Polyamine-functionalized imidazolyl poly(ionic liquid)s for the efficient conversion of CO₂ into cyclic carbonates

Abstract

Global warming is becoming a challenging issue due to the emission of a large number of greenhouse gases, mainly CO₂. The transformation of CO₂ into chemicals with high additional value is considered a promising and sustainable way to solve the “greenhouse effect”. Herein, a series of polyamine-functionalized imidazolyl poly(ionic liquid)s (PILs) modified by triethylenetetramine (TETA) were synthesized as heterogeneous catalysts to convert CO₂ into cyclic carbonates. The synergistic effect of nucleophile (bromide anions) and polyamine groups in promoting CO₂ conversion was explored by density functional theory (DFT) calculations, which is critical to improve the catalytic performance. When the anions of ionic liquids acted as nucleophiles to attack the epoxide from the C–O bond with less steric hindrance, the substrate epoxide and CO₂ can be activated by hydrogen bonding with amine group protons. Therefore, PILs modified by TETA (N₄-PIL-x) have been verified to possess high efficiency and stable reusability for the cycloaddition of epoxides with CO₂ without solvent, metal, and co-catalyst, of which N₄-PIL-2 can achieve 98.0% conversion of epichlorohydrin (ECH) with a turnover frequency (TOF) value as high as 42.4 h⁻¹ under ambient pressure; moreover, the complete conversion of epichlorohydrin is obtained in only 4 h at 1.0 MPa CO₂ pressure.

---

Introduction

The extensive consumption of fossil fuels and various activities of human beings are the roots of excessive emission of CO₂, which has remarkably increased the aggravation of environmental issues as a primary greenhouse gas in the atmosphere.^1,2 Nevertheless, in industrial processes, CO₂ has also been recommended as a nontoxic, omnipresent, and sustainable C1 resource. Widespread attention has been paid for the efficient conversion of excessive CO₂ emissions into fine chemicals with high additional value.^3–7 The transformation of CO₂ into cyclic carbonates is one of the most effective, economical and environmentally friendly applications of fixed CO₂.^8–10 It is worth noting that cyclic carbonates are also recognized as excellent chemicals and are widely used in the industrial field.

However, due to the intrinsic thermodynamic stability and kinetic inertness of CO₂, high temperature or high pressure are often required to overcome the energy barriers in the reaction pathways of converting CO₂ into cyclic carbonates.^2,11 Therefore, it still a challenge to find an excellent catalyst for the cycloaddition of epoxides with CO₂ under milder conditions. A variety of catalysts have been extensively studied. As a typical homogeneous catalyst, ionic liquids (ILs) exhibit excellent catalytic performance,^12,13 while it is usually necessary to add alkali metals as co-catalysts.^14,15 This is similar to other metal-bearing catalysts (metal–organic frameworks (MOFs), metal complexes), which may cause environmental pollution by toxic metals.^16–19 In addition, it is still a great challenge to separate the product from the catalyst in a homogeneous catalytic system.^20–22 As one of the representative achievements of heterogeneous catalysts, porous organic polymers (POPs) can be separated from the catalytic products more easily to some extent; however, they also have disadvantages, such as complicated preparation, limited effective active sites, and harsh catalytic conditions.^23–27 Therefore, it is an active area of research to narrow the divide between homogeneous and heterogeneous catalysis by the feasible construction of heterogeneous catalysts with simple preparation, high active site density and mild catalytic conditions.²⁸

In this work, by adjusting the ratio of imidazolyl ionic liquid to divinylbenzene, PILs with different densities of catalytic active sites were obtained. After post-synthetic modification, TETA was introduced into the PILs by nucleophilic substitution reaction. Imidazole-based PILs have been considered as excellent catalysts for conversion of CO₂, in which imidazolium cations can activate CO₂ and the customized counterions can be nucleophilic to promote the ring opening of epoxides.^29–32 Furthermore, there is a unique cross-linking structure present in PILs; the imidazole structure cell can be incorporated into the backbone chain, which provides the possibility of obtaining a catalyst with a high density of active sites by regulating and controlling the catalytic active centers. As deduced from previous research, the rate-limiting step of the cycloaddition reaction is usually regarded as ring-opening by a nucleophilic attack.^33,34 It is an advanced strategy to introduce task-specific active sites to activate epoxides to make them more vulnerable to nucleophilic attack.^35–39

Various task-specific groups (–OH, –NH₂, –COOH)^39–47 have been introduced into imidazole-based PILs as hydrogen-bond donors to activate CO₂/epoxides. The hydrogen-bond donors on cations can promote the CO₂/epoxide coupling when the charge-balanced bromide anions act as nucleophiles to attack the epoxides.^40,48–50 In this paper, a series of N₄-PIL-x materials were synthesized as heterogeneous catalysts for the cycloaddition reaction. The more robust hydrogen bonds can be cooperatively produced in N₄-PIL-x with multiple neighbouring hydrogen bond donors compared to isolated hydrogen-bond donors.⁵¹ TETA, as a typical small molecular organic amine, is widely used in the field of carbon dioxide adsorption because of its small molecular size, high boiling point and low viscosity.⁵² Moreover, the primary and secondary amine groups contained in TETA are considered to be excellent multiple adjacent hydrogen bond donors. The polyamine groups produce suitable basicity with a superior affinity for CO₂ to create a high-concentration CO₂ environment at the catalytic active site, which is conducive to accelerating the catalytic reaction process. Systematic tests for the cycloaddition reaction were investigated at atmospheric pressure without solvent, metal, and co-catalyst. Based on DFT calculations, the synergistic effect of the amine groups and bromide anions in N₄-PIL-x in promoting CO₂ conversion was explored.

Experimental

General

Synthesis of ionic liquid ([BVIM]Br). The ionic liquid (IL) [3-bromopropyl-1-vinylimidazolium]Br ([BVIM]Br) was synthesized on the basis of an improved version of a previous method.³²N-Vinylimidazole (0.941 g, 0.01 mol), 1,3-dibromopropane (8.08 g, 0.04 mol) and acetonitrile (10 mL) were mixed in a 50 mL flask. Under nitrogen atmosphere, the reactants were stirred at 80 °C for 12 h. The reaction solvent was evaporated in a rotary evaporator under reduced pressure. After ethyl acetate was added, the [BVIM]Br precipitated and was obtained by filtration and washing with ethyl acetate. ¹H NMR (400 MHz, D₂O, TMS) δ (ppm) = 9.06 (s, 1H), 7.73 (d, 1H), 7.57 (d, 1H), 7.07 (dd, 1H), 5.73 (d, 1H), 5.35 (t, 1H), 4.36 (t, 2H), 3.39 (t, 2H), 2.40–2.34 (m, 2H).

Synthesis of PIL-x (x = 0, 1/2, 1, 2). PIL-x (x = 0, 1/2, 1, 2) was synthesized by copolymerization of [BVIM]Br with DVB according to a modified literature procedure.⁵³ In a typical synthesis of PIL-2, [BVIM]Br (5 mmol), azobisisobutyronitrile (0.08 g 0.49 mmol) and DVB (2.5 mmol) were added to a mixed solution of AcOEt/EtOH/H₂O (10/30/5). The mixture was reacted at 80 °C for 24 h in nitrogen atmosphere. The PIL-x precipitate was obtained by filtration and washed with ethanol. The white powder was collected after drying at 40 °C for 24 h in a vacuum. The samples of PIL-1/2 and PIL-1 were prepared via adjusting the molar ratio of [BVIM]Br (5 mmol) and DVB (10 mmol) for PIL-1/2, [BVIM]Br (5 mmol) and DVB (5 mmol) for PIL-1. The structural formulas are shown in Scheme 1. To reveal the role of imidazole-based ionic liquid unit in the copolymer catalyst on the cycloaddition reaction, PIL-0 was synthesized without integrating the IL monomer for comparison.

Scheme 1 The synthesis of N₄-PIL-x (x = 1/2, 1, 2).

Synthesis of N₄-PIL-x (x = 1/2, 1, 2). N₄-PIL-x (x = 1/2, 1, 2) was synthesized by introducing TETA into the corresponding PIL-x (x = 0, 1/2, 1, 2) by a nucleophilic substitution reaction. In a typical procedure, PIL-x (500 mg) was ground and evenly dispersed in 25 mL TETA. After 20 minutes of ultrasonic dispersion, the mixture was reacted at 120 °C for 10 h. The N₄-PIL-x was obtained by filtration at high temperature and washed with ethanol until the unreacted TETA was removed. The pale yellow powder was dried at 80 °C for 24 h in a vacuum.

General procedure for the cycloaddition of CO₂ with epoxides. An autoclave reactor (50 mL) equipped with a magnetic stirrer was applied for the cycloaddition of epoxides and CO₂. In a general procedure, the respective epoxide and catalyst N₄-PIL-x were charged into the reactor and purged with CO₂ to drive air away three times. Then, CO₂ was slowly filled into the reactor under the desired pressure. When the reactor was heated to the required temperature, the reaction started with stirring at 400 rpm. When the predetermined reaction time was over, excess CO₂ was released after cooling the reactor in an ice bath. The product was obtained after removal of the catalyst by centrifugation. The conversion was identified by ¹H NMR spectroscopy using mesitylene as an internal standard.⁵⁴ The cycloaddition reaction at atmospheric pressure was performed in a 25 mL three-neck flask with a reflux condenser under CO₂ atmosphere.

Recovery and reuse of the catalyst. After reaction, the catalyst can be recovered by simple centrifugation. After washing with ethanol and chloroform, the recovered catalyst was dried in a vacuum at 80 °C for 24 h and reused in successive reactions.

Results and discussion

Synthesis and characterization

Scheme 1 shows the strategy for the synthesis of N₄-PIL-x (x = 1/2, 1, 2). PIL-x (x = 0, 1/2, 1, 2, x represents the molar ratio of [BVIM]Br to DVB) were synthesized by radical polymerization of [3-bromopropyl-1-vinylimidazolium]Br ([BVIM]Br) and divinylbenzene (DVB). N₄-PIL-x (x = 1/2, 1, 2) was synthesized by post-synthesis modification of PIL-x (x = 1/2, 1, 2) with TETA through a nucleophilic substitution reaction.

The as-synthesised PIL-2 and N₄-PIL-x (x = 1/2, 1, 2) were characterized by FT-IR spectroscopy (Fig. 1A). All the samples exhibit the characteristic peaks at 3134 and 2954 cm⁻¹, which can correspond to the C–H stretching vibrations.³¹ The imidazolium rings are determined by the peaks at 1587 and 1441 cm⁻¹, which are attributed to C=N and C=C, respectively.^23,55 The peak at 1171 cm⁻¹ can be attributed to the stretching vibration of C–N in the amino-alkyl chain, indicating that the PILs contain the imidazolium-IL moiety.^56,57 The band at 640 cm⁻¹ is for the C–Br stretching vibration, which vanishes in the spectra of polyamine-functionalized PILs with TETA (N₄-PIL-1/2, N₄-PIL-1, N₄-PIL-2). Meanwhile, a new peak appears at 825 cm⁻¹, which is attributed to the primary amine N–H bending vibration. Thus, the successful synthesis of these polyamine-functionalized PILs was confirmed. The N₄-PIL-x backbone is further indicated by the solid-state ¹³C-nuclear magnetic resonance (¹³C-NMR) spectrum in Fig. 1B. Obvious overlapping bands around 125 ppm are displayed in the ¹³C-NMR spectrum, which can be correlated to unsubmitted benzene carbons and the C3 and C4 atoms of the imidazolium ring. Moreover, the C2 carbon of the imidazolium ring is reflected by the resonance at 143 ppm. Meanwhile, the adjacent peak with a chemical shift of ca. 136 ppm is attributed to the methylene linker carbons produced by DVB polymerization.³⁶ Furthermore, the signals of carbons linked to the N atom of the imidazole ring and amino group are found at about 58 and 47 ppm, respectively, while the peak at 40 ppm is related to the methylene tethered on the phenyl ring. The peaks at around 28–31 ppm correspond to the alkyl methylene.

Fig. 1 (A) FT-IR spectra of the samples; (B) ¹³C MAS NMR spectrum of N₄-PIL-2.

The porous characteristics of the samples were studied by N₂ adsorption/desorption (Fig. 2A). The sorption isotherms of all the samples exhibit characteristic type IV isotherms with distinct hysteresis, suggesting that the as-synthesised materials have meso-porous structures.⁵⁸ In Fig. 2B, the pore size distribution curves also confirm that all samples exhibit meso-porous structures, with the most probable apertures between 3.34 and 3.83 nm. The calculated structural parameters, including the surface area, total pore volume, and average pore size, from the nitrogen adsorption–desorption isotherms are shown in Table 1. It is apparent that the modification of TETA results in a decrease of the specific surface area and total pore volume. Compared with PIL-2, the specific surface area and total pore volume of N₄-PIL-2 decreased significantly, from 484.340 to 22.939 m² g⁻¹ and 0.315 to 0.091 cm³ g⁻¹, respectively. The results showed that catalysts with different porosities were formed by copolymerization with DVB in different proportions, which also accelerated the diffusion of the substrate molecules and was beneficial to the reaction kinetics.^59,60

Fig. 2 N₂ sorption isotherms (A) and the corresponding pore size distribution curves (B) of N₄-PIL-x polymerized at 80 °C for 24 h, using AcOEt/EtOH/H₂O (10/30/5) as the solvent.

Table 1 Synthetic conditions and textural properties of the N₄-PIL-x samples

Entry	Catalyst	IL : DVB^a	S BET [m^2 g^−1]	V P [cm^3 g^−1]	D pr [nm]
a Molar ratio of IL to DVB in the gel. b BET surface area. c Total pore volume. d Most probable aperture.
1	PIL-2	2 : 1	484.340	0.315	3.72
2	N4-PIL-1/2	1 : 2	122.481	0.156	3.71
3	N4-PIL-1	1 : 1	102.136	0.192	3.83
4	N4-PIL-2	2 : 1	22.939	0.091	3.34

---

Fig. S1A† shows the X-ray diffraction (XRD) pattern of N₄-PIL-2, indicating that N₄-PIL-2 material has the amorphous structure of a polymer network. Thermogravimetric analysis (TGA) was performed to investigate the thermal stability of N₄-PIL-2 to ensure the safe temperature range for the subsequent catalysis process of CO₂ and epoxides. As shown in Fig. S1B,† the degradation process of N₄-PIL-2 can be split into three steps. The initial step can be attributed to the evaporation of captured moisture from air at temperature below 310 °C.⁵⁵ The second step is the decomposition of ionic liquid segments and the loading of grafted TETA in the temperature range from 310 °C to 400 °C, which will prove that the synthesized N₄-PIL-2 contains abundant catalytic activity. During the temperature from 400 °C to 500 °C, the final step is related to the decomposition of the polymer matrix.^61,62 Thus, the synthesized N₄-PIL-2 is thermally stable below 310 °C, under which the catalytic reaction of CO₂ with epichlorohydrin can be performed. The elemental analysis for the N₄-PIL-x can be seen in Table 2.

Table 2 Elemental analysis of N₄-PIL-x with varied molar ratios of VIMBr to DVB

Entry	Catalyst	IL content (mmol g^−1)	Found
			C (%)	N (%)	H (%)
1	N4-PIL-1/2	0.422	80.095	3.550	7.183
2	N4-PIL-1	0.801	66.525	6.730	7.137
3	PIL-2	1.228	58.650	3.440	6.525
4	N4-PIL-2	1.222	48.880	10.270	6.615

---

The morphologies of PIL-2 and N₄-PIL-2 were investigated by scanning electron microscopy (SEM) (Fig. 3). Obviously, both polymers present fluffy cotton-like clusters at the micron scale in Fig. 3A and C, in which PIL-2 is composed of many nano-scale primary spherical particles that are cross-linked with each other and has a loose and porous structure (Fig. 3B). In contrast, due to the graft introduction of TETA in N₄-PIL-2, the radius of the primary spherical particles forming the polymer cluster gradually increases, and finally the structure formed by the agglomeration of many irregular nano-blocks is obtained (Fig. 3D). Meanwhile, the graft introduction of TETA also leads to the blockage of a large of the pores in the polymer, thereby resulting in a significant reduction in the specific surface area, which corresponds to the results shown in the N₂ sorption isotherm curves in Fig. 2. Elemental distribution of PIL-2 and N₄-PIL-2 was observed by the energy-dispersive X-ray spectroscopy (EDX) mapping. Elemental mapping in Fig. 3E and F has shown that N and Br elements are uniformly distributed in the polymer skeleton, and it further confirms the existence of homogeneously distributed IL units, that is, the active functional sites.

Fig. 3 SEM images of samples PIL-2 (A and B) and N₄-PIL-2 (C and D) and N, Br elemental mapping images of PIL-2 (E), N₄-PIL-2 (F).

Catalyst evaluation

The cyclization of epoxides with CO₂ is one of the most economical, effective and environmentally friendly applications of CO₂.^4,6 In order to evaluate the catalytic activity of the as-synthesised catalysts, the cyclization of epichlorohydrin with CO₂ was conducted without solvents and co-catalysts. The results are listed in Table 3. The conversion of PIL-0 obtained by self-polymerization of DVB was only 4.5% (Table 3, entry 1).

Table 3 Cycloaddition of CO₂ with ECH^a

Entry	Catalyst	P (MPa)	t (h)	Con.^b (%)	Sel.^b (%)	TOF^c (h^−1)
a Reaction conditions: epichlorohydrin (12.7 mmol); catalyst (10 mg); 100 °C. b Conversion (con.) and selectivity (sel.) were determined by ^1H NMR. c Turnover frequency (TOF) = [mmol (product)]/[mmol (ionic sites) × (reaction time)].
1	PIL-0	1.0	4	4.5	>99	—
2	PIL-2	1.0	4	81.2	>99	209.9
3	N4-PIL-2	1.0	4	99.6	>99	258.8
4	N4-PIL-1	1.0	4	72.3	>99	286.6
5	N4-PIL-1/2	1.0	4	42.5	>99	319.8
6	N4-PIL-2	1.0	1	81.3	>99	844.9
7	N4-PIL-2	0.1	24	98.0	>99	42.4

---

When PIL-2 containing imidazole-based ionic liquid units was chosen as a catalyst, the conversion could reach 81.2% (Table 3, entry 2), which was attributed to the good catalytic activity of the imidazole-based ionic liquid units in PIL-2 for the cyclization of CO₂ and epichlorohydrin.³² After modification with TETA, the specific surface area and total pore volume of the obtained N₄-PIL-2 decreased significantly. However, the introduction of TETA significantly improved the density of the functional groups in N₄-PIL-2, and the ring-opening reaction of epoxide substrates (rate-limiting step) is promoted by the formation of hydrogen bonds between the multiple hydrogen bond donors and epoxides. Therefore, the obtained N₄-PIL-2 exhibited the highest conversion of 99.6% (selectivity >99%) (Table 3, entry 3). Its catalytic activity was greatly improved compared with that of PIL-2 without TETA modification. Even when the reaction time was curtailed to 1.0 h, the N₄-PIL-2 catalyst could still achieve 81.3% conversion, with an initial TOF value of 844.9 h⁻¹ (Table 3, entry 6). Conversion of 98.0% was identified under ambient pressure when the reaction time was only 24 h, and the TOF value was calculated to be as high as 42.4 h⁻¹ (Table 3, entry 7). The activity of N₄-PIL-2 was superior to that of reported ionic liquid-supported catalysts (Table S1†) because of the activation of CO₂/epoxides by the polyamine groups in N₄-PIL-2 through the formation of hydrogen bonds. With the decrease of the proportion of ionic liquid units in N₄-PIL-x, the catalytic activity of N₄-PIL-x gradually decreased. The conversions were 99.6%, 72.3% and 42.5% (Table 3 entries 3–5) when N₄-PIL-2, N₄-PIL-1 and N₄-PIL-1/2 were used as catalysts, respectively.

To select the optimal reaction conditions, N₄-PIL-2 was chosen to catalyze the cycloaddition of epichlorohydrin with CO₂ in the range from 40 °C to 120 °C. The effects of temperature on the conversion are shown in Fig. 4A. At low temperatures below 60 °C, the N₄-PIL-2 was less reactive. However, when the temperature reached 100 °C, conversion of up to 99.6% was obtained. This can be interpreted by the enhancement of the reaction kinetics at higher reaction temperature, which resulted in an increased reaction conversion over the same period of time.⁶³ The effects of the catalyst amount were studied at 100 °C and 1.0 MPa (Fig. 4B). The conversion increased with the increase of the N₄-PIL-2 dosage. When the catalyst amount was 2, 5 and 8 mg, the conversion was 35.2%, 72.5% and 90.3%, respectively. As the amount was further increased to 10 mg, a quantitative conversion (99.6%) was obtained.

Fig. 4 The cycloaddition reaction of CO₂ and ECH. (A) Effect of the reaction temperature. (B) Effect of catalyst loading. (C) Effect of CO₂ pressure. (D) The kinetic curves of N₄-PIL-2 and PIL-2. Unless otherwise stated, the reaction was performed under these conditions: ECH (12.7 mmol); catalyst (10 mg); CO₂ (1.0 MPa); 100 °C for 4 h.

The plots of conversion and selectivity versus initial CO₂ pressure are shown in Fig. 4C. With the increase of the initial CO₂ pressure from 0.1 MPa to 0.5 MPa, the reaction conversion increased significantly from 47.6% to 95.2%. When the CO₂ pressure was increased to 1.0 MPa, the reaction conversion reached 99.6%, suggesting that the initial CO₂ pressure of 1.0 MPa could provide the most appropriate concentration of CO₂ for the catalytic reaction.

Fig. 4D shows the function between conversion and reaction time in the CO₂ cycloaddition with epichlorohydrin. When the reaction catalyst was N₄-PIL-2, the conversion of epichlorohydrin was 81.3% in 1.0 h, and quantitative conversion was obtained in 4 h. The initial TOF value was as high as 844.9 h⁻¹ (Table 3, entry 6). Due to the decrease of the cyclopropane concentration, even if the reaction time was increased to more than 4 h, the corresponding increase of conversion was less significant.⁵⁵ In sharp contrast, when PIL-2 was used as the catalyst to perform the reaction under the same conditions, its activity was lower than that of N₄-PIL-2; 49.5% conversion was identified in the first 1.0 h, and the initial TOF was 511.9 h⁻¹, which was only 3/5 that of N₄-PIL-2. The superior catalytic activity of N₄-PIL-2 revealed that the polyamine groups in the polymeric networks can accelerate the cyclization of CO₂. Based on the above experimental results, the optimized reaction conditions could be set to 1.0 MPa of CO₂ pressure at 100 °C for 4.0 h. It is worth mentioning that the selectivity was maintained at more than 99%, which is unrelated to the reaction temperature, catalyst dosage, CO₂ pressure and reaction time.

In addition, the recyclability of the catalyst is vitally important when developing a new heterogeneous catalyst. After washing with ethanol and chloroform and drying in a vacuum at 80 °C for 24 h, the recovered N₄-PIL-2 catalyst was reused in successive reactions. After six-run recycling tests, the conversion could still reach more than 95% and the selectivity remained above 99% (Fig. 5A). The recovered catalyst after six runs was characterized by FT-IR analysis (Fig. 5B). The FT-IR spectrum was similar to that of the original catalyst. Therefore, the N₄-PIL-2 catalyst showed excellent stability and good recyclability.

Fig. 5 Catalytic reusability of N₄-PIL-2 for cycloaddition of CO₂ with ECH. (A) Results of the six-run recycling test. (B) Comparison of the FT-IR spectra of the fresh and recovered N₄-PIL-2 catalyst. Reaction conditions: ECH (12.7 mmol); catalyst (10 mg); CO₂ (1.0 MPa); 100 °C for 4 h.

In the process of popularization of the catalyst, the universality of N₄-PIL-2 for various epoxide substrates was explored under the optimal catalytic conditions discussed above (Table 4). N₄-PIL-2 exhibited excellent reactivity on terminal epoxides; the corresponding cyclic carbonates were obtained with above 97.0% conversion and >99% selectivity. As an internal epoxide, cyclohexene oxide has an obvious steric hindrance effect, which makes it a very challenging substrate for this reaction.⁶⁴ However, when cyclohexene oxide was used as the substrate, N₄-PIL-2 could still provide a conversion of up to 62.5% within 12 h (Table 4, entry 8). Duo to the strong inductive effect of the electron-withdrawing –CH₂X (X = Cl, Br) groups, quantitative conversion (>99%) of epichlorohydrin and 2-(bromomethyl)oxirane was achieved in only 4 h (Table 4, entries 1, 3). At atmospheric pressure, epichlorohydrin even still provided 98.0% conversion (Table 4, entry 2). In contrast, these aliphatic carbon-chain alkyl epoxide substrates with electron-rich groups usually show lower reactivity; however, N₄-PIL-2 could still catalyze these substrates efficiently, with conversion of 99.2% and 97.2% in 6 h and 12 h, respectively (Table 4, entries 4, 5), which is superior to that of similar reported heterogeneous catalysts.^29,65 Additionally, the tolerance of N₄-PIL-2 to long chain ether and alkene functionalities was explored, and the corresponding cyclic carbonates could be formed at high conversion in 8 h (Table 4, entries 6, 7). As noted above, our catalysts were proved to exhibit the potential of generalization.

Table 4 Cycloaddition of CO₂ with various epoxides catalyzed by N₄-PIL-2^a

Entry	Epoxide	Product	t (h)	Con.^c (%)
a Reaction conditions: epoxides (12.7 mmol); catalyst (10 mg); CO2 (1.0 MPa); 100 °C. b CO2 (0.1 MPa). c Conversion (con.) and selectivity (>99%) were determined by ^1H NMR.
1			4	99.6
2^b			24	98.0
3			4	99.5
4			6	99.2
5			12	97.2
6			8	98.3
7			8	97.0
8			12	62.5

---

In order to demonstrate the mechanism of the high efficiency for converting CO₂ into cyclic carbonates after introducing polyamine groups, the interactions of N₄-PIL-2, epichlorohydrin and CO₂ were simulated by DFT studies. As shown in Fig. 6, the A structure molecule served as a reasonable model catalyst because it had the same molecular features as the ionic liquid structure cell (the active sites) in the framework of N₄-PIL-2. In the optimized structure B, two -NH groups of the polyamine groups combined with the oxygen atom of epichlorohydrin to form hydrogen bonds. Due to the effect of the hydrogen bonds, the bond length of the C-O bonds in epichlorohydrin were extended from 1.437 Å and 1.436 Å to 1.441 Å and 1.439 Å, respectively, which facilitated the ring-opening upon attack by bromide anions. Meanwhile, there were another two hydrogen bonds established between CO₂ and the polyamine groups, which transformed the bond length of the C=O bonds of CO₂ from 1.175 Å to 1.178 Å and 1.172 Å in the optimized structure C. Moreover, it was observed that the bond angle of CO₂ changed from a linear molecule to 178.50°; the bend of 1.50° was higher than that reported for similar heterogeneous catalysts.^66,67 This confirmed that the introduction of polyamine groups can activate CO₂ to some extent. The interaction energy between CO₂ and the A structure molecule was as high as 26.16 kJ mol⁻¹, which indicated the existence of a superior affinity between the polyamine groups and CO₂. Combined with the above results, it was proved that the polyamine groups introduced by the post-modification not only activated the pre-inserted CO₂ effectively, but also promoted the conversion of CO₂ efficiently through the synergistic effect in the rate-limiting step of the reaction.

Fig. 6 The calculated interactions of the model molecule (A) with ECH (B) and CO₂ (C).

Based on the DFT calculations and previous relevant reports,^35,38 a possible N₄-PIL-2 catalyzed cycloaddition mechanism is proposed in Scheme 2. The epoxidation reaction follows two different pathways for activating CO₂/epoxides. As the mechanism in Scheme 2A shows, the substrate epoxide is adsorbed on the face of the catalyst and activated by coordination of its oxygen atom with two –NH of the polyamine group through double hydrogen bonds. Meanwhile, the bromide anion acts as a nucleophile and attacks the weakened C–O bond from the less sterically hindered side to promote the ring-opening reaction. Afterward, the polyamine-functionalized catalyst creates a high-concentration CO₂ environment at the catalytic active site and activates the pre-inserted CO₂ effectively by hydrogen bonds. The alkyl carbonate species are obtained by grafting CO₂ into the C–O covalent bond, which forms the ring-opened oxyanion intermediates. Finally, the corresponding cyclic carbonate is produced with closing of the intramolecular ring and regeneration of the catalyst. In the meantime, the mechanism in Scheme 2B was obtained from previous related reports, and it relies on the operation of imidazolium ionic liquid units. This is why PIL-2 containing imidazole-based ionic liquid units also has high catalytic activity. The synergistic effect of these two activation modes resulted in the superior catalytic activity of N₄-PIL-2.

Scheme 2 Two possible synergistic mechanisms (A and B) for the cycloaddition reaction of CO₂ and epoxide catalyzed by N₄-PIL-2.

Conclusions

Based on the copolymerization of imidazolium-based bromide salts and DVB in different molar ratios, a series of novel stable porous polyamine-functionalized heterogeneous catalysts, N₄-PIL-x, were successfully synthesized by creatively introducing task-specific polyamine groups through a post-modification method. The chemical composition and structure of copolymers were characterized. Their catalytic performance for cycloaddition of epoxides with CO₂ to cyclic carbonates was evaluated. Benefiting from abundant catalytic active sites and task-specific polyamine components, the N₄-PIL-2 has high efficiency for cycloaddition of epoxides with CO₂ under mild conditions. N₄-PIL-2 can be easily recovered by simple centrifugation and reused six times with little loss of activity. For the conversion of CO₂ and epichlorohydrin, the catalytic activity of N₄-PIL-2 is 1.7 times that of the counterpart PIL-2 without polyamine groups. Under ambient pressure, the conversion can even reach 98.0%, with a high TOF value of 42.4 h⁻¹. DFT simulations have elucidated that the synergistic effect of polyamine amine groups and bromide anions in N₄-PIL-x is critical in promoting CO₂ conversion. Polyamine groups can efficiently activate both substrate epoxide and CO₂ through hydrogen bonding interactions. In this work, a simple method of introducing task-specific groups to obtain the multifunctional materials was proposed, and some useful insights were put forward for the rational functionalization of ionic copolymers.

Author contributions

Yizhen Zou: conceptualization, date curation, methodology, investigation, writing-original draft. Yuansheng Ge: investigation and formal analysis. Qiang Zhang: formal analysis and validation. Wei Liu: date curation and supervision. Xiaoguang Li: formal analysis. Guoe Cheng: funding acquisition and supervision. Hanzhong Ke: funding acquisition and writing-review & editing.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported financially by the National Natural Science Foundation of China (Grant no. 21777149).

References

1. S. Chu, Science, 2009, 325, 1599.
2. W. Zhou, Q. W. Deng, G. Q. Ren, L. Sun, L. Yang, Y. M. Li, D. Zhai, Y. H. Zhou and W. Q. Deng, Nat. Commun., 2020, 11, 4481.
3. M. Jahandar Lashaki, S. Khiavi and A. Sayari, Chem. Soc. Rev., 2019, 48, 3320–3405.
4. M. D. Burkart, N. Hazari, C. L. Tway and E. L. Zeitler, ACS Catal., 2019, 9, 7937–7956.
5. B. Grignard, S. Gennen, C. Jerome, A. W. Kleij and C. Detrembleur, Chem. Soc. Rev., 2019, 48, 4466–4514.
6. Y. Chen and T. Mu, Green Chem., 2019, 21, 2544–2574.
7. K. Yu, P. Puthiaraj and W. S. Ahn, Appl. Catal., B, 2020, 273, 119059.
8. A. C. Deacy, E. Moreby, A. Phanopoulos and C. K. Williams, J. Am. Chem. Soc., 2020, 142, 19150–19160.
9. F. Zhang, S. Bulut, X. Shen, M. Dong, Y. Wang, X. Cheng, H. Liu and B. Han, Green Chem., 2021, 23, 1147–1153.
10. Y. Sang and J. Huang, Chem. Eng. J., 2020, 385, 123973.
11. I. Omae, Coord. Chem. Rev., 2012, 256, 1384–1405.
12. Y. Wu, X. Song, S. Xu, J. Zhang, Y. Zhu, L. Gao and G. Xiao, Catal. Lett., 2019, 149, 2575–2585.
13. S. Shi, Y. Wu, M. Zhang, Z. Zhang, O. Oderinde, L. Gao and G. Xiao, J. Mol. Liq., 2021, 339, 117278.
14. M. K. Leu, I. Vicente, J. A. Fernandes, I. de Pedro, J. Dupont, V. Sans, P. Licence, A. Gual and I. Cano, Appl. Catal., B, 2019, 245, 240–250.
15. J. W. Comerford, I. D. V. Ingram, M. North and X. Wu, Green Chem., 2015, 17, 1966–1987.
16. C. K. Ng, R. W. Toh, T. T. Lin, H.-K. Luo, T. S. A. Hor and J. Wu, Chem. Sci., 2019, 10, 1549–1554.
17. C. M. Miralda, E. E. Macias, M. Zhu, P. Ratnasamy and M. A. Carreon, ACS Catal., 2011, 2, 180–183.
18. P. Puthiaraj, S. Ravi, K. Yu and W.-S. Ahn, Appl. Catal., B, 2019, 251, 195–205.
19. L. He, J. K. Nath and Q. Lin, Chem. Commun., 2019, 55, 412–415.
20. J. Wang and Y. Zhang, ACS Catal., 2016, 6, 4871–4876.
21. M. Liu, L. Liang, X. Li, X. Gao and J. Sun, Green Chem., 2016, 18, 2851–2863.
22. Y. Wu, Y. Zhao, R. Li, B. Yu, Y. Chen, X. Liu, C. Wu, X. Luo and Z. Liu, ACS Catal., 2017, 7, 6251–6255.
23. G. Chen, X. Huang, Y. Zhang, M. Sun, J. Shen, R. Huang, M. Tong, Z. Long and X. Wang, Chem. Commun., 2018, 54, 12174–12177.
24. J. Li, D. Jia, Z. Guo, Y. Liu, Y. Lyu, Y. Zhou and J. Wang, Green Chem., 2017, 19, 2675–2686.
25. J. Wang, W. Sng, G. Yi and Y. Zhang, Chem. Commun., 2015, 51, 12076–12079.
26. G. Chen, Y. Zhang, K. Liu, X. Liu, L. Wu, H. Zhong, X. Dang, M. Tong and Z. Long, J. Mater. Chem. A, 2021, 9, 7556–7565.
27. S. T. Kostakoğlu, Y. Chumakov, Y. Zorlu, A. E. Sadak, S. Denizaltı, A. G. Gürek and M. M. Ayhan, Mater. Adv., 2021, 2, 3685–3694.
28. R. Luo, X. Liu, M. Chen, B. Liu and Y. Fang, ChemSusChem, 2020, 13, 3945–3966.
29. X. Wang, Q. Dong, Z. Xu, Y. Wu, D. Gao, Y. Xu, C. Ye, Y. Wen, A. Liu, Z. Long and G. Chen, Chem. Eng. J., 2021, 403, 126460.
30. J. Hu, J. Ma, H. Liu, Q. Qian, C. Xie and B. Han, Green Chem., 2018, 20, 2990–2994.
31. M. Ding and H.-L. Jiang, ACS Catal., 2018, 8, 3194–3201.
32. Q. Yi, T. Liu, X. Wang, Y. Shan, X. Li, M. Ding, L. Shi, H. Zeng and Y. Wu, Appl. Catal., B, 2021, 283, 119620.
33. H. Zhong, Y. Su, X. Chen, X. Li and R. Wang, ChemSusChem, 2017, 10, 4855–4863.
34. E. D. Bates, R. D. Mayton, I. Ntai and J. H. Davis, J. Am. Chem. Soc., 2002, 124, 926–927.
35. G. Chen, Y. Zhang, J. Xu, X. Liu, K. Liu, M. Tong and Z. Long, Chem. Eng. J., 2020, 381, 122765.
36. J. Zhang, X. Li, Z. Zhu, T. Chang, X. Fu, Y. Hao, X. Meng, B. Panchal and S. Qin, Adv. Sustainable Syst., 2020, 5, 2000133.
37. D. H. Lan, Y. X. Gong, N. Y. Tan, S. S. Wu, J. Shen, K. C. Yao, B. Yi, C. T. Au and S. F. Yin, Carbon, 2018, 127, 245–254.
38. S. Motokucho, Y. Takenouchi, R. Satoh, H. Morikawa and H. Nakatani, ACS Sustainable Chem. Eng., 2020, 8, 4337–4340.
39. M. Liu, X. Wang, Y. Jiang, J. Sun and M. Arai, Catal. Rev.: Sci. Eng., 2019, 61, 214–269.
40. Z. Guo, Q. Jiang, Y. Shi, J. Li, X. Yang, W. Hou, Y. Zhou and J. Wang, ACS Catal., 2017, 7, 6770–6780.
41. S. Wu, B. Wang, Y. Zhang, E. H. M. Elageed, H. Wu and G. Gao, J. Mol. Catal. A: Chem., 2016, 418–419, 1–8.
42. C. Yue, D. Su, X. Zhang, W. Wu and L. Xiao, Catal. Lett., 2014, 144, 1313–1321.
43. Y. Wu, Y. Xiao, H. Yuan, Z. Zhang, S. Shi, R. Wei, L. Gao and G. Xiao, Microporous Mesoporous Mater., 2021, 310, 110578.
44. C. Bao, Y. Jiang, L. Zhao, D. Li, P. Xu and J. Sun, New J. Chem., 2021, 45, 13893–13901.
45. Z. Shi, Q. Su, T. Ying, X. Tan, L. Deng, L. Dong and W. Cheng, J. CO2 Util., 2020, 39, 101162.
46. Y. Qu, J. Lan, Y. Chen and J. Sun, Sustainable Energy Fuels, 2021, 5, 2494–2503.
47. X. Zhang, W. Geng, C. Yue, W. Wu and L. Xiao, J. Environ. Chem. Eng., 2016, 4, 2565–2572.
48. Y. Zhang, G. Chen, L. Wu, K. Liu, H. Zhong, Z. Long, M. Tong, Z. Yang and S. Dai, Chem. Commun., 2020, 56, 3309–3312.
49. Y. Zhou, W. Zhang, L. Ma, Y. Zhou and J. Wang, ACS Sustainable Chem. Eng., 2019, 7, 9387–9398.
50. J. Cao, W. Shan, Q. Wang, X. Ling, G. Li, Y. Lyu, Y. Zhou and J. Wang, ACS Appl. Mater. Interfaces, 2019, 11, 6031–6041.
51. C. Martín, G. Fiorani and A. W. Kleij, ACS Catal., 2015, 5, 1353–1370.
52. F. M. Yang, Y. Liu, L. Chen, C. T. Au and S. F. Yin, J. Chem. Technol. Biotechnol., 2016, 91, 2340–2348.
53. Z. Guo, X. Cai, J. Xie, X. Wang, Y. Zhou and J. Wang, ACS Appl. Mater. Interfaces, 2016, 8, 12812–12821.
54. S. Shi, Y. Wu, M. Zhang, R. Wei, L. Gao and G. Xiao, Mol. Catal., 2021, 509, 111659.
55. W. Zhang, F. Ma, L. Ma, Y. Zhou and J. Wang, ChemSusChem, 2020, 13, 341–350.
56. Z. Z. Yang, Y. Zhao, H. Zhang, B. Yu, Z. Ma, G. Ji and Z. Liu, Chem. Commun., 2014, 50, 13910–13913.
57. W. Zhang, T. Liu, H. Wu, P. Wu and M. He, Chem. Commun., 2015, 51, 682–684.
58. R. Qu, Z. Ren, N. Li, F. Zhang, Z. J. Zhang and H. Zhang, J. CO2 Util., 2020, 38, 168–176.
59. Y. Chen, R. Luo, J. Bao, Q. Xu, J. Jiang, X. Zhou and H. Ji, J. Mater. Chem. A, 2018, 6, 9172–9182.
60. S. Ghazali-Esfahani, H. Song, E. Păunescu, F. D. Bobbink, H. Liu, Z. Fei, G. Laurenczy, M. Bagherzadeh, N. Yan and P. J. Dyson, Green Chem., 2013, 15, 1584–1589.
61. W. L. Dai, L. Chen, S. F. Yin, W. H. Li, Y. Y. Zhang, S. L. Luo and C. T. Au, Catal. Lett., 2010, 137, 74–80.
62. L. Han, H. J. Choi, D. K. Kim, S. W. Park, B. Liu and D. W. Park, J. Mol. Catal. A: Chem., 2011, 338, 58–64.
63. D. J. Darensbourg, J. C. Yarbrough, C. Ortiz and C. C. Fang, J. Am. Chem. Soc., 2003, 125, 7586–7591.
64. Y. Y. Zhang, G. W. Yang, R. Xie, L. Yang, B. Li and G. P. Wu, Angew. Chem., 2020, 59, 23291–23298.
65. H. Song, Y. Wang, Y. Liu, L. Chen, B. Feng, X. Jin, Y. Zhou, T. Huang, M. Xiao, F. Huang and H. Gai, ACS Sustainable Chem. Eng., 2021, 9, 2115–2128.
66. X. L. Meng, Y. Nie, J. Sun, W. G. Cheng, J. Q. Wang, H. Y. He and S.-J. Zhang, Green Chem., 2014, 16, 2771–2778.
67. W. Zhang, Q. Wang, H. Wu, P. Wu and M. He, Green Chem., 2014, 16, 4767–4774.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1cy01765a

---