L.W. Xu, W. Sun, G.X. Chen, Chem. Comm. 34 (2003) 2042–2043.
[
[
16] A. Decortes, A.M. Castilla, A.W. Kleij, Angew. Chem. Int. Ed. 49 (2010) 9822–9837.
17] R. Babu, A.C. Kathalikkattil, R. Roshan, J. Tharun, D.W. Kim, D.W. Park, Green
Chem. 18 (2016) 232–242.
[
weak acidity and acts as HBD to activate epoxide through forming
hydrogen bond with the oxygen atom of epoxide, while the halide anion
serves as a nucleophilic group.
[18] Z. Zhou, C. He, J. Xiu, L. Yang, C. Duan, J. Am. Chem. Soc. 137 (2015)
15066–15069.
[
19] J.A. Kozak, J. Wu, X. Su, F. Simeon, T.A. Hatton, T.F. Jamison, J. Am. Chem. Soc.
35 (2013) 18497–18501.
20] R. Ma, L.N. He, X.F. Liu, X. Liu, M.Y. Wang, J. CO
1
The reaction mechanism was proposed, as illustrated in Scheme 3.
Firstly, the interaction between SI and KI adduct takes increases the
activity of iodine ion. Secondly, Imide NH activates the epoxide mole-
cule by forming a hydrogen-bond. KI Subsequently attacks the less-
hindered β-carbon atom of the activated epoxide, resulting to ring-
opening of epoxide. The addition of II to a carbon dioxide molecule
generates an alkyl carbonate anion (III). Finally, intramolecular ring
closure of III happen to generate the corresponding cyclic carbonate and
the catalyst for the completion of the catalytic cycle.
The synergistic effect occurs when both iodide anion and SI are
involved in the reaction. The presence of SI not only enhances activa-
tion of epoxide, but also increase the activity of iodine ion due to the
interaction between KI and SI.
[
2
Util. 19 (2017) 28–32.
[21] J.L. Song, Z.F. Zhang, B.X. Han, S.Q. Hu, W.J. Li, Y. Xie, Green Chem. 10 (2018)
337–1341.
22] S.G. Liang, H.Z. Liu, T. Jiang, J.L. Song, G.Y. Yang, B.X. Han, Chem. Commun. 47
2011) 2131–2133.
[23] Z.L. Wu, H.B. Xie, X. Yu, E.H. Liu, ChemCatChem 5 (2013) 1328–1333.
1
[
(
[
[
[
[
24] Y.B. Sun, C.Y. Cao, S.L. Yang, P.P. Huang, C.R. Wang, W.G. Song, Chem. Commun.
5
0 (2014) 10307–10310.
25] L. Zhou, Y. Liu, Z. He, Y. Luo, F. Zhou, E. Yu, Z. Hou, W. Eli, J. Chem. Res. 37 (2013)
102–104.
26] W. Chen, L.X. Zhong, X.W. Peng, R.C. Sun, F.C. Lu, ACS Sustain. Chem. Eng. 3
(2015) 147–152.
27] R. Luo, W. Zhang, X. Zhou, H. Ji, Chin. J. Chem. 35 (2017) 659–664.
[28] J. Tharun, G. Mathai, A.C. Kathalikkattil, R. Roshan, J.Y. Kwak, D.W. Park, Green
Chem. 15 (2013) 1673–1677.
[
29] Z. Xue, X. Zhao, J. Wang, T. Mu, Chem. Asian J. 12 (2017) 2271–2277.
[
30] H. Chang, Q. Li, X. Cui, H. Wang, C. Qiao, Z. Bu, T. Lin, Mol. Catal. 449 (2018)
2
5–30.
[31] B.N. Xiao, J. Sun, J.Q. Wang, C.Y. Liu, W.G. Cheng, Synth. Commun. 43 (2013)
985–2997.
32] T. Werner, N. Tenhumberg, J. CO
33] Z. Yang, J. Sun, X. Liu, Q. Su, Y. Liu, Q. Li, S. Zhang, Tetrahedron Lett. 55 (2014)
3239–3243.
4
. Conclusion
2
[
[
2
Util. 7 (2014) 39–45.
We have shown that SI and KI can form an efficient binary catalyst
2
system for cycloaddition of CO to epoxides. A synergistic effect was
[
[
[
[
34] K.R. Roshan, A.C. Kathalikkattil, J. Tharun, D.W. Kim, Y.S. Won, D.W. Park, Dalton
observed between SI and KI that largely increased the reaction yield. It
enabled the addition reaction to take place at 70 °C and 0.4 MPa pres-
sure and a short reaction time to achieve a high reaction yield (as high
as 97.5%). The SI-KI catalyzed reaction can be conducted in a condition
milder than the same reaction catalyzed by most of the other KI in-
volved co-catalyst systems reported in literature. The binary catalytic
system was applicable to different epoxides. It may offer an in-
expensive, environmentally-friendly route to synthesis of propylene
Trans. 43 (2014) 2023–2031.
35] Z.F. Yang, J. Sun, W.G. Cheng, J.K. Wang, Q. Li, S.J. Zhang, Catal. Commun. 44
(2014) 6–9.
36] H. Chang, Q. Li, X. Cui, H. Wang, Z. Bu, C. Qiao, T. Lin, J. CO2 Util. 24 (2018)
1
74–179.
37] N. Kihara, N. Hara, T. Endo, J. Org. Chem. 58 (1993) 6198–6202.
[38] J.H. Shi, J.L. Song, J.L. Ma, Z.F. Zhang, H.L. Fan, B.X. Han, Pure Appl. Chem. 85
(2013) 1633–1641.
[
[
39] D. Willi, K. Christina, F. Marcus, T. Werner, ChemSusChem 8 (2015) 3815–3822.
40] J. Song, B. Zhang, P. Zhang, J. Ma, J. Liu, H. Fan, T. Jiang, B. Han, Cataly. Today
183 (2012) 130–135.
carbonate from CO
2
.
[
41] J.L. Song, B.B. Zhang, T. Jiang, G.Y. Yang, B.X. Han, Front. Chem. China 6 (2011)
2
1–30.
Acknowledgments
[
42] D. Bai, G. Nian, G. Wang, Z. Wang, Appl. Organometal. Chem. 27 (2013) 184–187.
[
43] A. Chen, T. Zhao, H. Gao, L. Chen, J. Chen, Y. Yu, Catal. Sci. Technol. 6 (2016)
7
80–790.
This work was supported from National Natural Science Foundation
of China (NSFC) program (21204018, U1204518, U1704144), National
Science and Technology support program (2013BAB11B02), and
Natural Science Foundation of Henan Province (15A150033).
[
[
44] S. Kumar, S.L. Jain, Ind. Eng. Chem. Res. 53 (2014) 541–546.
45] S. Kaneko, S. Shirakawa, ACS Sustain. Chem. Eng. 5 (2017) 2836–2840.
[46] A.M. Hardman-Baldwin, A.E. Mattson, ChemSusChem 7 (2014) 3275–3278.
[47] J. Wang, Y. Zhang, ACS Catal. 6 (2016) 4871–4876.
48] L. Wang, G. Zhang, K. Kodama, T. Hirose, Green Chem. 18 (2016) 1229–1233.
49] a) P. Riviere, K. Koga, Tetrahedron Lett. 38 (1997) 7589–7592;
Predicted value from SciFinder;.
[
[
Appendix A. Supplementary data
b) F.G. Bordwell, Accounts Chem. Res. 21 (1988) 456–463;
c) K.R. Darnall, L.B. Townsend, R.K. Robins, Proc. Natl. Acad. Sci. U. S. A. 57
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2019.03.012.
(
1967) 548–553;
d) I. Koppel, J. Koppel, F. Degerbeck, L. Grehn, U. Ragnarsson, J. Org. Chem. 56
1991) 7172–7174;
(
f) D.E. Ames, T.F. Grey, J. Chem. Soc. 1 (1955) 631–636;
g) D.D. Perrin, Dissociation constants of organic bases in aqueous solution,
Butterworths (1965) 187–536;
References
[
1] S. Lin, C.S. Diercks, Y.B. Zhang, N. Kornienko, E.M. Nichols, Y. Zhao, A.R. Paris,
D. Kim, P. Yang, O.M. Yaghi, C.J. Chang, Science 349 (2015) 1208–1213.
2] A. Banerjee, G.R. Dick, T. Yoshino, M.W. Kanan, Nature 531 (2016) 215–219.
3] X.D. Lang, L.N. He, Chem. Rec. 16 (2016) 1337–1352.
h) H. Taniguchi, Y. Kirino, J. Am. Chem. Soc. 99 (1977) 3625–3631
[50] L.F. Xiao, D.W. Lv, D. Su, W. Wu, H.F. Li, J. Clean. Prod. 67 (2014) 285–290.
[51] Y. Zhou, S. Hu, X. Ma, S. Liang, T. Jiang, B. Han, J. Mol. Catal. A Chem. 284 (2008)
52–57.
[52] L. Han, S.J. Choi, M.S. Park, S.M. Lee, Y.J. Kim, M. Kim, B. Liu, D.W. Park, React.
Kinet. Mech. Cat. 106 (2012) 25–35.
[53] C.X. Miao, J.Q. Wang, Y. Wu, Y. Du, L.N. He, ChemSusChem 1 (2008) 236–241.
[54] W. Clegg, R.W. Harrington, M. North, R. Pasquale, Chem. Eur. J. 16 (2010)
6828–6843.
[55] J. Ma, J. Liu, Z. Zhang, B. Han, Green Chem. 14 (2012) 2410–2420.
[56] J. Sun, J.Q. Wang, W.G. Cheng, J.X. Zhang, X.H. Li, S.J. Zhang, Y.B. She, Green
Chem. 14 (2012) 654–660.
[
[
[
[
4] Q.W. Song, Z.H. Zhou, L.N. He, Green Chem. 19 (2017) 3707–3728.
5] J.W. Comerford, I.D.V. Ingram, M. North, X. Wu, Green Chem. 17 (2015)
1
966–1987.
[
[
[
6] Q. Li, J. Chen, L. Fan, X.Q. Kong, Y.Y. Lu, Green Energy Environ. 1 (2016) 18–24.
7] J.H. Clements, Ind. Eng. Chem. Res. 42 (2003) 663–674.
8] F. Suriano, O. Coulembier, J.L. Hedrick, P. Dubois, Polym. Chem. 2 (2011)
5
28–533.
9] K. Yamaguchi, K. Ebitani, T. Yoshida, H. Yoshida, K. Kaneda, J. Am. Chem. Soc. 121
1999) 4526–4527.
[
(
[57] W.G. Cheng, Q. Su, J.Q. Wang, J. Sun, F.T.T. Ng, Catalysts 3 (2013) 878–901.
[58] W.L. Wang, C.Y. Li, L. Yan, Y.Q. Wang, M. Jiang, Y.J. Ding, ACS Catal. 6 (2016)
6091–6100.
[
[
[
10] M. Alves, B. Grignard, R. Mereau, C. Jerome, T. Tassaing, C. Detrembleur, Catal. Sci.
Technol. 7 (2017) 2651–2684.
11] D.H. Lan, N. Fan, Y. Wang, X. Gao, P. Zhang, L. Chen, C.T. Au, S.F. Yin, Chin. J.
Catal. 37 (2016) 826–845.
[59] W.L. Wang, Y.Q. Wang, C.Y. Li, L. Yan, M. Jiang, Y.J. Ding, ACS Sustain. Chem. Eng.
5 (2017) 4523–4528.
12] B.H. Xu, J.Q. Wang, J. Sun, Y. Huang, J.P. Zhang, X.P. Zhang, S.J. Zhang, Green
117