Graphene oxide immobilized with ionic liquids: facile preparation and efficient catalysis for solvent-free cycloaddition of CO₂ to propylene carbonate

Abstract

Functionalized imidazolium-based ionic liquids (ILs) with different halides (Cl, Br, and I) were successfully immobilized on the surface of graphene oxide materials by one step through covalent condensation between alcoholic hydroxyl groups of GO and alkoxyl groups of functionalized ILs. Several characterization including TG, Raman, AFM, FT-IR, and XPS techniques have been applied to characterize the physicochemical properties of the synthesized GO-[SmIm]X materials. In the solvent-free cycloaddition reactions of CO₂ to propylene oxide, GO-[SmIm]I showed remarkably catalytic activity, affording a maximum yield of propylene carbonate as ca. 96%. The heterogeneous catalyst could be reused for at least four runs without any significant loss in activity, and demonstrated versatile catalysis for a wide range of substrates. A possible catalytic mechanism has been proposed, wherein epoxides were activated by the oxygen-containing groups of GO and the halide anions of the grafted ILs acted as key active species for the catalytic cycloaddition reactions.

---

1. Introduction

CO₂ is the primary byproduct of worldwide fossil-fuel-based energy production and is believed to be partly responsible for the global warming and climate change thereof.^1,2 Alternatively, CO₂ is also an abundant, economic, and nontoxic C1 resource.³ Therefore, the transformation of CO₂ as a renewable building block to valuable chemicals has attracted increasing attention in the past several decades.^4,5 Among the various pathways explored to transform CO₂, in terms of green chemistry and atomic economy, the cycloaddition of CO₂ with epoxides to five-membered cyclic carbonates has been regarded as the most promising process.^6,7 Such products are industrially important and widely used as polar solvents, electrolyte components in lithium ion batteries, and monomers for manufacturing polycarbonates,⁸ etc.

Up to now, a wide range of catalytic systems have been developed for the cycloaddition reactions of CO₂ to cyclic carbonates, including alkali metal halides,^7,9 quaternary ammonium and phosphonium salts,^10–12 ion-exchange resins,^13,14 metal–salen complexes,^15,16 etc. Among these candidates, ionic liquids (ILs) have been reported as the most efficient catalyst for the reactions.^17,18 Unfortunately, due to their inherently homogeneous nature, the crucial disadvantage in the use of such ILs lies in the difficulty in product separation as well as catalyst recovery.^17,19,20 To successfully circumvent this problem, mesoporous siliceous materials (e.g. SBA-15,²¹ and MCM-41 ²²) have been utilized as supports to immobilize the ILs and then served as heterogeneous catalysts for the cycloaddition of CO₂ to cyclic carbonates. Nevertheless, the high cost of mesoporous silicas and tedious preparation procedure of such supported ILs still restricted their further practical application. In this sense, it is of high interest to exploit a new material to immobilize ILs, which could be facilely prepared and more importantly, afford efficient catalytic activity towards the cycloaddition of CO₂.

Graphene oxide (GO) and its related materials have attracted tremendous attention in wide communities,^23–25 including adsorption, photocatalysis, optics, heterogeneous catalysis, and biosensors.^26,27 In particular, the high solubility, high surface area, and the fact that there are almost no barriers to mass transfer for substrates enable GO also a novel and metal-free catalyst in the fields of heterogeneous catalysis.^28,29 During the synthetic procedures of GO, a large quantity of oxygen-containing functionalities (e.g. alcoholic hydroxyl, epoxy, and carboxyl) were inevitably introduced to the graphene plane, thereby making GO an excellent catalyst for a broad range of organocatalysis, such as oxidative coupling of amines to imines,³⁰ oxidation of alcohols,³¹ C–H oxidation,³² and ring opening of epoxides.^5,33

Very recently, a series of reports^34–37 have revealed that GO could be functionalized with oxysilanes via the condensation of alcoholic hydroxyl and/or epoxy groups of GO and alkoxyl groups of oxysilanes. However, to the best of our knowledge, the application of such grafted GO material as a single catalyst for the cycloaddition of CO₂ is rarely reported. Since GO showed its potential in promoting the ring opening of epoxides, inspired by this, in the present work, we have reported one-step synthesis of IL-immobilized GO via the reaction of GO and functionalized imidazolium-based IL. The synthesized catalyst exhibited highly efficient and stable catalytic activity in the cycloaddition of CO₂ with epoxides to cyclic carbonates in absence of any solvent.

2. Experimental

2.1 Preparation of ILs

The ILs were prepared by the method reported previously.³⁸ In brief, 18.6 mL of 3-choropropyltrimethoxysilane and 8.0 mL g of N-methylimidazole were mixed well and refluxed at 80 °C for 48 h. The resultant mixture was washed for three times with ether to remove unreacted impurities. Then, the solution was dried under vacuum to remove excess ether. Finally, a viscous and light yellow IL, i.e. 1-(trimethoxysilyl)propyl-3-methylimidazolium chloride, was obtained and denoted as [SmIm]Cl. Likewise, the 1-(trimethoxysilyl)propyl-3-methylimidazolium bromide, was synthesized using 3-bromopropyltrimethoxysilane and N-methylimidazole as precursors, and denoted as [SmIm]Br. Moreover, 3 g of [SmIm]Cl was added into 100 mL of acetone solution containing KI (0.1 mol L⁻¹). The mixture was stirred for 3 h, and subjected to filtration to remove white precipitate and rotary evaporation. The resultant light gray IL prepared through an ion-exchange method was 1-(trimethoxysilyl)propyl-3-methylimidazolium iodide, labeled as [SmIm]I.

2.2 Preparation of immobilized ILs on GO

GO material was synthesized using natural graphite through a modified Hummers' method.²⁷ For the immobilized ILs on GO, 0.1 g of GO was added into 100 mL of ethanol containing 1.0 g of the above IL. Next, the mixture was ultrasonicated for three times; each ultrasonication lasted for ca. 30 min. After that, the well dispersed mixture was refluxed for 24 h, and then filtrated, rinsed by ethanol for several times, and subsequently dried overnight under 80 °C. Finally, the obtained immobilized IL was denoted as GO-[SmIm]X, where X represented the halide anion.

2.3 Sample characterization

Thermal gravimetric (TG) measurements were conducted on a Perkin-Elmer TGA 7 analyzer. The sample was placed in an α-Al₂O₃ chamber and heated in flowing N₂ (50 mL min⁻¹) from room temperature to 800 °C at a rate of 20 °C min⁻¹.

Raman spectra were recorded on a Raman spectrometer (Jobin Yvon Lab Ram HR evolution) using 532 nm line as an excitation source.

Atomic force microscopy (AFM) images were obtained using a Park Scientific CP-Research model (VEECO) with a Si tip in the tapping mode. To prepare the samples for AFM measurement, GO was diluted with ethanol (0.25 mg mL⁻¹) and the mixture was drop-casted on a silicon substrate.

Transmission electron microscopy (TEM) experiments of the materials were conducted on a JEOL 2100 electron microscope operating at 200 kV. Scanning electron microscopy (SEM) was recorded on a field emission scanning electron microscope (ZEISS SUPRA55).

Fourier transform infrared (FT-IR) spectra of the samples were collected in transmission mode from KBr pellets at room temperature on a Bruker Tensor 27 spectrometer with a resolution of 4 cm⁻¹, using 32 scans per spectrum in the region of 400–4000 cm⁻¹. The mass ratio of every sample to KBr was constant at 1 : 200.

X-ray photoelectron spectroscopy (XPS) measurements were performed using a Perkin-Elmer PHI 5000C spectrometer working in the constant analyzer energy mode with Mg K_α radiation as the excitation source. The carbonaceous C 1s line (284.6 eV) was used as the reference to calibrate the binding energies.

2.4 Catalytic test

The cycloaddition of CO₂ with PO to PC was carried out in 80 mL stainless steel autoclave equipped with a magnetic stirrer. In a typical reaction process, 15 mL of PO, and 0.6 g of the catalyst were added into the reactor. Then, the reactor was pressurized with CO₂ to a desired pressure and heated to 100 °C under stirring for 4 h. After the reaction, the autoclave was cooled in ice water and the excess CO₂ was vented. The liquid product was separated by centrifugation and analyzed using a GC equipped with a SE-54 capillary column and FID. The liquid mixture consisted of PO, PC, and 1,2-propylene glycol (PG) as the byproduct originating from the hydrolysis of PO with trace H₂O. No other substance was detected. The carbon balance was nearly 100%. Their quantitative calculation (i.e. conversion of PO, and selectivity to PC) was based on an area-normalization method. The PO conversion and selectivity to PC were calculated as follows:
where A, and f were the peak area of GC, and response factor for each product.

The filtered catalyst was washed with methanol (50 mL) for two times, dried for 2 h under vacuum, and then investigated for its next running.

3. Results and discussions

3.1 Structure characterization

Initially, the thermal stability of the pure and grafted GO materials was analyzed by TG under N₂ atmosphere and the result was shown in Fig. 1, GO is thermally unstable; it started to decompose at even ca. 120 °C. Moreover, two significant weight losses have been detected at ca. 250 and 600 °C, which were associated with the decomposition of oxygen-containing functionalities (e.g. alcoholic hydroxyl, epoxy, and carboxyl) and pyrolysis of carbon skeleton,^34,39 respectively. In comparison, the TG curve of GO-[SmIm]I presented a broad weight loss in the range of 200–350 °C, accounting for ca. 20% of the total weight loss. More importantly, the loss value collected at 800 °C was only 43%, much lower than that of the pure GO material (92%). Given in these TG results, it can be inferred that the incorporation of functionalized ILs have reacted with the original active oxygen-containing groups of GO, therefore enhancing the thermal stability of GO.

Fig. 1 TG curves of GO (a), and the fresh (b), and spent GO-[SmIm]I (c) materials.

Fig. 2 shows Raman spectra of GO and GO-[SmIm]I. Both the specimens displayed two apparent peaks at ca. 1363 and 1594 cm⁻¹, which were attributed to the D band (vibrations of sp³-hybrized carbon atoms of defects and disorder), and G band (vibrations of sp²-hybrized carbon atoms in 2D graphitic hexagonal lattices),⁴⁰ respectively. Apart from this analogy, the overall intensity of Raman spectrum from GO-[SmIm]I was appreciably lower than that of the pure GO material. Furthermore, in the case of their intensity ratios of D to G peaks (i.e. I_D/I_G), the value obtained over GO-[SmIm]I was ca. 1.25, much higher than that of GO (1.13). The variation manifested that, after the immobilization of IL, the percentage of sp² carbon species underwent a decline by the formation of disordered sp³ carbon atoms.

Fig. 2 Raman spectra of GO (a) and GO-[SmIm]I (b) materials.

AFM measurement was used to observe the graphitic sheet morphology of the materials. The AFM image of GO (Fig. S1†) revealed that its thickness was estimated at 0.5–1.5 nm, equivalent to 1–2 graphene sheets, very close to the values of exfoliated GO materials reported previously.^41,42 On the other hand, the average distance between the graphitic sheets of GO-[SmIm]I was ca. 3 nm (Fig. 3). Obviously, the expansion of the average interlayer spacing of GO-[SmIm]I resulted from the immobilization of long-chain IL on the graphitic sheets of GO. The similar phenomenon has also been reported involving GO materials grafted or incorporated by long-chain components.^35,36,39 Also, the electrostatic repulsion of the grafted IL could contribute to the change of the interlayer spacing.³⁶

Fig. 3 AFM image (left) of GO-[SmIm]I material and its height profiles (right) along the two lines marked in the image.

The TEM and SEM images of GO-[SmIm]I are shown in Fig. 4. The TEM image of GO-[SmIm]I illustrated that the IL-grafted GO sample was thin and transparent, in good agreement with the result observed in the above AFM measurement. Also, the SEM image displayed laminated but wrinkled structures. The two images were similar to those of the bare GO sample (Fig. S2†), illuminating that the introduction of functionalized IL has not damaged the original thin sheets of GO material. Furthermore, the EDX mapping based on the SEM image of GO-[SmIm]I (Fig. S3†) indicated that the carbon, oxygen, silicon, and iodine elements dispersed well on surface of the GO-[SmIm]I sample.

Fig. 4 TEM (a) and SEM (b) images of GO-[SmIm]I material.

The chemical functional groups before and after the immobilization of the ILs were recorded by FT-IR spectroscopy. The spectrum (Fig. 5) of the pristine GO showed four intensive peaks at 1720, 1612, 1228, and 1072 cm⁻¹, indexed as the stretching vibration of C=O, C=C, C–O, and C–O groups of GO,⁴³ respectively. The spectrum information agreed well with those of GO-based materials reported elsewhere.^44,45 After the immobilization of [SmIm]I, the bands about the C=O and C=C groups have been well retained, whereas the intensity of the peaks corresponding to C–OH and C–O chemical functions decreased noticeably. Additionally, a sharp peak located at 1126 cm⁻¹, and two tiny yet pronounced peaks at 2941 and 2883 cm⁻¹ emerged in the case of GO-[SmIm]I, which corresponded to imidazole motifs, and symmetric and asymmetric vibrations of CH₂ of the alkyl chains,^22,46 respectively. Compared with the spectrum of pure [SmIm]I IL (Fig. S4†), these unique peaks of the spectrum of GO-[SmIm]I were derived from the original IL. In addition, a shoulder band has been also found at 1076 cm⁻¹ in the case of GO-[SmIm]I. This was the proof of the presence of C–O–Si functions in the GO-[SmIm]I materials,³⁹ confirming that the IL have been immobilized on the GO via the condensation between C–OH groups of GO and alkoxyl groups of oxysilanes. In this study, we have also conducted a control experiment by preparing a silanized GO sample (GO-sil) using trimethylchlorosilane as a silylation agent (see ESI† for the detailed procedure). The corresponding FT-IR spectrum (Fig. S5†) revealed that the featured C=O and C=C groups remained on the GO-sil sample. However, the band pertaining to C–OH bonds disappeared, and meanwhile two notable peaks centered at 1240 and 1072 cm⁻¹ have been detected. The two bands were assigned to the Si–O–Si and C–O–Si motifs, which undoubtedly resulted from the self-condensation of trimethylchlorosilane, and condensation between trimethylchlorosilane and GO, respectively.

Fig. 5 FT-IR spectra of GO (a), and GO-[SmIm]I (b) materials.

XPS was further employed to analyze the surface information of the GO and GO-[SmIm]I materials. The XPS survey spectrum (Fig. S6†) indicated that the surface of the bare GO material was only constituted by C and O elements. For GO-[SmIm]I, besides the two elements, I, N, and Si have been detected in its XPS survey (Fig. 6A), suggesting that the [SmIm]I IL has been grafted on the surface of GO. In order to elucidate the detailed distribution of C species, deconvolution of C 1s spectra for both GO and GO-[SmIm]I was performed. As shown in Fig. 6B, the C 1s spectrum could be separated in four independent peaks. The lowest peak with a binding energy of 284.7 eV is attributed to the graphitic C (sp²) atoms. The peak centered at 285.4, 286.7, and 288.4 eV were contributed to C–OH, epoxide, C=O groups,^35,39 respectively (Table S1†). By comparison, a minor peak located at 285.9 eV appeared at the C 1s spectrum of GO-[SmIm]I (Fig. 6C), which was indexed as C–N groups,^39,47 definitely originating from the imidazole units of [SmIm]I. Moreover, the intensity of peaks associated with C–OH and epoxy C species decreased apparently. This indicates that a certain number of C–OH groups and epoxides of GO have undergone reactions during the introduction of [SmIm]I. Herein, combining the C 1s spectra and the above FT-IR results, it can be concluded that the [SmIm]I have been successfully covalently grafted on the surface of GO, as illustrated in Scheme 1.

Fig. 6 XPS survey of GO-[SmIm]I (A), and C 1s spectra of GO (B) and GO-[SmIm]I (C) materials.

Scheme 1 A possible structure of GO-[SmIm]X material.

3.2 Catalyst activity

A blank test without any catalyst was performed for the cycloaddition of CO₂ with PO (Table 1); the corresponding PO conversion was less than 3%. The pure GO material also gave a poor catalytic activity for the reaction. This meant that, under such reaction conditions applied in this study, GO could not promote the cycloaddition of CO₂ with PO. In sharp contrast, the functionalized imidazolium-based IL, i.e. [SmIm]Cl showed an outstanding catalytic PO conversion (97.5%), and the product was almost solely PC. The catalytic result was in good agreement with the previous work involving homogeneous IL-catalyzing cycloaddition of CO₂ with PO,^46,48 where the halide cations were proved as the catalytically active species. In the case of the immobilized GO-[SmIm]Cl complex, under the same catalytic conditions, the heterogeneous catalyst demonstrated a moderate PC yield of ca. 68.7%, less than that obtained over the pure [SmIm]Cl IL. Besides GO-[SmIm]Cl, we have also investigated the catalytic performances of the supported IL with different halide anions. As listed in Table 1, the GO-[SmIm]I complex gave the highest catalytic activity, followed by GO-[SmIm]Br and GO-[SmIm]I. Apparently, the order of their activity depended on the anions as the halide with higher leaving ability showed higher catalytic activity (discussed below). Additionally, the GO-sil sample immobilized with [SmIm]Cl, namely GO-sil-[SmIm]Cl, was also submitted to the catalytic examination, and the resultant catalytic activity was very poor. As discussed above, it is evident that silylation sacrificed the crucial C–OH groups on the surface of GO, and thereafter the catalytically active IL component could not be grafted on GO.

Table 1 Catalytic activity of GO and GO-[SmIm]X materials in cycloaddition reactions of CO₂^a

Catalyst	Conv. (%)	Sel. (%)	Yield (%)
a Reaction conditions: VPO = 15 mL, pCO2 = 2.0 MPa, T = 140 °C, t = 4 h, and Wcatal. = 0.6 g.
—	<3	—	—
GO	<5	—	—
[SmIm]Cl	97.5	>99.5	97.0
GO-[SmIm]Cl	68.7	>99.5	68.3
GO-[SmIm]Br	85.0	>99.5	84.6
GO-[SmIm]I	96.4	>99.5	96.0
GO-sil-[SmIm]Cl	<5	—	—

---

The influence of reaction conditions on the catalytic performances was further examined. At the first 1 h, the PO conversion achieved over GO-[SmIm]I was as high as ca. 62% (Fig. 7A), suggesting that with the aid of the immobilized IL catalyst, the reaction proceeded fast in a short time. Upon prolonging the catalytic test, the PO conversion increased progressively while leveled off after 4 h. Likewise, the reaction temperature was found to be sensitive to the catalytic performances of the cycloaddition of CO₂ with PO (Fig. 7B). Under a low temperature of 80 °C, the PO conversion was only ca. 21%. As the temperature was elevated, the conversion increased drastically and reached its maximum at 140 °C, affording a PO conversion of ca. 96%. However, excessive temperature would bring out a negative impact on the reaction, which was plausibly due to fractional polymerization of PC.⁴⁹ Fig. 7C displays the effect of the catalyst amount on catalytic activity. Using more catalysts could introduce more catalytically active species, and therefore facilitate the catalytic PO conversion. When the mass of the GO-[SmIm]I catalyst was up to 0.6 g, the PO conversion increased smoothly. Considering the above effects of various reaction conditions, a reaction temperature of 140 °C at 5 h and 0.6 g of the catalyst were chosen as optimal catalytic conditions for achieving the highest PC yield.

Fig. 7 Effects of reaction time (A), temperature (B), and catalyst amount (C) on the catalytic activity of GO-[SmIm]I in cycloaddition reactions of CO₂. In all cases, the selectivities to PC were all above 99.5%.

In addition to the reaction conditions, recyclability and reproducibility were also key issues to evaluate a heterogeneous catalyst. Regarding this point, a series of repetitious experiments were carried out over the three GO-immobilized IL catalysts. After each run, the mass of the rinsed and dried catalyst was found ca. 5% lower than that of previous one, which should be due to the partial loss during the rising procedure, however confirming that the immobilized IL was chemically stable on the GO support. During the four consecutive runs, all the three different GO-immobilized catalysts revealed no obvious decline in terms of their catalytic activities (Fig. 8). The fourth PO conversion acquired over GO-[SmIm]I, for instance, was still as high as 92%. According to the TG analysis, the spent GO-[SmIm]I catalyst subjected to four runs exhibited the same TG curve to the fresh one (Fig. 1). The test clearly excluded the potential leaching problem of the active species for the present heterogeneous systems, and the excellent recyclability and reproducibility was undoubtedly owing to the firm connection between the GO support and the functionalized ILs. Also, in order to further verify the chemical stability of the catalyst, FT-IR and XPS measurements were further employed to characterized the used GO-[SmIm]I catalyst, and the corresponding spectra were provided as Fig. S7 and S8,† respectively. For the GO-[SmIm]I catalyst subjected to one, two, and three runs, their spectra exhibited the same bands (including vibrations of CH₂ groups and characteristic signals of C=O, C=C, imidazole, and C–O–Si functions) to the fresh GO-[SmIm]I sample (Fig. 5), and no additional bands have been detected, suggesting that the grafted IL component has not undergone any significant loss during the several catalytic examination. In the case of XPS analysis, the survey scan (Fig. S8A†) manifested the presence of C, N, O, Si, and I elements. Moreover, the deconvolution of C 1s spectrum (Fig. S8B†) of the spent GO-[SmIm]I sample revealed similar peak distribution to that of the fresh catalyst. Combining the above characterization of TG, FT-IR, and XPS, the chemical composition and functionalities of GO-[SmIm]I have remained during the consecutive catalytic tests, which is directly responsible for its excellent catalytic reproductivity.

Fig. 8 Recycling tests for GO-[SmIm]X (a: X = Cl; b: X = Br; and c: X = I) catalysts in the cycloaddition reactions of CO₂. Reaction conditions: V_PO = 15 mL, pCO₂ = 2.0 MPa, T = 140 °C, t = 4 h, and W_catal. = 0.6 g.

As proved by the above FT-IR and XPS characterization results, the pristine GO possessed abundant C–OH groups on its surface, and the immobilization of the functionalized ILs on the GO support was derived from the condensation between the C–OH and Si–O–CH₃ groups. Even after the grafting procedure, there inevitably existed a small number of uncondensed C–OH groups on the surface of GO, as revealed by the XPS result (Fig. 6C). Based on the reports published previously,^49–51 a possible catalytic mechanism of the cycloaddition of CO₂ with PO under the present GO-immobilized catalysts have been proposed, as illustrated in Scheme S1.† Initially, the PO molecule was adsorbed and then activated by the free C–OH group of GO via the hydrogen bond (Step I). Next, the halide anion of the immobilized IL nucleophilically attacked the less hindered carbon atom of the PO (Step II), leading to the ring opening and thus generating a halogenoalkoxy anion. After that, the anion attacked the CO₂, and thus initiated the CO₂-inserting reaction (Step III), generating a halogenocarbonate intermediate. Afterward, the intermediate underwent a ring closing and transformed in a PC thorough intramolecular substitution (Step IV). Eventually, the hydrogen bond broke and the PC detached from the GO surface. Since Step IV was an intramolecular substitution of the halogen, the iodide anion with bigger bulk volume and higher leaving ability afforded superior catalytic activity to other halide anion, which can well explain the above variation (Table 1) of the activity using the immobilized ILs with different halides.

In order to test the catalytic versatility of the present IL-grafted GO materials, several catalytic cycloaddition reactions using other cyclic epoxides were conducted. As listed in Table 2, under the same reaction conditions above, various epoxides including ethylene (entry 1) and chloropropylene (entry 3) could be catalytically utilized to synthesize the corresponding cyclic carbonates. The moderated catalytic conversion adopting styrene oxide (entry 6) might be due to its steric hindrance. Notwithstanding, the above results clearly verified that the GO-[SmIm]I catalyst was active and selective for the cycloaddition of CO₂ with a wide range of epoxides. Furthermore, the abovementioned order of catalytic activity of GO-[SmIm]X with different halides (entries 3–5) was also reflected in Table 2.

Table 2 Catalytic performances of GO-[SmIm]X for the cycloaddition reactions using various cyclic epoxides^a

Entry	Cyclic epoxide	Conv. (%)	Yield (%)
a Reaction conditions: Vcyclic epoxide = 15 mL, pCO2 = 2.0 MPa, T = 140 °C, t = 4 h, and Wcatal. = 0.6 g. In each case, the selectivity to the target cyclic carbonate was higher than 99.5%.b Catalyzed by GO-[SmIm]I.c Catalyzed by GO-[SmIm]Br.d Catalyzed by GO-[SmIm]Cl.
1^b		98.2	97.7
2^b		96.4	95.9
3^b		94.7	94.2
4^c		83.6	83.2
5^d		65.3	65.1
6^b		62.1	61.0

---

Table 3 summarizes almost all the GO-based materials reported for the cycloaddition reactions of CO₂. In addition, the catalytic performances of carbon nitride materials have also been incorporated. Although the reaction conditions adopted in every entry deviate from each other, a rough comparison is still feasible. As listed in Table 3, the modified carbon nitride and GO materials, which contain quaternary ammonium or ionic liquid (IL) component, demonstrate superior catalytic activity to those pristine samples (entries 1, 2, and 4). Apparently, as mentioned above, quaternary ammonium and IL materials have been reported as the most active catalyst for the cycloaddition reaction of CO₂ with PC, as the halide anions thereof play an important role in activating the epoxide molecule. In our previous work, we synthesized mesoporous carbon nitride (mp-C₃N₄) materials grafted with n-bromobutane (i.e. n-butBr/mp-C₃N₄), and thereafter employed it as a catalyst for the cycloaddition reactions of CO₂ with PO.⁵² Although its catalytic activity (entry 3) is very close to the value gained in GO-[SmIm]I (entry 7), it is worth noting that, in terms of their preparation process, mp-C₃N₄ materials were synthesized using nanosized silica particles as hard templates via a nanocasting approach, which was time-consuming and not eco-friendly.⁵³ By contrast, the synthetic routes for the grafted GO materials are relatively simple, only involving one-step condensation.

Table 3 Comparison of the catalytic performances of GO-based and carbon nitride materials in the cycloaddition reactions of CO₂ with PO

Entry	Catalyst	T (°C)	t (h)	VPO (mL)	Wcatal. (mg)	Yield (%)	Catalyst feeding^g (mg mL^−1)
a Mesoporous carbon nitride prepared using mesoporous silica as a hard template and melamine as a precursor.b Mesoporous carbon nitride prepared using urea as a raw material without addition of any template.c Mesoporous carbon nitride grafted with n-butyl bromide, and containing quaternary ammonium units.d GO as a catalyst in the presence of 0.715 mmol of tetrabutylammonium bromide (Bu4NBr) as a co-catalyst.e GO material grafted with tetramethylethylenediamine, containing quaternary ammonium and iodide units.f This work.g Calculated based on the catalyst amount per mL of PO fed.
1^a ^54	MS-MCN	140	10	1.5	20	30.6	13.3
2^b ^8	u-g-C3N4-480	130	4	1.5	50	23.7	33.3
3^c ^52	n-butBr/mp-C3N4	140	6	10	200	87.7	20
4 ^43	GO	135	24	2	50	29.9	25
5^d ^43	GO	100	1	2	50	81.3	25
6^e ^47	GO-TMEDA-I	120	4	2	150	97.0	75
7^f	GO-[SmIm]I	140	4	15	600	96.0	40

---

In the case of entries 6 and 7, interestingly, both the two GO-based catalysts are grafted with quaternary ammonium species, using halogen-containing silanes for the first procedure of functionality on the surface of GO. Although the present GO-[SmIm]I demands slightly higher reaction temperatures and catalyst amounts, it should be noted that, the volume of feedstock, i.e. PO (entry 7) is much higher than it counterpart (entry 6). That is, under the similar catalytic activity, the catalytic feeding amount of GO-[SmIm]I (40 mg mL⁻¹) is significantly lower than that of GO-TMEDA-I (75 mg mL⁻¹). Therefore, combining their high activity and selectivity, and excellent reproductivity along with wide versatility, the IL-grafted GO materials in this study could be utilized as potentially efficient candidates towards the catalytic cycloaddition of CO₂ with epoxides to cyclic carbonates.

4. Conclusions

In summary, GO materials were directly grafted with functionalized imidazolium-based ILs with different halides via the covalent condensation between alcoholic C–OH of GO and alkoxyl groups of oxysilanes in ILs. As heterogeneous catalysts, GO-[SmIm]X samples exhibited potential catalytic application for the solvent-free cycloaddition of CO₂ with PO, where GO-[SmIm]I material demonstrated superior catalytic activity to GO-[SmIm]Br and GO-[SmIm]Cl. Under optimal reaction conditions, the PO yields obtained over GO-[SmIm]I was 95.9%. Also, various cycloaddition reactions using other cyclic epoxides could be selectively catalyzed by GO-[SmIm]I. In the light of the facile preparation, together with high activity and reproductivity, it is anticipated that the IL-grafted GO materials could serve new robust heterogeneous catalysts for the synthesis of cycloaddition reactions of CO₂.

Acknowledgements

This work was supported by National Natural Science Foundation of China (21376032), Jiangsu Key Laboratory of Advanced Catalytic Materials and Technology (BM2012110), and the Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions. J. Xu thanks Dr Xiazhang Li of Changzhou University for his kind help in the TEM characterization.

Notes and references

1. J. Roeser, K. Kailasam and A. Thomas, ChemSusChem, 2012, 5, 1793–1799.
2. S. Ravi, D. H. Kang, R. Roshan, J. Tharun, A. C. Kathalikkattil and D. W. Park, Catal. Sci. Technol., 2015, 5, 1580–1587.
3. W.-L. Dai, S.-L. Luo, S.-F. Yin and C.-T. Au, Appl. Catal., A, 2009, 366, 2–12.
4. J. Ma, N. Sun, X. Zhang, N. Zhao, F. Xiao, W. Wei and Y. Sun, Catal. Today, 2009, 148, 221–231.
5. R. Luo, X. Zhou, Y. Fang and H. Ji, Carbon, 2015, 82, 1–11.
6. W.-L. Dai, B. Jin, S.-L. Luo, S.-F. Yin, X.-B. Luo and C.-T. Au, J. CO₂ Util., 2013, 3–4, 7–13.
7. S. Zhong, L. Liang, B. Liu and J. Sun, J. CO₂ Util., 2014, 6, 75–79.
8. Q. Su, J. Sun, J. Wang, Z. Yang, W. Cheng and S. Zhang, Catal. Sci. Technol., 2014, 4, 1556–1562.
9. J. Xu, F. Wu, Q. Jiang, J.-K. Shang and Y.-X. Li, J. Mol. Catal. A: Chem., 2015, 403, 77–83.
10. S. Baj, T. Krawczyk, K. Jasiak, A. Siewniak and M. Pawlyta, Appl. Catal., A, 2014, 488, 96–102.
11. J. Sun, L. Wang, S. Zhang, Z. Li, X. Zhang, W. Dai and R. Mori, J. Mol. Catal. A: Chem., 2006, 256, 295–300.
12. J.-Q. Wang, K. Dong, W.-G. Cheng, J. Sun and S.-J. Zhang, Catal. Sci. Technol., 2012, 2, 1480–1484.
13. Y. Du, F. Cai, D.-L. Kong and L.-N. He, Green Chem., 2005, 7, 518–523.
14. A. H. Liu, L. N. He, S. Y. Peng, Z. D. Pan, J. L. Wang and J. Gao, Sci. China: Chem., 2010, 53, 1578–1585.
15. C.-X. Miao, J.-Q. Wang, Y. Wu, Y. Du and L.-N. He, ChemSusChem, 2008, 1, 236–241.
16. Y. Xie, T. T. Wang, X. H. Liu, K. Zou and W. Q. Deng, Nat. Commun., 2013, 4, 1960.
17. R. A. Watile, K. M. Deshmukh, K. P. Dhake and B. M. Bhanage, Catal. Sci. Technol., 2012, 2, 1051–1055.
18. J. Tharun, Y. Hwang, R. Roshan, S. Ahn, A. C. Kathalikkattil and D.-W. Park, Catal. Sci. Technol., 2012, 2, 1674–1680.
19. L. Han, H. Li, S.-J. Choi, M.-S. Park, S.-M. Lee, Y.-J. Kim and D.-W. Park, Appl. Catal., A, 2012, 429, 67–72.
20. X. Zhang, D. Wang, N. Zhao, A. S. Al-Arifi, T. Aouak, Z. A. Al-Othman, W. Wei and Y. Sun, Catal. Commun., 2009, 11, 43–46.
21. W. Cheng, X. Chen, J. Sun, J. Wang and S. Zhang, Catal. Today, 2013, 200, 117–124.
22. S. Udayakumar, M.-K. Lee, H.-L. Shim, S.-W. Park and D.-W. Park, Catal. Commun., 2009, 10, 659–664.
23. G. Vasilios, O. Michal, A. B. Bourlinos, C. Vimlesh, K. Namdong, K. K. Christian, H. Pavel, Z. Radek and K. S. Kim, Chem. Rev., 2012, 112, 6156–6214.
24. R. Zbořil, F. Karlický, A. B. Bourlinos, T. A. Steriotis, A. K. Stubos, V. Georgakilas, K. Šafářová, D. Jančík, C. Trapalis and M. Otyepka, Small, 2010, 6, 2885–2891.
25. M. B. Gawande, V. D. B. Bonifácio, R. Luque, P. S. Branco and R. S. Varma, ChemSusChem, 2014, 7, 24–44.
26. K. P. Loh, Q. L. Bao, G. Eda and M. Chhowalla, Nat. Chem., 2010, 2, 1015–1024.
27. D. R. Dreyer, S. Park, C. W. Bielawski and R. S. Ruoff, Chem. Soc. Rev., 2010, 39, 228–240.
28. L. R. Radovic, C. Mora-Vilches and A. J. A. Salgado-Casanova, Chin. J. Catal., 2014, 35, 792–797.
29. B. F. Machado and P. Serp, Catal. Sci. Technol., 2012, 2, 54–75.
30. C. Su, R. Tandiana, J. Balapanuru, W. Tang, K. Pareek, C. T. Nai, T. Hayashi and K. P. Loh, J. Am. Chem. Soc., 2015, 137, 685–690.
31. D. R. Dreyer, H.-P. Jia and C. W. Bielawski, Angew. Chem., Int. Ed., 2010, 49, 6813–6816.
32. G. K. P. Dathar, Y. T. Tsai, K. Gierszal, Y. Xu, C. D. Liang, A. J. Rondinone, S. H. Overbury and V. Schwartz, ChemSusChem, 2014, 7, 483–491.
33. A. Dhakshinamoorthy, M. Alvaro, P. Concepción, V. Fornés and H. Garcia, Chem. Commun., 2012, 48, 5443–5445.
34. W.-S. Ma, J. Li, B.-J. Deng and X.-S. Zhao, J. Mater. Sci., 2013, 48, 156–161.
35. W.-S. Ma, J. Li and X.-S. Zhao, J. Mater. Sci., 2013, 48, 5287–5294.
36. H. Yang, C. Shan, F. Li, D. Han, Q. Zhang and L. Niu, Chem. Commun., 2009, 3880–3882.
37. Y.-Y. Xu, X.-Y. Niu, Y.-L. Dong, H.-G. Zhang, X. Li, H.-L. Chen and X.-G. Chen, J. Chromatogr. A, 2013, 1284, 180–187.
38. J. Xu, H.-T. Wu, C.-M. Ma, B. Xue, Y.-X. Li and Y. Cao, Appl. Catal., A, 2013, 464, 357–363.
39. H. Yang, F. Li, C. Shan, D. Han, Q. Zhang, L. Niu and A. Ivaska, J. Mater. Chem., 2009, 19, 4632–4638.
40. F. Liu, J. Sun, L. Zhu, X. Meng, C. Qi and F.-S. Xiao, J. Mater. Chem., 2012, 22, 5495–5502.
41. Z. Lin, Y. Yao, Z. Li, Y. Liu, Z. Li and C.-P. Wong, J. Phys. Chem. C, 2010, 114, 14819–14825.
42. G. Eda and M. Chhowalla, Adv. Mater., 2010, 22, 2392–2415.
43. D.-H. Lan, F.-M. Yang, S.-L. Luo, C.-T. Au and S.-F. Yin, Carbon, 2014, 73, 351–360.
44. S. Thakur and N. Karak, Carbon, 2012, 50, 5331–5339.
45. B. Xue, J. Zhu, N. Liu and Y. Li, Catal. Commun., 2015, 64, 105–109.
46. S. Udayakumar, V. Raman, H.-L. Shim and D.-W. Park, Appl. Catal., A, 2009, 368, 97–104.
47. D.-H. Lan, L. Chen, C.-T. Au and S.-F. Yin, Carbon, 2015, 93, 22–31.
48. F. Li, L. Xiao, C. Xia and B. Hu, Tetrahedron Lett., 2004, 45, 8307–8310.
49. J. Sun, W. Cheng, W. Fan, Y. Wang, Z. Meng and S. Zhang, Catal. Today, 2009, 148, 361–367.
50. W.-L. Dai, L. Chen, S.-F. Yin, S.-L. Luo and C.-T. Au, Catal. Lett., 2010, 135, 295–304.
51. Y. Zhou, S. Hu, X. Ma, S. Liang, T. Jiang and B. Han, J. Mol. Catal. A: Chem., 2008, 284, 52–57.
52. J. Xu, F. Wu, Q. Jiang and Y.-X. Li, Catal. Sci. Technol., 2015, 5, 447–454.
53. J. Xu, K.-Z. Long, Y. Wang, B. Xue and Y.-X. Li, Appl. Catal., A, 2015, 496, 1–8.
54. M. B. Ansari, B.-H. Min, Y.-H. Mo and S.-E. Park, Green Chem., 2011, 13, 1416–1421.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra13533h

---