Ruthenium complex immobilized on supported ionic-liquid-phase (SILP) for alkoxycarbonylation of olefins with CO2

Article abstract of DOI:10.1039/d0gc04386a

In this study, the heterogeneously catalyzed alkoxycarbonylation of olefins with CO₂based on a supported ionic-liquid-phase (SILP) strategy is reported for the first time. An [Ru]@SILP catalyst was accessed by immobilization of ruthenium complex on a SILP, wherein imidazolium chloride was chemically integrated at the surface or in the channels of the silica gel support. An active Ru site was generated through reacting Ru₃(CO)₁₂with the decorated imidazolium chloride in a proper microenvironment. Different IL films, by varying the functionality of the side chain at the imidazolium cation, were found to strongly affect the porosity, active Ru sites, and CO₂adsorption capacity of [Ru]@SILP, thereby considerably influencing its catalytic performance. The optimized [Ru]@SILP-A-2 displayed enhanced catalytic performance and prominent substrate selectivity compared to an independent homogeneous system under identical conditions. These findings provide the basis for a novel design concept for achieving both efficient and stable catalysts in the coupling of CO₂with olefins.

Full text of DOI:10.1039/d0gc04386a

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Ruthenium complex immobilized on supported
ionic-liquid-phase (SILP) for alkoxycarbonylation
of olefins with CO2†
Cite this: Green Chem., 2021, 23,
3073
Shi-Ping Xia,a,b Guang-Rong Ding,b,c Rui Zhang, *a Li-Jun Han,b
b,c
Bao-Hua Xu
*
b,c and Suo-Jiang Zhang
In this study, the heterogeneously catalyzed alkoxycarbonylation of olefins with CO2 based on a sup-
ported ionic-liquid-phase (SILP) strategy is reported for the first time. An [Ru]@SILP catalyst was accessed
by immobilization of ruthenium complex on a SILP, wherein imidazolium chloride was chemically inte-
grated at the surface or in the channels of the silica gel support. An active Ru site was generated through
reacting Ru3(CO)12 with the decorated imidazolium chloride in a proper microenvironment. Different IL
films, by varying the functionality of the side chain at the imidazolium cation, were found to strongly
affect the porosity, active Ru sites, and CO2 adsorption capacity of [Ru]@SILP, thereby considerably
influencing its catalytic performance. The optimized [Ru]@SILP-A-2 displayed enhanced catalytic per-
formance and prominent substrate selectivity compared to an independent homogeneous system under
identical conditions. These findings provide the basis for a novel design concept for achieving both
efficient and stable catalysts in the coupling of CO2 with olefins.
Received 28th December 2020,
Accepted 25th March 2021
DOI: 10.1039/d0gc04386a
rsc.li/greenchem
used to promote the conversion of CO2 to fine chemicals.15–19
To achieve the alkoxycarbonylation of olefins using CO2, sup-
Introduction
Converting CO2 into high value-added chemicals and fuels is ported ionic-liquid-phase (SILP)20–22 heterogeneous catalysts
regarded as one of the most promising technologies by which have the potential for use as sustainable heterogeneous cata-
to alleviate the energy and environmental crisis.1–6 To date, a lysts with a homogeneous microenvironment around the active
variety of strategies on using CO2 as a C1 surrogate in the site, but to date these have remained unexplored.
alkoxycarbonylation of olefins have been documented.7–10
SILP candidates can be accessed by the confinement or
Therein, the reductive conversion of CO2 to CO with proper covalent tethering of a well-stabilized thin layer of ionic liquid
reductants11–13 imitating the reverse water gas shift process, (IL) in a highly porous matrix. In principle, the properties,
followed by the alkoxycarbonylation of olefins, has attracted including the structural patterns and functionality at or near
the most attention. In this regard, researchers have focused on interfaces, porosity, and interaction force between the IL film
the construction of novel homogeneous systems consisting of and the carrier, can be readily modified by judiciously coup-
non-noble metals and phosphine-free ligands over the past ling selected ILs and porous materials. Upon incorporation
decade to meet the global goal of sustainable development.14 with
various
transition-metal
complexes23–26
or
However, heterogeneous catalysts are preferred industrially nanoparticles,27–30 SILPs have been employed in flow pro-
due to their low cost and ease of separation and reuse. cesses for gas phase reactions such as hydroformylation,31–34
Therefore, supported transition-metal nanoparticles have been hydrogenation,35–38 the water–gas shift reaction,39–41 the
reverse water–gas shift reaction42 and hydroaminomethyl-
ation.43 Experimental investigations into the approach of the
SILP-mediated alkoxycarbonylation of olefins with CO2 and
aState Key Laboratory of Chemical Engineering, East China University of Science and
Technology, Shanghai 200237, China. E-mail: r.zhang@ecust.edu.cn
bBeijing Key Laboratory of Ionic Liquids Clean Processes, CAS Key Laboratory of
Green Processes and Engineering, State Key Laboratory of Multiphase Complex
Systems, Institute of Process Engineering, Chinese Academy of Sciences, 100190
Beijing, China. E-mail: bhxu@ipe.ac.cn
studies on the reactivity difference between selected IL films
with the same carriers would lead to a better understanding of
how the parameters of SILPs influence the catalyst and further
advance the potential of such strategies.
Based on the present understanding on the homogeneous
Ru3(CO)12-catalyzed alkoxycarbonylation of olefins with CO2 in
imidazolium chlorides, either abnormal C4/5 or normal C2
cSchool of Chemistry and Chemical Engineering, University of Chinese Academy of
Sciences, Beijing 100049, China
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0gc04386a
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 3073–3080 | 3073

View Article Online
Paper
Green Chemistry
the type of IL incorporated (1-R-3-(trimethoxysilyl)propyl imid-
azolium chloride (R
= Me (X = A), C12H25 (X = B),
(CH2CH2O)3CH3 (X = C)) and m indicates the amount of
Ru3(CO)12 added (m = 1, Ru3(CO)12 = 19 mg; m = 2, Ru3(CO)12
38 mg; m = 3, Ru3(CO)12 = 76 mg).
=
The porosity of [Ru]@SILP-X-2 (X = A, B, C) and the silica
gel support was investigated through carrying out nitrogen-
sorption isotherms at 77 K (Fig. 1). All of the samples dis-
played a typical type IV isotherm featuring a rapid uptake at a
low relative pressure at P/P0 < 0.01, followed by a sharp step at
0.40 < P/P0 < 0.80, which is characteristic of a mesoporous skel-
eton (Fig. 1a). The Brunauer–Emmett–Teller (BET) surface area
of [Ru]@SILP-X-2 was determined to be 324 m2 g−1 (X = A),
384 m2 g−1 (X = B), 113 m2 g−1 (X = C), respectively, suggesting
pore accessibility therein, although the values were all lower
Scheme 1 Schematic illustration of the preparation of [Ru]@SILP
hybrids.
mode NHC carbene ligated Ru species have been detected
in situ as the active site.44–48 Besides this, the presence of Cl−
facilitates the deprotonation of the transient ruthenium
hydride species to enable the activation of CO2 by coordi-
compared to those of the silica gel support (514 m2 g−1
,
nation. It is also supposedly crucial for the formation of
Table S1†). Besides this, the average pore size of
[Ru]@SILP-X-2 is almost the same (3.6 nm) with different IL
moieties (Fig. 1b), but the volume decreases in the order of A
−
[Ru3(Cl)(CO)12−n
]
(n = 1–3) species, where the strong anti-
effect of Cl− therein accelerates CO dissociation and olefin
association for alkoxycarbonylation.49–53 We thus envisioned
the design and fabrication of novel [Ru]@SILP hybrids in an
adaptable and controlled manner, wherein task-specific imida-
zolium chloride are covalently decorated in the porous
support, followed by the immobilization of Ru via direct reac-
tion with Ru3(CO)12 (Scheme 1). The quantity of Ru active sites
can be controlled through the ratio of Ru3(CO)12 to imidazo-
lium units located within a proper microenvironment. As a
proof of concept, commercially available porous silica gel was
examined as a carrier. The difference between representative
functionalities (Me (X = A), C12H25 (X = B), (CH2CH2O)3CH3
(X = C)) at the side chain of the imidazolium ring on the
physicochemical properties of the SILP and thereby the cata-
lytic performance of [Ru]@SILP was investigated. Impressively,
[Ru]@SILP-A-2 exhibits durable catalytic activity, recyclability,
and applicability in the alkoxycarbonylation of olefins with
CO2, which was performed in the presence of methanol
(MeOH) and provides methoxyl carbonylate products in good
yields. Moreover, [Ru]@SILP-A-2 exhibits enhanced catalytic
performance compared to an individual homogeneous cata-
lyst, attributed to its highly dispersed catalytic sites, porous
framework with high surface area, and large CO2 adsorption
capacity and enrichment. Besides this, it shows prominent
substrate selectivity, wherein small cyclic olefins and short-
chain linear olefins are preferred, corresponding to the steric
hindrance effect of the pore channels. These findings provide
a platform for novel design concept to achieve both efficient
and stable SILP catalysis in the coupling of CO2 with olefins.
(0.443 cm3 g−1) > B (0.404 cm3 g−1) > C (0.128 cm3 g−1
)
(Table S1†). We also found that both the BET surface area and
Barrett–Joyner–Halenda (BJH) pore size distribution of
[Ru]@SILP-A-m deceases slightly with increasing Ru content
(Fig. S1†), wherein the maximum Ru per mass unit of SILP[Cl]-
A was controlled at less than 2.5 wt% (Table S2†). These results
suggest that both the surface and channel of the silica gel is
partially occupied by the ILs and Ru.
The morphologies of [Ru]@SILP-X-2 and the silica gel
support were characterized via a combination of scanning elec-
tron microscopy (SEM) and transmission electron microscopy
(TEM) (Fig. 2). Apparently, the silica gel exhibits an amorphous
and porous structure consisting of some aggregates of small
particles (Fig. 2a). In comparison, the surface of [Ru]@SILP-X-2
is much rougher since the size and quantity of the aggregates
increased, suggesting the IL is partially immobilized on the
silica gel. However, no significant difference in morphology
was detected between [Ru]@SILP-A-2, [Ru]@SILP-B-2 and
[Ru]@SILP-C-2. The EDS elemental mapping clearly reveals
that Ru is uniformly located on the surface of the hybrids
(Fig. 2b), thus indicating that both the IL moieties and Ru
species are highly dispersed on the silica gel support. In
addition, the density of Ru in [Ru]@SILP-B-2 bin the EDS
Results and discussion
Synthesis and characterization of [Ru]@SILP hybrids
The synthesis of [Ru]@SILP hybrids, denoted as [Ru]@SILP-X-m,
was conducted through the chemical functionalization of
silica gel with IL-X to form SILP[Cl]-X, followed by reaction
Fig. 1 (a) N2 adsorption isotherms and (b) BJH pore size distribution of
with a certain amount of Ru3(CO)12. In the label, X represents silica gel; [Ru]@SILP-A-2; [Ru]@SILP-B-2 and [Ru]@SILP-C-2.
3074 | Green Chem., 2021, 23, 3073–3080
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
Fig. 3 FT-IR spectra of IL-A, silica gel, SILP[Cl]-A and [Ru]@SILP-A-2.
vibration of Si–O–Si, and the peaks at 804 and 470 cm−1
belong to the symmetric stretching vibration and bending
vibration of Si–O, respectively. Besides these, a broad, asym-
metric feature is detected at 3457 cm−1, which can be attribu-
ted to O–H vibrations of either the silanol or water in silica
gel. The main absorption peaks of IL-A can be observed at
1575 and 1461 cm−1, assigned to the CvN and CvC stretch-
ing vibrations of the imidazole ring, respectively. Additionally,
typical aliphatic C–H stretching can be observed at 2948 and
2842 cm−1. As expected, the FT-IR spectrum of SILP[Cl]-A
shows both characteristic peaks of silica gel and IL-A,
suggesting the successful incorporation of IL-A into the silica
gel. In comparison, two new absorption peaks at 2051 and
1980 cm−1 are present in the spectrum of [Ru]@SILP-A-2,
which can be assigned to Ru-hydride-carbonyl-carbene species
according to the literature.44 The other spectra of IL-X, SILP
[Cl]-X and @SILP-X-2 (X = B and C) are displayed and com-
pared in Fig. S2.† Both display identical variations in the func-
tional groups compared to when X = A.
Fig. 2 The morphology and microstructure of [Ru]@SILP-X-2 (a) SEM
images of silica gel and [Ru]@SILP-X-2 (X = A, B and C); (b) EDS analysis
of Ru in [Ru]@SILP-X-2 (X = A, B and C); (c) HR-TEM images of
[Ru]@SILP-X-2 (X = A, B and C).
The surface modification was further studied from the 29Si
and 13C cross-polarization magic angle spinning nuclear mag-
netic resonance (CP-MAS NMR) spectra of [Ru]@SILP-A-2, SILP
[Cl]-A and silica gel (Fig. 4). The spectrum of silica gel displays
three 29Si resonances at −111 (Q4), −102 (Q3) and −91 (Q2)
ppm, which can be assigned to Si(OSi)4, Si(OSi)3OH and Si
(OSi)2(OH)2 species, respectively (Fig. 4A-a). Upon sequential
functionalization with IL-A and incorporation of Ru species,
elemental mapping is apparently higher than that in
[Ru]@SILP-A-2 and [Ru]@SILP-C-2. This demonstrates that
[Ru]@SILP-B-2 has the maximum Ru per mass unit of SILP[Cl]-X,
which is consistent with the ICP results (Table S2†). TEM
images of [Ru]@SILP with the different ILs show that the Ru
species has the most even size and is distributed uniformly in
[Ru]@SILP-A-2 compared to the other two samples (Fig. 2c).
Besides this, the lattice fringe of Ru in [Ru]@SILP-A-2 is
clearly observed to have a spacing of 0.195 nm,54 which further
confirms the successful immobilization of Ru. In contrast to
their round shape in [Ru]@SILP-A-2 and [Ru]@SILP-B-2, the
Ru in the [Ru]@SILP-C-2 exhibits randomly ellipsoid shapes
with a particle size of more than 20 nm, indicating that the Ru
therein is agglomerated.
The FT-IR spectra of IL, SILP[Cl] and [Ru]@SILP were
recorded to detect the potential variation of the functional
groups therein. The set of IL-A, SILP[Cl]-A and [Ru]@SILP-A-2
was selected as representatives for discussion and their spectra
are displayed in Fig. 3. Meanwhile, the spectrum of silica gel
was also studied for comparison. The broad band at
Fig. 4 Solid-state NMR studies. (A) Solid-state 29Si CP-MAS NMR
spectra of (a) silica gel and (b) [Ru]@SILP-A-2; (B) solid-state 13C
1236–1043 cm−1 can be assigned to the asymmetric stretching CP-MAS NMR spectra of (c) SILP[Cl]-A and (d) [Ru]@SILP-A-2.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 3073–3080 | 3075

View Article Online
Paper
Green Chemistry
both the Q3 and Q2 peaks disappear, while the Q4 peak is lower than those of [Ru]@SILP-B-2 (197.6 and 199.1 eV)
retained (Fig. 4A-b), indicating the condensation of singular (Fig. S4†). This suggests that there are more chloride anions at
and geminal silanol groups with the trimethoxysilane moiety. or near the surface of [Ru]@SILP-A-2 and [Ru]@SILP-C-2 com-
Besides this, the spectrum of [Ru]@SILP-A-2 displays three pared to [Ru]@SILP-B-2. In addition, the binding energy of
additional signals at −52 (T1), −61 (T2), and −67 (T3) ppm, Ru 3p3/2 and 3p1/2 can be observed at 462.4 and 484.5 eV,
assigned to R-Si(OSi)(OMe)2, R-Si(OSi)2(OMe), and R-Si(OSi)3, respectively, confirming the presence of Ru (0) in
respectively. These results unambiguously confirm the [Ru]@SILP-A-2 (Fig. 5d).55 Remarkably, both peaks show an
covalent attachment of IL-A on the surface and in the channels upshift of ca. 0.5 eV compared to those of Ru3(CO)12,
of the silica gel support. Noteworthily, no resonance was suggesting the formation of a Ru active site with a strong inter-
detected at the lower field over −50 ppm, suggesting the action between the imidazolium unit and Ru(0).
absence of physically adsorbed silane molecules. Meanwhile,
The amount of ILs decorated in the [Ru]@SILP-X-2 was
the 13C CP-MAS NMR spectrum of SILP[Cl]-A further con- quantified by elemental analysis (Table S3†). Taking nitrogen
firmed the successful modification of silica gel with IL-A as the indicator atom, the IL concentration in [Ru]@SILP-X-2
(Fig. 4B-c). Evidently, typical imidazolium 13C peaks at 114.1 was recorded as 29.4 wt% (X = A), 24.3 wt% (X = B), and
(C3, C4) and 128.3 (C2) ppm were detected. Besides these, the 41.7 wt% (X = C), respectively. Thus, the corresponding molar
signals at 26.0 (C1), 0.34 (C7), 14.51 (C6) and 41.8 (C5) ppm can percentages of IL-X decorated on the surface and in the chan-
be attributed to the methyl (C1) and three methylene (C5, C6, nels of the silica gel are 88.2% (X = A), 56.0% (X = B), and
C7) groups in IL-A, respectively. Upon incorporation of 98.1% (X = C) under identical conditions. The apparently low
Ru3(CO)12, a new peak at 210.3 ppm appears in the 13C spec- value for [Ru]@SILP-B-2 can be probably attributed to the
trum of [Ru]@SILP-A-2 (Fig. 4B-d), which can be assigned to hydrophobic characteristics of IL-B, thereby leading to signifi-
the ligated carbonyl group at the Ru site.
cant resistance upon diffusion onto the surface and into the
X-ray photoelectron spectroscopy (XPS) was employed to channels of the hydrophilic silica gel. All these results are con-
further characterize the chemical composition and state of the sistent with the BET characterization. [Ru]@SILP-B-2 exhibits a
newly prepared hybrids (Fig. 5). The XPS survey profile of higher surface area compared to [Ru]@SILP-A-2, which can be
[Ru]@SILP-A-2 features a combination of characteristic peaks of attributed to the lower percentage of IL decoration in the
SILP[Cl]-A and Ru (Fig. 5a). Noteworthily, the quite weak peak former. However, the flexible hydrocarbon C12 chains on the
of Ru 3p in the profile is due to a low concentration of Ru. The imidazole ring of IL-B can seriously block the channels,
two resolved peaks of N 1s at around 399.7 and 401.8 eV can leading to a poorer pore volume. Similarly, [Ru]@SILP-C-2
be respectively assigned to the amine nitrogen atom and exhibits the lowest surface area and porosity when the
ammonium nitrogen atom within the imidazolium ring percentage of decorated IL is almost the same as that of
(Fig. 5b). Besides these, the peaks at 197.4 eV (Cl 2p3/2) and [Ru]@SILP-A-2.
198.8 eV (Cl 2p1/2) in the high-resolution spectrum suggest the
The thermal stability of the newly prepared [Ru]@SILP-X-2
presence of chloride anions (Fig. 5c). These values were found and the silica gel support was investigated by carrying out
to be almost the same as that for [Ru]@SILP-C-2, but relatively thermogravimetric analysis (TGA) under a nitrogen atmo-
sphere (Fig. 6). The first weight loss of all of the samples
occurred at about 50 °C, which can be attributed to the phys-
ically adsorbed water and residual organic solvents present
from the preparation of the hybrids being lost. The second
weight loss of [Ru]@SILP-X-2 in the range of between 230 and
350 °C can likely be attributed to the partial decomposition of
the ILs, such as the end group alkyl chains decorated on the
imidazolium ring. Singh56 supposed that this follows a phe-
Fig. 5 XPS studies. (a) XPS survey spectra of silica gel, SILP[Cl]-A and
[Ru]@SILP-A-2; (b) N 1s; (c) Cl 2p and (d) Ru 3p in [Ru]@SILP-A-2.
Fig. 6 TGA curves of silica gel and [Ru]@SILP-X-2 (X = A, B and C).
3076 | Green Chem., 2021, 23, 3073–3080
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
nomenological ‘hinged spring’ model, wherein the surface with CO2 as a C1 surrogate, together with some controlled
oxygen atom of the silica gel interacts with the C–H groups of experiments for comparison. Taking the methoxycarbonyla-
the imidazolium ring to ‘hinge’ it better to the pore walls tion of cyclohexene (1a) with CO2 as the benchmark reaction
while the end alkyl chains will be exposed outside the pore. (Fig. 8a), [Ru]@SILP-A-2 displays outstanding catalytic per-
Thus, the alkyl chains are more easily broken under thermoly- formance under 4 MPa CO2 at 160 °C for 20 h, leading to the
sis than the imidazolium ring. Noteworthily, the decompo- formation of ester (2a) in 74% yield with a TON value of 81
sition of ILs is accompanied by that of Ru3(CO)12, according to (Fig. 8b). In comparison, neither [Ru]@SILP-B-2 nor
the literature.57 The last weight reduction in the range of [Ru]@SILP-C-2 provide a comparable yield under identical
between 350 °C and 800 °C is likely due to the full decompo- conditions. We supposed that the higher loading amount
sition of the ILs. As a result, the weight loss of [Ru]@SILP-X-2 of Ru but lower percentage of decorated IL in
was calculated to be up to 21.5% (X = A), 17.9% (X = B), and [Ru]@SILP-B-2 meant that it does not have sufficient reactive
29.5% (X = C), values that are less than the mass of the respect- sites for the reaction. Besides this, the flexible hydrocarbon
ive IL and Ru3(CO)12 loaded. We therefore suppose that the ILs C12 chains of IL-B not only block the pores but also inhibit
and Ru3(CO)12 are partially confined in the inner pores of the the access of reactants to the catalytic active sites. The
silica gel, which has good pyrolysis resistance. Collectively, latter is also responsible for the poor performance of
these results demonstrate that the immobilized ILs are ther- [Ru]@SILP-C-2. We also found that a appropriate loaded
mally stable at temperatures of lower than 230 °C, ensuring amount of Ru with SILP[Cl]-A as the support was crucial to
the reliability of catalytic performance screening on [Ru]@SILP the reactivity (Fig. 8c). As a result, either insufficient or
within this limit.
excess loading of Ru is not conducive to the number of active
The CO2 adsorption capability of the silica gel and sites per unit mass of substrate. Varying of the reaction con-
[Ru]@SILP-X-2 were investigated and the CO2 adsorption iso- ditions, such as the solvents, temperatures, chloride addi-
therms were measured at pressures of up to 1.0 bar at 273 and tives and pressure of CO2, was examined (Fig. 8d and
298 K, respectively (Fig. 7). Clearly, the silica gel exhibits a rela- Fig. S5†). We supposed that a similar Ru species was formed
tively higher adsorption capacity, which can be attributed to to that in either the water–gas shift reaction58 or the reverse
the strong affinity between the CO2 and the hydroxyl groups water–gas shift reaction42 catalyzed by [Ru]@SILP, wherein
decorated in the channel of the porous silica gel and higher the presence of both Cl− and CO is highly likely. Under opti-
pore volume therein. Upon the introduction of ILs, the CO2 mized conditions, a higher catalytic activity than the homo-
uptake of [Ru]@SILP-X-2 decreased compared to that of the geneous systems can be achieved by [Ru]@SILP-A-2, leading
silica gel support (Fig. 7a). Besides this, [Ru]@SILP-A-2 has a to an 87% yield of 2a with a TON value of 94. In addition, CO
better storage capacity than [Ru]@SILP-B-2 and [Ru]@SILP-C-2 and H2 can be detected in the gas mixture, leading to TONs
under identical conditions. As expected, the CO2 uptake of 3.6 and 1.8, respectively (Fig. S6†).
capacity of all of the samples decreased when the temperature
increased from 273 to 298 K. The isosteric heat of adsorption
(Qst) value of CO2 for [Ru]@SILP-X-2 was subsequently calcu-
lated using the virial method. Evidently, [Ru]@SILP-A-2 has
the highest value of 21.3–29.0 kJ mol−1 compared to the other
two hybrids (Fig. 7b).
Catalytic performance evaluation for the alkoxycarbonylation
of olefins with CO2
The catalytic performance of the [Ru]@SILP-X hybrids was sub-
sequently evaluated towards the alkoxycarbonylation of olefins
Fig. 8 Catalytic performance evaluation. (a) Scheme of the benchmark
reaction. Reaction conditions: 1a (2 mmol), catalyst (2.8 mol% Ru), LiCl
(4 mmol, 2 equiv.), MeOH (4 ml), CO2 (4 MPa), 160 °C, 20 h; yield was
determined by GC-FID analysis using isooctane as an internal standard;
(b) reactivity comparison among [Ru]@SILP-X-2; (c) reactivity compari-
son among [Ru]@SILP-A-m;
m = 1 (0.028 mmol Ru g m = 2
−1),
Fig. 7 CO2 adsorption capability evaluation. (a) CO2 adsorption iso-
therms of silica gel and [Ru]@SILP-X-2 recorded at 273 and 298 K;
(b) the respective isosteric heats of adsorption for CO2 of [Ru]@SILP-X-2.
(0.056 mmol Ru g−1), m = 3 (0.104 mmol Ru g−1); (d) other cross
experiments.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 3073–3080 | 3077

View Article Online
Paper
With the optimal [Ru]@SILP-A-2 catalyst in hand, the
Green Chemistry
olefin tolerance of this newly developed protocol was next
examined (Fig. 9). All reactions were conducted in the pres-
ence of [Ru]@SILP-A-2 (2.8 mol% Ru) at 160 °C under 4 MPa
CO2 for 20 h. We found that the small cyclic olefin performed
better than the large one (1b vs. 1c), which might be attribu-
ted to the difference in the steric hindrance in the pores of
[Ru]@SILP-A-2. Besides this, the yield of ester from the long-
chain terminal olefin (1e) was worse than that from the short-
chain olefin (1d) due to high diffusion resistance in the pore
channels. To our surprise, a branched ester is slightly pre-
ferred over a linear one in both cases. It is worth noting that
the opposite selectivity in the carbonylation of olefins with
Fig. 10 Examination of the recyclability of [Ru]@SILP-A-2. Reaction
conditions: 1a (2 mmol), [Ru]@SILP-A-2 (2.8 mol% Ru), MeOH (6 ml),
CO2 (4 MPa), 160 °C, 20 h; yield was determined by GC-FID analysis
using isooctane as an internal standard.
14
CO2 has been well documented using both Ru3(CO)12 and
59
Rh-PR3
homogeneous catalysts. We suppose that this is
probably due to the steric confinement effect of the support
and the acidity of the silica gel.60 Besides this, significant iso-
merization was detected in the reaction of 2-octene (1e′),
leading to an identical result to that of 1-octene (1e). It is
rationally supposed that the isomerization of 2-octene is
much faster than methoxycarbonylation with CO2 using this
protocol. Finally, diene substrates, such as isoprene (1f) and
1,3-cyclohexadiene (1g), were also examined, which provided
the corresponding mono carboxylic ester rather than double
methoxycarbonylation product with the second olefin site
being reduced.
the SEM images showed no obvious changes in the used
[Ru]@SILP-A-2, suggesting that the silica skeleton is less affected
during the recycling (Fig. S9†). However, the elemental analysis
results show that the nitrogen content decreased by 0.55%,
suggesting the leaching of IL-A from the catalyst (Table S3†).
Conclusions
The recyclability of [Ru]@SILP-A-2 (Fig. 10) and
[Ru]@SILP-B-2 (Fig. S7†) were finally examined by repeated
addition of 1a after each run. After four runs, the yields of
ester product 2a were 82% and 35%, respectively, suggesting
In summary, the first example of heterogeneous catalysis in
the alkoxycarbonylation of olefins with CO2 is reported. The
[Ru]@SILP catalyst was synthesized by immobilization of a
ruthenium complex on SILP, wherein imidazolium chloride
was used to decorate the surface or channels of the silica gel
support. Different IL films, by varying the functionality at the
imidazolium cation, strongly affect the porosity, active Ru
sites, and CO2 adsorption capacity of [Ru]@SILP, thereby con-
siderably affecting the activity of the resulting catalysts. The
optimized [Ru]@SILP-A-2 catalysts displays enhanced catalytic
performance compared to an independent homogeneous
system for the alkoxycarbonylation of olefins with CO2 under
identical conditions, attributed to the highly dispersed cata-
lytic sites, porous framework with high surface area, and large
CO2 absorption capacity and enrichment. Besides this, it
shows prominent substrate selectivity, wherein small cyclic
olefins and short-chain linear olefins were preferred, corres-
ponding to the steric hindrance effect of the pore channels.
The fabrication of heterogeneous IL-based catalysts suitable
for the coupling of CO2 with olefins based on this newly devel-
oped novel design concept is now under way in our lab.
only
a
slight reduction in the catalytic activity of
[Ru]@SILP-A-2 and [Ru]@SILP-B-2. To address this issue, full
characterization of the used [Ru]@SILP-A-2 was conducted.
FT-IR study demonstrated that the characteristic peaks of the
silica gel, IL-A and Ru complexes were retained in their
spectra, compared to the fresh hybrids (Fig. S8†). Besides this,
Experimental section
Fig. 9 Screening the olefin scope of the [Ru]@SILP-A-2-catalyzed
alkoxycarbonylation with CO2. Reaction conditions:
1 (2 mmol), Synthesis of SILP[Cl]-X
[Ru]@SILP-A-2 (2.8 mol% Ru), LiCl (4 mmol, 2 equiv.), MeOH (6 ml), CO2
(4 MPa), 160 °C, 20 h; the yield was determined by GC analysis using the
area normalization method. Only linear products are shown, and the
bold bonds represents the hydrogenation of the double bond.
The synthesis of 1-R-3-(trimethoxysilyl)propyl imidazolium
chloride (IL-X; R = Me (X = A), C12H25 (X = B), (CH2CH2O)3CH3
(X = C)) is described in detail in the ESI.† SILP[Cl]-X was sub-
3078 | Green Chem., 2021, 23, 3073–3080
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
sequently synthesized by chemical functionalization of silica then opened inside the glovebox and a mixture of fresh 1a
gel with IL-X according to previous reports, wherein X denotes (2 mmol), MeOH (1 ml) and isooctane (1 ml) was added for
the type of IL incorporated. In a typical procedure, silica gel the next run. Before the start of each run, one drop was taken
(5 g) and IL-X (8.9 mmol) were dispersed in toluene (50 mL) in out from the mixture with a syringe to detect the distribution
a 100 ml Schlenk bottle. The mixture was refluxed at 110 °C of the composition by GC analysis. After the appointed runs,
for 24 h under a nitrogen atmosphere. After filtration and the used catalyst was separated from the mixture upon fil-
washing with MeOH (6 × 10 ml), the solid was subjected to tration, which was then washed with MeOH (6 × 20 ml) and
Soxhlet extraction with MeOH (100 ml) for 24 h to remove dried under high vacuum at 80 °C for 24 h.
excess ILs. The obtained SILP[Cl]-X was dried at 90 °C for 12 h
under high vacuum and then stored in a glove box.
Conflicts of interest
Synthesis of [Ru]@SILP-X-m
There are no conflicts to declare.
[Ru]@SILP-X-m was synthesized by reacting SILP[Cl]-X (2 g)
with a certain amount of Ru3(CO)12, based on the IL loading
(m = 1, Ru3(CO)12 = 19 mg; m = 2, Ru3(CO)12 = 38 mg; m = 3,
Ru3(CO)12 = 76 mg). Taking [Ru]@SILP-X-2 as an example,
Acknowledgements
degassed THF (8 mL) was introduced into a mixture of SILP
Financial support from the Excellent Young Scientists Fund
[Cl]-X (2 g) and Ru3(CO)12 (38 mg, 0.06 mmol) in a 50 mL
(22022815), the Key Research Program of Frontier Sciences,
Schlenk flask under a nitrogen atmosphere. After stirring at
CAS (ZDBS-LY-JSC022), the National Natural Science
Foundation of China (U1704251, 22078336, 21374029), and
120 °C for 16 h, the solid was isolated by filtration and washed
thoroughly with THF (3 × 10 ml) and MeOH (3 × 10 ml) to
Zhengzhou High Level Talent Certificate (20180200052) is
remove unreacted Ru3(CO)12. The residue was then dried
gratefully acknowledged.
under vacuum at 80 °C for 12 h to give [Ru]@SILP-X-2.
General procedure for the catalytic performance examination
of [Ru]@SILP
References
A typical experiment was carried out under argon using a stan-
1 R.-P. Ye, J. Ding, W. Gong, M. D. Argyle, Q. Zhong, Y. Wang,
dard glovebox, and [Ru]@SILP-X-2 (2.8 mol% Ru), LiCl
C. K. Russell, Z. Xu, A. G. Russell, Q. Li, M. Fan and
(170 mg, 4 mmol), MeOH (4 ml) and olefins (2 mmol) were
Y.-G. Yao, Nat. Commun., 2019, 10, 5698–5712.
2 W. Zhou, J. Guo, S. Shen, J. Pan, J. Tang, L. Chen, C. Au
and S. Yin, Acta Phys.–Chim. Sin., 2020, 36, 1906048–
1906058.
3 A. Weilhard, M. I. Qadir, V. Sans and J. Dupont, ACS Catal.,
2018, 8, 1628–1634.
4 M. I. Qadir, F. Bernardi, J. D. Scholten, D. L. Baptista and
J. Dupont, Appl. Catal., B, 2019, 252, 10–17.
added to a 30 mL autoclave equipped with a Teflon lining.
Then, CO2 (40 bar) was introduced, and the autoclave was
heated to 160 °C. After reaction for 20 h, the mixture was
cooled down and the pressure was released. The reaction solu-
tion was diluted with MeOH, followed by filtration using an
organic filter. The filtrate was then analyzed by GC using iso-
octane as an internal standard. GC analysis was performed using
an Agilent GC-7890B spectrometer equipped with a capillary
5 J. Klankermayer, S. Wesselbaum, K. Beydoun and
column (DB-FFAP, 30 m × 0.32 mm) using a flame ionization
W. Leitner, Angew. Chem., Int. Ed., 2016, 55, 7296–7343.
6 J. Artz, T. E. Mueller, K. Thenert, J. Kleinekorte, R. Meys,
A. Sternberg, A. Bardow and W. Leitner, Chem. Rev., 2018,
118, 434–504.
7 Q. Liu, L. Wu, R. Jackstell and M. Beller, ChemCatChem,
2014, 6, 2805–2809.
8 P. H. Gehrtz, V. Hirschbeck and I. Fleischer, Chem.
detector. The temperatures of the injector and FID were both
controlled at 250 °C. The components were separated accord-
ing to the following program: an initial temperature of 100 °C
held for 2 min, heated to 200 °C at a rate of 25 °C min−1 and
then held for 5 min. The calculation equations for ester yield
and turnover number (TON) are as follows:
Commun., 2015, 51, 12574–12577.
9 S. C. Stouten, T. Noel, Q. Wang, M. Beller and V. Hessel,
Catal. Sci. Technol., 2016, 6, 4712–4717.
Y
ester% ¼ ðmoles of the produced ester
=moles of initial olefinsÞ ꢀ 100%
TONester ¼ moles of the produced ester
=moles of Ru3ðCOÞ12in the catalyst
10 X. Zhang, C. Shen, C. Xia, X. Tian and L. He, Green Chem.,
2018, 20, 5533–5539.
11 T. Fujihara, T. H. Xu, K. Semba, J. Terao and Y. Tsuji,
Angew. Chem., Int. Ed., 2011, 50, 523–527.
12 S. H. Li, W. M. Yuan and S. M. Ma, Angew. Chem., Int. Ed.,
2011, 50, 2578–2582.
Recyclability examination of [Ru]@SILP
The recyclability of [Ru]@SILP was examined according to the
literature.61 Specifically, after the first run, the mixture was 13 C. M. Williams, J. B. Johnson and T. Rovis, J. Am. Chem.
cooled down and the pressure released. The autoclave was
Soc., 2008, 130, 14936–14937.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 3073–3080 | 3079

View Article Online
Paper
Green Chemistry
14 L. Wu, Q. Liu, I. Fleischer, R. Jackstell and M. Beller, Nat. 38 M. J. Schneider, M. Haumann and P. Wasserscheid, J. Mol.
Commun., 2014, 5, 3091–3096. Catal. A: Chem., 2013, 376, 103–110.
15 J. Cored, A. Garcia-Ortiz, S. Iborra, M. J. Climent, L. Liu, 39 D. Blaumeiser, R. Stepic, P. Wolf, C. R. Wick, M. Haumann,
C. H. Chuang, T. S. Chan, C. Escudero, P. Concepcion and
P. Wasserscheid, D. M. Smith, A.-S. Smith, T. Bauer and
A. Corma, J. Am. Chem. Soc., 2019, 141, 19304–19311.
J. Libuda, Catal. Sci. Technol., 2020, 10, 252–262.
16 J. Liu, C. Li, F. Wang, S. He, H. Chen, Y. Zhao, M. Wei, 40 S. Werner, N. Szesni, A. Bittermann, M. J. Schneider,
D. G. Evans and X. Duan, Catal. Sci. Technol., 2013, 3,
2627–2633.
P. Haerter, M. Haumann and P. Wasserscheid, Appl. Catal.,
A, 2010, 377, 70–75.
17 L. Roldan, Y. Marco and E. Garcia-Bordeje, ChemSusChem, 41 P. Wolf, M. Aubermann, M. Wolf, T. Bauer, D. Blaumeiser,
2017, 10, 1139–1144.
R. Stepic, C. R. Wick, D. M. Smith, A.-S. Smith,
P. Wasserscheid, J. Libuda and M. Haumann, Green Chem.,
2019, 21, 5008–5018.
18 J. Polanski, T. Siudyga, P. Bartczak, M. Kapkowski,
W. Ambrozkiewicz, A. Nobis, R. Sitko, J. Klimontko,
J. Szade and J. Lelatko, Appl. Catal., B, 2017, 206, 16–23.
19 Y. Zhou, N. Han and Y. Li, Acta Phys.–Chim. Sin., 2020, 36,
2001041–2001051.
42 T. Yasuda, E. Uchiage, T. Fujitani, K.-i. Tominaga and
M. Nishida, Appl. Catal., B, 2018, 232, 299–305.
43 M. J. Schneider, M. Lijewski, R. Woelfel, M. Haumann and
P. Wasserscheid, Angew. Chem., Int. Ed., 2013, 52, 6996–6999.
20 C. P. Mehnert, Chem. – Eur. J., 2005, 11, 50–56.
21 A. Riisager, R. Fehrmann, M. Haumann and 44 M. Ali, A. Gual, G. Ebeling and J. Dupont, ChemCatChem,
P. Wasserscheid, Top. Catal., 2006, 40, 91–102.
2014, 6, 2224–2228.
22 B. Xin and J. Hao, Chem. Soc. Rev., 2014, 43, 7171–7187.
23 J. Bruenig, Z. Csendes, S. Weber, N. Gorgas, R. W. Bittner,
45 M. Ali, A. Gual, G. Ebeling and J. Dupont, ChemSusChem,
2016, 9, 2129–2134.
A. Limbeck, K. Bica, H. Hoffmann and K. Kirchner, ACS 46 M. I. Bruce, M. L. Cole, R. S. Fung, C. M. Forsyth,
Catal., 2018, 8, 1048–1051.
24 R. Castro-Amoedo, Z. Csendes, J. Bruenig, M. Sauer,
M. Hilder, P. C. Junk and K. Konstas, Dalton Trans., 2008,
31, 4118–4128.
A. Foelske-Schmitz, N. Yigit, G. Rupprechter, T. Gupta, 47 H. Lebel, M. K. Janes, A. B. Charette and S. P. Nolan, J. Am.
A. M. Martins, K. Bica, H. Hoffmann and K. Kirchner,
Catal. Sci. Technol., 2018, 8, 4812–4820.
25 L.-L. Lou, X. Peng, K. Yu and S. Liu, Catal. Commun., 2008,
9, 1891–1893.
Chem. Soc., 2004, 126, 5046–5047.
48 J. A. Cabeza, I. del Rio, D. Miguel and M. G. Sanchez-Vega,
Chem. Commun., 2005, 3956–3958.
49 S. Jaaskelainen and M. Haukka, Appl. Catal., A, 2003, 247,
95–100.
26 Y. Yang, C. Deng and Y. Yuan, J. Catal., 2005, 232, 108–116.
27 X. Guo, Z. Peng, A. Traitangwong, G. Wang, H. Xu, 50 G. Lavigne, Eur. J. Inorg. Chem., 1999, 917–930.
V. Meeyoo, C. Li and S. Zhang, Green Chem., 2018, 20, 51 V. K. Srivastava and P. Eilbracht, Catal. Commun., 2009, 10,
4932–4945.
1791–1795.
28 K. L. Luska, A. Bordet, S. Tricard, I. Sinev, W. Gruenert, 52 K. Tominaga and Y. Sasaki, J. Mol. Catal. A: Chem., 2004,
B. Chaudret and W. Leitner, ACS Catal., 2016, 6, 3719–
3726.
29 K. L. Luska, P. Migowski, S. El Sayed and W. Leitner, ACS
Sustainable Chem. Eng., 2016, 4, 6186–6192.
30 L. Offner-Marko, A. Bordet, G. Moos, S. Tricard,
S. Rengshausen, B. Chaudret, K. L. Luska and W. Leitner,
Angew. Chem., Int. Ed., 2018, 57, 12721–12726.
31 M. Haumann, M. Jakuttis, R. Franke, A. Schoenweiz and
P. Wasserscheid, ChemCatChem, 2011, 3, 1822–1827.
32 M. Jakuttis, A. Schoenweiz, S. Werner, R. Franke,
220, 159–165.
53 L. Wu, Q. Liu, R. Jackstell and M. Beller, Org. Chem. Front.,
2015, 2, 771–774.
54 Y. Yao, D. S. He, Y. Lin, X. Feng, X. Wang, P. Yin, X. Hong,
G. Zhou, Y. Wu and Y. Li, Angew. Chem., Int. Ed., 2016, 55,
5501–5505.
55 B. M. Faroldi, J. F. Múnera and L. M. Cornaglia, Appl.
Catal., B, 2014, 150–151, 126–137.
56 M. P. Singh, R. K. Singh and S. Chandra, J. Phys. D: Appl.
Phys., 2010, 43, 092001–092004.
K.-D. Wiese, M. Haumann and P. Wasserscheid, Angew. 57 T. Huang, B. Liu, Z. Zhang, Y. Zhang and J. Li, RSC Adv.,
Chem., Int. Ed., 2011, 50, 4492–4495. 2014, 4, 28529–28536.
33 S. Shylesh, D. Hanna, S. Werner and A. T. Bell, ACS Catal., 58 S. Werner, N. Szesni, R. W. Fischer, M. Haumann and
2012, 2, 487–493.
P. Wasserscheid, Phys. Chem. Chem. Phys., 2009, 11, 10817–
34 S. Walter, H. Spohr, R. Franke, W. Hieringer,
10819.
P. Wasserscheid and M. Haumann, ACS Catal., 2017, 7, 59 T. G. Ostapowicz, M. Schmitz, M. Krystof, J. Klankermayer
1035–1044.
and W. Leitner, Angew. Chem., Int. Ed., 2013, 52, 12119–
12123.
60 Z. He, Z. Hou, Y. Luo, Y. Dilixiati and W. Eli, Catal. Sci.
Technol., 2014, 4, 1092–1103.
61 J. Brünig, Z. Csendes, S. Weber, N. Gorgas, R. W. Bittner,
A. Limbeck, K. Bica, H. Hoffmann and K. Kirchner, ACS
Catal., 2018, 8, 1048–1051.
35 S. El Sayed, A. Bordet, C. Weidenthaler, W. Hetaba,
K. L. Luska and W. Leitner, ACS Catal., 2020, 10, 2124–2130.
36 D. Geier, P. Schrnitz, J. Walkowiak, W. Leitner and
G. Francio, ACS Catal., 2018, 8, 3297–3303.
37 G. Moos, M. Emondts, A. Bordet and W. Leitner, Angew.
Chem., 2020, 59, 11977–11983.
3080 | Green Chem., 2021, 23, 3073–3080
This journal is © The Royal Society of Chemistry 2021