Bifunctional catalyst of a metallophthalocyanine-carbon nitride hybrid for chemical fixation of CO₂ to cyclic carbonate

Abstract

Chemical fixation of carbon dioxide (CO₂) to cyclic carbonates was investigated by using bifunctional nucleophile–electrophile catalysts of a metallophthalocyanine–carbon nitride hybrid [MPc/g-C₃N₄ (M = Co, Cu)] in the absence of any co-catalysts and organic solvents. MPc/g-C₃N₄ was readily obtained by direct calcination of a mixture of dicyandiamide and metallophthalocyanine under a flowing-nitrogen atmosphere, and the MPc/g-C₃N₄ prepared at 480 °C (MPc/g-C₃N₄-480) showed the highest catalytic performance toward the cycloaddition reaction of CO₂ to epichlorohydrin (ECH). For bifunctional MPc/g-C₃N₄, the MPc species function as Lewis acidic centers for ECH activation via electrophilic attack; while, the g-C₃N₄ moiety, possessing abundant and uncondensed species with the forms of primary amine (NH₂) groups and secondary amine (C–NH–C) groups at the edges of graphitic sheets as edge defects, acts as an organic base for CO₂ activation through nucleophilic attack. The developed MPc/g-C₃N₄ is stable and insoluble in any commonly used organic solvents and behaves as heterogeneous catalyst, leading to facile separation and recycling in a CO₂ fixation reaction.

---

Introduction

The chemistry of transformation of carbon dioxide (CO₂) to valuable chemicals has recently been extensively researched from both an economical and environmental point of view.^1,2 Although one of the major greenhouse gases, CO₂ is an abundant, economic, renewable and alternative C1 building block for our depleting and unrenewable fossil-based feedstock. Among various CO₂ utilizations, the cycloaddition of CO₂ with epoxides giving cyclic carbonates is one of the most efficient ways for CO₂ fixation.¹ The cyclic carbonates have been extensively employed as aprotic polar solvents, electrolytes for lithium-ion batteries, synthetic intermediates, precursors for biomedical applications, and raw materials for engineering plastics, pharmaceuticals and fine chemicals.²

For the cycloaddition reaction, a number of catalysts, such as transition metal complexes, organometallic compounds, ionic liquids, alkali metal salts, quaternary ammonium salts and phosphonium salts have been developed so far.^1j–s Among them, metallophthalocyanine (MPc) complexes are very attractive, not only because of their bio-inspired chemistry and structural analogy to metalloporphyrin in the active sites of various enzymes which are responsible for peroxide destruction, catalytic oxidation and reduction, and dioxygen transport in nature, but also due to their chemical and thermal stability as well as their facile preparation on a large scale.³ In addition, MPc complexes have been widely used in various material fields, such as liquid crystal, electrochromic and non-linear optical devices, semiconductor and information storage systems.⁴

In the literature research for MPc-promoted reactions, a series of MPc-based catalysts such as copper phthalocyanine (CuPc),⁵ MPc complexes (M = Cu²⁺, Co²⁺, Ni²⁺ and Al³⁺) encapsulated in zeolite-Y,⁶ and aluminum phthalocyanine complex covalently bonded to MCM-41 silica⁷ were previously investigated for cycloaddition of CO₂ and epoxides to produce cyclic carbonates. Moreover, cobalt phthalocyanine (CoPc) covalently immobilized on cellulose was reported for aerobic oxidation of alkylarenes and alcohols.⁸ CoPc immobilized on graphene oxide was investigated for photo induced carbon dioxide reduction to methanol.⁹ In situ synthesized CoPc/TiO₂ was explored for photo-induced CO₂ reduction to formic acid, methanol, formaldehyde in an aqueous solution using visible light irradiation.¹⁰ The methods of encapsulation and covalent immobilization of MPc species realize heterogenization of catalyst, provide easy separation and recycling of the MPc-derived catalyst, promote sufficient interactions between substrate and catalyst, and, most importantly, maintain MPc complex as a catalytically active monomer form by preventing it from self-dimerization.

Recent research demonstrated possibility of incorporating metalloporphyrin or metallophthalocyanine with sp²-hybridized carbon materials such as graphene via π–π interaction for various catalytic application. For instance, catalyst of graphene–hemin hybrid material was investigated for selective oxidation of pyrogallol, L-arginine and toluene.¹¹ Graphene-iron phthalocyanine (g-FePc) composite was reported as a promising Pt-free catalyst for oxygen reduction reactions.¹² Such kind of hybrids generally possess homogeneously distributed metalloporphyrin or metallophthalocyanine species throughout the surface of the graphene. The sp²-hybridized graphene behaves as supports to protect the metalloporphyrin and metallophthalocyanine species from aggregation or destruction in the catalytic reactions, and functions as a π donor to the metalloporphyrin and metallophthalocyanine species through cation–π or π–π interaction between them to modify the catalytic performance of the hybrids.¹³ In addition, the heterostructures of these hybrids provide a convenient preparation, separation and recycling of the conjugates from the reaction solution.¹⁴

In addition to graphene, graphitic carbon nitride (g-C₃N₄), as an analogue to graphite, possesses two-dimensional stacked structures with π-conjugated planar layers. Moreover, g-C₃N₄ is a typical solid base material and a series of mesoporous g-C₃N₄-derived materials were reported for various CO₂-related transformations.¹⁵ For instance, urea-derived graphitic carbon nitride,¹⁶ mesoporous carbon nitride grafted with n-bromobutane,¹⁷ and metal halides supported mesoporous carbon nitride¹⁸ were investigated for cycloaddition of CO₂ to epoxides to afford cyclic carbonates.

In the case of urea-derived graphitic carbon nitride, edge defects, the incompletely-coordinated nitrogen atoms, in the lower crystallinity and polymerization degree of u-g-C₃N₄ are presumably served as the main active sites in the cycloaddition reaction.¹⁶ For mesoporous carbon nitride grafted with n-bromobutane, the catalytic active sites are suggested to be uncondensed amines and Br anions which are obtained from the reaction between n-bromobutane and the N-containing heterocycles of mp-C₃N₄.¹⁷ Finally, a proposed mechanism of metal halides supported mesoporous carbon nitride-promoted cycloaddition reaction of CO₂ to propylene oxide generally involves a basic support of mp-C₃N₄ material to adsorb CO₂; whereas the active sites of supported metal halides promote the activation of epoxide and catalyze its transform into carbonates.¹⁸

In addition to catalyst, in some cases, co-solvent and co-catalysts play key roles on the cycloaddition reaction. For instance, the presence of co-solvent can facilitate the epoxides diffusion, promote the catalyst dispersion and reduce the reaction system viscosity.^1d While, co-catalysts such as 4-dimethylaminopyridine (DMAP), potassium iodide (KI), phenyltrimethylammonium tribromide (PTAT), tetrabutylammonium bromide (TBAB) and tetraphenylphosphonium bromide (TPPB) can function as nucleophilic reagents showing synergetic effect in promoting the cycloaddition reactions.^1s,t

Inspired by the above works, in this research, we prepared metallophthalocyanine–carbon nitride hybrids [MPc/g-C₃N₄ (M = Co, Cu)] as effective and bifunctional catalysts for cycloaddition of CO₂ to epichlorohydrin (ECH) to give cyclic 3-chloro-1,2-propylenecarbonate (CPC) under co-catalyst and solvent free conditions (Fig. 1 and 2). In the hybrid MPc/g-C₃N₄, the MPc species can function as mild Lewis acid site for epoxide activation by coordination of oxygen atom and subsequent polarization of the C–O bond of ECH in the cycloaddition of CO₂ to epoxides. While, the g-C₃N₄ moiety can act as solid base and behave as co-catalyst such as DMAP owing to the presence of a large number of uncondensed terminal amine and amine groups at the edges of the graphitic sheets.¹⁹ Moreover, the investigated MPc/g-C₃N₄ is readily available, air- and moisture-tolerant, recyclable catalyst for the cycloaddition reaction.

Fig. 1 MPc/g-C₃N₄ (M = Co, Cu) hybrid materials.

Fig. 2 MPc/g-C₃N₄ (M = Co, Cu)-promoted chemical fixation of CO₂ to cyclic carbonate.

Experimental

Materials

Unless otherwise stated, all chemicals in this work were commercially available and used without further purification. Dicyandiamide (98%) and epichlorohydrin (99.5%) were purchased from Aladdin Industrial Inc. (Shanghai, P.R. China). Cobalt(II) phthalocyanine (92%) and copper(II) phthalocyanine (90%) were supplied by J & K Chemical Technology (Beijing, P.R. China). Absolute ethanol was chemically pure and obtained from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, P.R. China). Carbon dioxide (CO₂ > 99.999%) was obtained from Huate Co. Ltd. (Foshan, P.R. China).

Synthesis of MPc/g-C₃N₄ (M = Co, Cu)

In a typical preparation of CoPc/g-C₃N₄ (CoPc 0.66 mmol g⁻¹), 1.5 g of dicyandiamide dissolved in 100 mL of ethanol was heated with cobalt phthalocyanine (CoPc, 600 mg, 1.05 mmol) and stirred at reflux temperature for 2 h. The mixed solution was then heated at 80 °C to remove ethanol. The resulting mixtures were then heated at a rate of 2.5 °C min⁻¹ to reach a temperature of 480 °C, and tempered at this temperature for 4 h under a flowing-nitrogen atmosphere. This was followed by cooling the sample naturally to room temperature under nitrogen gas. The final powder was collected and labelled as CoPc/g-C₃N₄-480 (0.66 mmol g⁻¹). CuPc/g-C₃N₄ was prepared following the same synthetic procedure as for CoPc/g-C₃N₄ used except that CoPc (600 mg) was replaced by CuPc (600 mg).

General procedure for coupling reaction of CO₂ and epoxide

All coupling reactions were performed in a 25 mL stainless steel autoclave equipped with a magnetic stir bar. In a typical reaction, the autoclave reactor was successively charged with CoPc/g-C₃N₄ (200 mg) and epichlorohydrin (3.70 g, 40.0 mmol). After purging the reactor several times with CO₂, the outlet valve was then closed to maintain at 3.0 MPa (room temperature) in the system controlled by a large CO₂ gas reservoir which was connected to the reaction cell. The reaction was performed at a stirring speed of 600 rpm at 130 °C for 24 h. After the reaction was halted, the reactor was cooled down to room temperature. The reaction mixture was subsequently extracted with CH₂Cl₂ and analyzed by GC (Agilent 6890) equipped with a FID detector and a KB-5 capillary column (0.32 mm × 30 m) with He as carrier gas. The reaction products were identified by GC-MS analysis (Thermo Quest Trace 2000). 3-Chloro-1,2-propylenecarbonate (CPC): ¹H NMR (400 MHz, 25 °C, CDCl₃) δ 4.98 (dddd, J = 8.4, 5.7, 4.8, 3.7 Hz, 1H), 4.64–4.47 (m, 1H), 4.36 (dd, J = 8.9, 5.7 Hz, 1H), 3.74 (ddd, J = 16.0, 12.3, 4.2 Hz, 2H). ¹³C {¹H} NMR (101 MHz, 25 °C, CDCl₃) δ 154.5, 74.5, 67.0, 44.1 (Fig. S4 and S5, ESI†).

Characterization

Fourier transform infrared (FT-IR) spectra of KBr wafers were recorded at room temperature in the 500–2200 cm⁻¹ region with a Bruker Tensor 27 spectrophotometer equipped with a Data Station at a spectral resolution of 1 cm⁻¹ and accumulations of 128 scans. Powder X-ray diffraction (XRD) patterns of CoPc/g-C₃N₄ and CuPc/g-C₃N₄ were obtained with a PANalytical X'pert Pro MPD diffractometer operated at 40 kV and 40 mA using Ni-filtered Cu-Kα radiation. X-ray photoelectron spectroscopy (XPS) spectra was performed with a Kratos Axis Ultra (DLD) photoelectron spectrometer operated at 15 kV and 10 mA at a pressure of about 5 × 10⁻⁹ torr using Al-Kα as the exciting source (hν = 1486.6 eV). C1s photoelectron peak (BE = 284.2 eV) was used for the binding energy calibration. Co and Cu contents of the samples were determined quantitatively by inductively coupled plasma mass spectrometry (ICP-AES) on an IRIS Advantage 1000 instrument. Thermogravimetric Analysis (TGA) of the CoPc/g-C₃N₄ and CuPc/g-C₃N₄ were carried out in a TGA Instruments thermal analyzer TA-SDT Q-600.

Results and discussion

Metallophthalocyanine–carbon nitride hybrid

In this research, MPc/g-C₃N₄ (M = Co, Cu) was readily obtained by a direct calcination of a mixture of dicyandiamide and metallophthalocyanine under flowing-nitrogen atmosphere. Herein, the MPc/g-C₃N₄ catalyst with various MPc contents and terminal calcination temperatures is denoted as MPc/g-C₃N₄-x (y) with x and y indicating terminal calcination temperature and MPc content in the unit of mmol g⁻¹, respectively. Proofs of the MPc/g-C₃N₄ (M = Co, Cu) formations were systematically revealed by Fourier transform infrared (FT-IR), X-ray diffraction (XRD), thermal gravimetric-differential thermal analysis, transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), and inductively coupled plasma-atomic emission spectroscopy (ICP-AES).

FT-IR spectra of g-C₃N₄-480, CoPc, and CoPc/g-C₃N₄-480 with various CoPc contents were compared in Fig. 3a. The FT-IR spectra of g-C₃N₄-480 show several absorption bands in the region of 1240 to 1650 cm⁻¹ corresponding to the typical stretching vibration of CN heterocycles.^16,20 The characteristic peak at 803 cm⁻¹ is assigned to the triazine units (C₆N₇), suggesting the formation of typical structure of g-C₃N₄.^20b,21 The broad peaks at around 3000–3400 cm⁻¹ were indicative of primary (terminal –NH₂ groups) and secondary (=NH) amines linked to the edges of graphitic CN sheets and physically adsorbed water (H–O–H) molecules.^18,20b,22

Fig. 3 (a) FT-IR spectra of (i) g-C₃N₄-480, (ii) CoPc/g-C₃N₄-480 (0.15), (iii) CoPc/g-C₃N₄-480 (0.36), (iv) CoPc/g-C₃N₄-480 (0.66), (v) CoPc; (b) FT-IR spectra of (i) CoPc/g-C₃N₄-450 (0.78), (ii) CoPc/g-C₃N₄-480 (0.66), (iii) CoPc/g-C₃N₄-520 (1.21), and (iv) CoPc/g-C₃N₄-550 (1.60).

The FT-IR spectra of CoPc/g-C₃N₄-480 with various CoPc contents and g-C₃N₄-480 closely resemble each other. As shown in Fig. 3a, with increasing content of CoPc in CoPc/g-C₃N₄-480, the characteristic absorption peaks of CoPc at 732 cm⁻¹ assigned as the Pc ring vibration become stronger,¹⁰ while that of the –NH₂ and –NH vibration between 3000 and 3400 cm⁻¹ become weaker, confirming the incorporation of CoPc species in CoPc/g-C₃N₄-480 and decrease in polymerization degree of g-C₃N₄ in CoPc/g-C₃N₄-480.

Fig. 3b shows CoPc/g-C₃N₄ samples with various terminal calcination temperatures. The multiple absorption bands of the CN heterocycles in the region of 1240–1650 cm⁻¹ become broader with the increasing temperature, while the broad bands located at 3000–3400 cm⁻¹ become weaker, further confirming the increase in polymerization degree. FT-IR analysis of CuPc/g-C₃N₄ with various CuPc contents and terminal calcination temperatures reveal similar tendency with that of CoPc/g-C₃N₄ samples (Fig. S1, ESI†).

XRD patterns of g-C₃N₄-480, CoPc, and CoPc/g-C₃N₄-480 with various CoPc contents are shown in Fig. 4a. In the case of g-C₃N₄-480, the strongest peak at 27.41° is a characteristic interlayer stacking peak of aromatic systems, indexed for graphitic materials as the (002) peak (Fig. 4a). While, the small angle peak at 13.08°, corresponding to interplanar distance, is indexed as (100), which is associated with interlayer stacking.²³ For the composite CoPc/g-C₃N₄-480, the intensity of characteristic diffraction peaks significantly decreased with increasing CoPc contents if compared with pristine g-C₃N₄-480, suggesting a slightly distorted structure of nitride pores after the increasing incorporation of CoPc species (Fig. 4a).²⁴ Notably, the intensity of diffraction peaks corresponding to CoPc species at 7.04, 9.26, 23.84, 33.08 and 40.43 (Fig. 4a, JCPDS file no. 44-1994) were observed increasing with CoPc contents in CoPc/g-C₃N₄-480 samples. In the case of CoPc/g-C₃N₄ samples obtained with various terminal calcinations temperatures, the XRD patterns show the same peak positions (Fig. 4b). However, these patterns generally display a slight broadening width and a decreasing intensity of the overall peaks following higher calcination temperatures. XRD analysis of CuPc/g-C₃N₄ with various CuPc contents and terminal calcinations temperatures reveal similar tendency with that of CoPc/g-C₃N₄ sample (Fig. S2, ESI†).

Fig. 4 (a) XRD patterns of (i) g-C₃N₄-480, (ii) CoPc/g-C₃N₄-480 (0.15), (iii) CoPc/g-C₃N₄-480 (0.36), (iv) CoPc/g-C₃N₄-480 (0.66), (v) CoPc; (b) XRD patterns of (i) CoPc/g-C₃N₄-450 (0.78), (ii) CoPc/g-C₃N₄-480 (0.66), (iii) CoPc/g-C₃N₄-520 (1.21), and (iv) CoPc/g-C₃N₄-550 (1.60).

The TEM images of both CoPc/g-C₃N₄ and CuPc/g-C₃N₄ showed the typical layered platelet-like morphology of g-C₃N₄, indicating the MPc (M = Co, Cu) species were found to be uniformly distributed through g-C₃N₄ support in MPc/g-C₃N₄ without any agglomeration (Fig. 5).

Fig. 5 TEM micrographs of (a) CoPc/g-C₃N₄-480 (0.66) and (b) CuPc/g-C₃N₄-480 (0.80).

The thermal stabilities of CoPc/g-C₃N₄-480 (0.66) and CuPc/g-C₃N₄-480 (0.80) were investigated by thermogravimetric analysis to confirm the stability of the hybrid materials. The curves of weight loss versus temperature for CoPc/g-C₃N₄-480 and CuPc/g-C₃N₄-480 were plotted in Fig. S3 (ESI†). The significant weight loss of hybrid samples, which corresponds to sample degradation, starts at about 615 °C for CoPc/g-C₃N₄-480 and 670 °C for CuPc/g-C₃N₄-480, respectively.

The surface compositions of CoPc/g-C₃N₄ and CuPc/g-C₃N₄ were further investigated with XPS. Peaks corresponding to carbon, nitrogen, cobalt for CoPc/g-C₃N₄ and copper for CuPc/g-C₃N₄ were presented in the survey scan (Fig. 6a). Fig. 6b shows the N(1s) XPS spectrum of CoPc/g-C₃N₄ and CuPc/g-C₃N₄. The N(1s) line of CoPc/g-C₃N₄ was deconvoluted into three superimposed peaks at 398.7, 400.4 and 401.7 eV corresponding to the aromatic nitrogen bonded to two carbon (C=N–C) in triazine or heptazine rings,¹⁶ the sp² hybridized nitrogen connected to three atoms (C–N(–C)–C or C–N(–H)–C),²⁵ and the sp³ hybridized terminal nitrogen (such as N–H₂ and N–O) of the heptazine rings.²⁶ In the case of the N(1s) XPS of CuPc/g-C₃N₄, two peaks at 398.4 and 400.4 eV was evidently observed as depicted in Fig. 6b.

Fig. 6 (a) XPS scan survey and (b) N(1s) XPS spectra for CoPc/g-C₃N₄-480 (0.66) and CuPc/g-C₃N₄-480 (0.80); (c) Co(2p) XPS spectra of CoPc/g-C₃N₄; and (d) Cu(2p) XPS spectra of CuPc/g-C₃N₄.

As shown in Fig. 6c, the Co(2p) XPS spectrum of CoPc/g-C₃N₄ shows two main peaks located at 796.0 and 780.8 eV corresponding to Co(2p_1/2) and Co(2p_3/2), respectively, indicating the presence of Co(II)-centered CoPc species in CoPc/g-C₃N₄.²⁷ In the case of CuPc/g-C₃N₄, the presence of Cu(II) species was supported by the appearance of two main peaks at 954.6 and 934.7 eV assigned to Cu(2p_1/2) and Cu(2p_3/2), respectively, as shown in Fig. 6d, which is closer to the Cu(II) position of a Cu(II)Pc.²⁸ The Co(2p) XPS analysis of CoPc/g-C₃N₄ and Cu(2p) XPS analysis of CuPc/g-C₃N₄ therefore confirmed successful formation of MPc/g-C₃N₄ composites.

Cycloaddition of CO₂ to epichlorohydrin

Initially, MPc/g-C₃N₄ (M = Co, Cu) catalysts with various MPc contents and terminal calcination temperatures were employed to investigate the effects of MPc contents and calcination temperatures on the cycloaddition reaction of CO₂ to ECH. The influences of CoPc content on the catalytic performances of CoPc/g-C₃N₄ catalysts indicated that the CPC yield increased from 82.3% to 97.6% with the CoPc content ranging from 0.15 mmol g⁻¹ to 0.66 mmol g⁻¹ in CoPc/g-C₃N₄-480 under the investigated conditions (Fig. 7). While, the effect of calcination temperatures on the catalyst to the cycloaddition reaction revealed that both CoPc/g-C₃N₄ and CuPc/g-C₃N₄ exhibit an increased catalytic performance with a decreasing calcination temperature from 550 °C to 480 °C and then a decrease at 450 °C. Among all the MPc/g-C₃N₄ (M = Co, Cu) catalysts obtained at various temperatures, both CoPc/g-C₃N₄-480 and CuPc/g-C₃N₄-480 were found to be the best one, achieving CPC yields 97.6% and 93.6%, respectively. In contrast, the cycloaddition reactions promoted by CoPc and g-C₃N₄ provided CPC yields of 29.9% and 11.7% respectively, under the investigated conditions. Our observation of the influence of calcinations temperature on the MPc/g-C₃N₄ (M = Co, Cu) catalysts is in line with the urea-derived graphitic carbon nitride catalyst u-g-C₃N₄ in the cycloaddition of CO₂ to epoxides.¹⁶ The above results thus indicate that a lower calcination temperature leads to a lower crystallinity and polymerization degree with more edge defects generated.¹⁶ While, the edge defects, mainly consisting of the uncondensed amino groups including the primary amine (NH₂) groups and secondary amine (C–NH–C) groups, play a key role on the cycloaddition reaction.

Fig. 7 MPc/g-C₃N₄ (M = Co, Cu) promoted cycloaddition reaction of CO₂ to ECH and the influence of the MPc content (in M mmol g⁻¹), terminal calcination temperatures (in °C). Reaction conditions: ECH (3.70 g, 40.0 mmol), MPc/g-C₃N₄ (200 mg), CO₂ (3.0 MPa), 130 °C, 24 h in a 25 mL autoclave.

Fig. 8a shows the influence of reaction temperature and time on the CoPc/g-C₃N₄-promoted coupling reaction of CO₂ and ECH without using co-catalyst and solvent. CPC yield generally increased with reaction time at all reaction temperatures investigated. Reaction temperature dramatically promoted the cycloaddition reaction and CPC yield significantly increased from 56.2% at 110 °C to 100% at 130 °C after 28 h. Notably, an induction period around 0–12 h was observed when the cycloaddition reaction was performed at low reaction temperature of 110 °C. In addition, the induction period decreased with reaction temperature. ECH has chloromethyl group which might undergo some reactions with the amino group of g-C₃N₄, giving off the Cl⁻ anion. Thus, Cl⁻ anion is suggested to function as a nucleophile for the epoxide ring opening, which might be related to the observed induction period. To rule out this possibility, KCl was added to the cycloaddition reaction at 110 °C and the induction period disappeared with increased CPC yield. However, KCl shows very limited promotion effect on the CPC yield after the induction period. While, co-catalyst KI can significant enhance CPC yield under the investigated conditions, suggesting the presence of the synergistic effect between CoPc/g-C₃N₄ and I⁻ anion. Moreover, several other epoxides having no chloromethyl group, such as propylene oxide and styrene oxide were examined and reduced yields of the corresponding cyclic carbonates were observed (Table 1). The effect of initial CO₂ pressure on the cycloaddition reaction revealed that the increase in CO₂ pressure from 1.0 to 3.0 MPa promoted the CPC yield from 70.3% to 97.6% (Fig. 8b).

Fig. 8 Effect of (a) reaction temperature and time, and (b) initial CO₂ pressure on the cycloaddition reaction of CO₂ to ECH. Reaction conditions: ECH (3.70 g, 40.0 mmol), CoPc/g-C₃N₄ (200 mg, CoPc 0.66 mmol g⁻¹), in a 25 mL autoclave, [CO₂ (3.0 MPa), 110–130 °C, 6–32 h], for (a) and [CO₂ (1.0–3.5 MPa), 130 °C, 24 h] for (b). In the cases of (iv) and (v) of (a), KI (1 mmol) and KCl (1 mmol) were added as co-catalyst, respectively.

Table 1 Synthesis of cyclic carbonates using CoPc/g-C₃N₄ catalyst^a

Entry	Substrate	Product	Yield^b [%]
a Reaction conditions: epoxide (40.0 mmol), CoPc/g-C3N4 (200 mg, CoPc 0.66 mmol g^−1), CO2 (3.0 MPa), 130 °C, 24 h in a 25 mL autoclave.b Isolated yields.
1			97.6
2			72.1
3			69.3
4			11.5

---

The influence of catalyst loading amounts on the cycloaddition reaction revealed that the CPC yield steeply increased with the catalyst loading levels, with a significant rise to CPC yield of 97.6% for a 0.33 mol% [CoPc] species relative to ECH (Fig. 9). Further increase of the CoPc/g-C₃N₄ loading led to a decrease in CPC yield.

Fig. 9 Effect of catalyst loading amount on the cycloaddition reaction of CO₂ to ECH. Reaction conditions: ECH (3.70 g, 40.0 mmol), CoPc/g-C₃N₄ (CoPc 0.66 mmol g⁻¹), CO₂ (3.0 MPa), 130 °C, 24 h.

Scheme 1 shows a proposed mechanism of bifunctional MPc/g-C₃N₄ (M = Co, Cu) catalyst-promoted cycloaddition reaction of CO₂ to ECH under co-catalyst and solvent free conditions. ECH is activated by a Lewis acidic center of MPc/g-C₃N₄ (M = Co, Cu) through coordination of oxygen atom of ECH to MPc species and subsequent polarization of C–O bond of ECH to form metal alkoxide intermediate. Owing to insufficient polycondensation, g-C₃N₄ moiety of MPc/g-C₃N₄ possesses abundant uncondensed species in the forms of primary amine (NH₂) groups and secondary amine (C–NH–C) groups at the edges of graphitic sheets as edge defects.^16,18,19 CO₂ molecules could presumably be activated by these edge defects of g-C₃N₄ via O–H bonding or nucleophilic attack. Under such condition, the g-C₃N₄ moiety of MPc/g-C₃N₄ can function as co-catalyst such as organic base DMAP to activate CO₂.^18,19 The activated CO₂ complex attacks the metal alkoxide intermediate to yield the metal carbonate that leads to CPC formation and regeneration of catalyst MPc/g-C₃N₄.

Scheme 1 Proposed mechanism for the coupling of CO₂ and ECH promoted by MPc/g-C₃N₄ (M = Co, Cu).

The process of CO₂ activation and fixation was investigated by using homogeneous system of CoPc and melamine in CH₃CN and monitored by UV-vis and FT-IR. CoPc–melamine–CH₃CN had a weak d–d band at around 600 nm (Fig. 10a). When the solution was treated with CO₂, a new band was observed at 426 nm and can be assigned to metal-to-ligand charge transfer transitions [Co (d orbitals) → CO₂ (π* orbitals)], indicating the formation of an activated CO₂ species.⁵ The observed d–d band further shifted to 494 nm when CoPc–melamine–CO₂–CH₃CN was reacted with ECH, suggesting the structure change of the Co complex due to the formation of CPC (Fig. 10a). The FT-IR analysis revealed that the treatment of CO₂ with the CoPc–melamine system showed two new IR peaks at 1606 and 1226 cm⁻¹, suggesting the formation of an activated CO₂ complex (Fig. 10b).⁵ When ECH reacted with CoPc–melamine–CO₂ in CH₃CN, additional peaks were observed at 1803, 1168 and 1074 cm⁻¹ with the disappearance of CO₂-complex (at 1606 and 1226 cm⁻¹), indicating the formation of CPC.

Fig. 10 (a) UV-vis spectra in CH₃CN: (i) CoPc + melamine, (ii) CoPc + melamine + CO₂, (iii) CoPc + melamine + ECH, (iv) CoPc + melamine + CO₂ + ECH; and (b) FT-IR spectra in CH₃CN: (i) CoPc + melamine, (ii) CoPc + melamine + CO₂, (iii) CoPc + melamine + CO₂ + ECH. Asterisk indicates the characteristic IR peaks for CPC.

Under the optimized reaction parameters, the CoPc/g-C₃N₄ catalyst was further investigated for the cycloaddition of CO₂ to various epoxides under solvent and co-catalyst free conditions (Table 1). In the cases of mono-substituted terminal epoxides with aliphatic, aromatic, electron-withdrawing and electron-donating substituents, the CoPc/g-C₃N₄ catalyst provided the corresponding cyclic carbonate products in excellent to moderate yields (Table 1, Entries 1–3). For cyclohexene oxide, the corresponding cyclic carbonate was obtained in very low yield of 11.5% (Table 1, Entry 4), presumably due to the steric hindrance of the cyclohexene oxide.

A six-cycle experiment was performed to probe the reusability of CoPc/g-C₃N₄ in the cycloaddition reactions. Fig. 11 shows the obtained CPC yields reduced from 97.6 to 89.5%. The decreased catalytic performance of CoPc/g-C₃N₄ in the recycling can presumably be related to the CoPc species leaching from 0.66 mmol g⁻¹ for the fresh catalyst to 0.60 mmol g⁻¹ after the cycle experiment based on ICP-AES analysis.

Fig. 11 Catalyst recycling. Reaction conditions: ECH (3.70 g, 40.0 mmol), CoPc/g-C₃N₄ (200 mg, CoPc 0.66 mmol g⁻¹), CO₂ (3.0 MPa), 130 °C, 24 h, in a 25 mL autoclave.

Conclusion

In this research, metallophthalocyanine–carbon nitride hybrid [MPc/g-C₃N₄ (M = Co, Cu)] have been successfully synthesized by a direct calcination of a mixture of dicyandiamide and metallophthalocyanine. The MPc/g-C₃N₄ are bifunctional nucleophile–electrophile catalysts for cycloaddition reaction of CO₂ to epichlorohydrin (ECH) to give cyclic carbonate 3-chloro-1,2-propylenecarbonate (CPC) in the absence of any co-catalysts and organic solvents. In addition, the MPc/g-C₃N₄ prepared at 480 °C (MPc/g-C₃N₄-480) showed the highest catalytic performance toward the cycloaddition reaction. Under optimal reaction conditions, CPC can be obtained with the yields of 97.6% with full conversion of ECH. The catalyst MPc/g-C₃N₄ was readily separated and effectively recycled for up to five cycles without a significant loss in its activity under identical conditions.

Acknowledgements

We are grateful for the financial support from National Natural Science Foundation of China (21172219 and 51408522) and Natural Science Foundation of Guangdong Province, China (2015A030312007).

Notes and references

1. (a) T. Sakakura, J. C. Choi and H. Yasuda, Chem. Rev., 2007, 107, 2365–2387; (b) D. J. Darensbourg, Chem. Rev., 2007, 107, 2388–2410; (c) S. N. Riduan and Y. Zhang, Dalton Trans., 2010, 39, 3347–3357; (d) S. Klaus, M. W. Lehenmeier, C. E. Anderson and B. Rieger, Coord. Chem. Rev., 2011, 255, 1460–1479; (e) M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann and F. E. Kühn, Angew. Chem., Int. Ed., 2011, 50, 8510–8537; (f) Y. Tsuji and T. Fujihara, Chem. Commun., 2012, 48, 9956–9964; (g) D. J. Darensbourg and S. J. Wilson, Green Chem., 2012, 14, 2665–2671; (h) I. Omae, Coord. Chem. Rev., 2012, 256, 1384–1405; (i) N. Kielland, C. J. Whiteoak and A. W. Kleij, Adv. Synth. Catal., 2013, 355, 2115–2138; (j) M. North, R. Pasquale and C. Young, Green Chem., 2010, 12, 1514–1539; (k) C. Maeda, Y. Miyazaki and T. Ema, Catal. Sci. Technol., 2014, 4, 1482–1497; (l) M. Mikkelsen, M. Jørgensen and F. C. Krebs, Energy Environ. Sci., 2010, 3, 43–81; (m) T. Sakakura and K. Kohno, Chem. Commun., 2009, 1312–1330; (n) C. J. Whiteoak and A. W. Kleij, Synlett, 2013, 24, 1748–1756; (o) M. Q. Zhu and M. A. Carreon, J. Appl. Polym. Sci., 2014, 131, 39738; (p) J. Sun, X. Q. Yao, W. G. Cheng and S. J. Zhang, Green Chem., 2014, 16, 3297–3304; (q) J. Sun, W. G. Cheng, Z. F. Yang, J. Q. Wang, T. T. Xu, J. Y. Xin and S. J. Zhang, Green Chem., 2014, 16, 3071–3078; (r) S. G. Liang, H. Z. Liu, T. Jiang, J. L. Song, G. Y. Yang and B. X. Han, Chem. Commun., 2011, 47, 2131–2133; (s) A. B. Chen, Y. Z. Zhang, J. Z. Chen, L. M. Chen and Y. F. Yu, J. Mater. Chem. A, 2015, 3, 9807–9816; (t) A. B. Chen, T. C. Zhao, H. Gao, L. M. Chen, J. Z. Chen and Y. F. Yu, Catal. Sci. Technol., 2015 10.1039/c5cy01024a.
2. (a) D. R. Moore, M. Cheng, E. B. Lobkovsky and G. W. Coates, Angew. Chem., Int. Ed., 2002, 41, 2599–2602; (b) B. Schäffner, J. Holz, S. P. Verevkin and A. Börner, ChemSusChem, 2008, 1, 249–253; (c) P. Tundo and M. Selva, Acc. Chem. Res., 2002, 35, 706–716; (d) H. Sugimoto and K. Kuroda, Macromolecules, 2008, 41, 312–317; (e) J. Bayardon, J. Holz, B. Schäffner, V. Andrushko, S. Verevkin, A. Preetz and A. Börner, Angew. Chem., Int. Ed., 2007, 46, 5971–5974; (f) S. Mang, A. I. Cooper, M. E. Colclough, N. Chauhan and A. B. Holmes, Macromolecules, 2000, 33, 303–308; (g) L. M. Stamp, S. A. Mang, A. B. Holmes, K. A. Knights, Y. R. de Miguel and I. F. McConvey, Chem. Commun., 2001, 2502–2503.
3. A. B. Sorokin, Chem. Rev., 2013, 113, 8152–8191.
4. (a) G. de la Torre, C. G. Claessens and T. Torres, Chem. Commun., 2007, 2000–2015; (b) F. D'Souza and O. Ito, Chem. Commun., 2009, 4913–4928; (c) M. V. Martinez-Diaz, G. de la Torre and T. Torres, Chem. Commun., 2010, 46, 7090–7108; (d) G. Bottari, G. de la Torre, D. M. Guldi and T. Torres, Chem. Rev., 2010, 110, 6768–6816; (e) G. Bottari, J. A. Suanzes, O. Trukhina and T. Torres, J. Phys. Chem. Lett., 2011, 2, 905–913.
5. R. Srivastava, T. H. Bennur and D. Srinivas, J. Mol. Catal. A: Chem., 2005, 226, 199–205.
6. R. Srivastava, D. Srinivas and P. Ratnasamy, Catal. Lett., 2003, 89, 81–85.
7. X. B. Lu, H. Wang and R. He, J. Mol. Catal. A: Chem., 2002, 186, 33–42.
8. A. Shaabani, S. Keshipour, M. Hamidzad and S. Shaabani, J. Mol. Catal. A: Chem., 2014, 395, 494–499.
9. P. Kumar, A. Kumar, B. Sreedhar, B. Sain, S. S. Ray and S. L. Jain, Chem.–Eur. J., 2014, 20, 6154–6161.
10. Z. H. Zhao, J. M. Fan, M. M. Xie and Z. Z. Wang, J. Cleaner Prod., 2009, 17, 1025–1029.
11. (a) Y. J. Li, X. Q. Huang, Y. J. Li, Y. X. Xu, Y. Wang, E. B. Zhu, X. F. Duan and Y. Huang, Sci. Rep., 2013, 3, 1787; (b) T. Xue, S. Jiang, Y. Q. Qu, Q. Su, R. Cheng, S. Dubin, C. Y. Chiu, R. Kaner, Y. Huang and X. F. Duan, Angew. Chem., Int. Ed., 2012, 124, 3888–3891.
12. Y. Y. Jiang, Y. Z. Lu, X. Y. Lv, D. X. Han, Q. X. Zhang, L. Niu and W. Chen, ACS Catal., 2013, 3, 1263–1271.
13. (a) M. E. Ragoussi, G. Katsukis, A. Roth, J. Malig, G. de la Torre, D. M. Guldi and T. Torres, J. Am. Chem. Soc., 2014, 136, 4593–4598; (b) G. I. Cárdenas-Jirón, P. León-Plata, D. Cortes-Arriagada and J. M. Seminario, J. Phys. Chem. C, 2013, 117, 23664–23675.
14. J. X. Geng and H. T. Jung, J. Phys. Chem. C, 2010, 114, 8227–8234.
15. Z. Huang, F. Li, B. Chen, T. Lu, Y. Yuan and G. Yuan, Appl. Catal., B, 2013, 136, 269–277.
16. Q. Su, J. Sun, J. Q. Wang, Z. F. Yang, W. G. Cheng and S. J. Zhang, Catal.: Sci. Technol., 2014, 4, 1556–1562.
17. J. Xu, F. Wu, Q. Jiang and Y. X. Li, Catal.: Sci. Technol., 2015, 5, 447–454.
18. J. Xu, F. Wu, Q. Jiang, J. K. Shang and Y. X. Li, J. Mol. Catal. A: Chem., 2015, 403, 77–83.
19. R. L. Paddock, Y. Hiyama, J. M. McKay and S. T. Nguyen, Tetrahedron Lett., 2004, 45, 2023–2026.
20. (a) D. Mitoraj and H. Kisch, Chem.–Eur. J., 2010, 16, 261–269; (b) G. D. Ding, W. T. Wang, T. Jiang, B. X. Han, H. L. Fan and G. Y. Yang, ChemCatChem, 2013, 5, 192–200.
21. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2009, 25, 10397–10401.
22. J. Xu, H. T. Wu, X. Wang, B. Xue, Y. X. Li and Y. Cao, Phys. Chem. Chem. Phys., 2013, 15, 4510–4517.
23. (a) B. Yue, Q. Li, H. Iwai, T. Kako and J. H. Ye, Sci. Technol. Adv. Mater., 2011, 12, 034401; (b) W. Wang, G. Ding, T. Jiang, P. Zhang, T. Wu and B. X. Han, Green Chem., 2013, 15, 1150–1154; (c) J. Z. Chen, Y. Y. Guo, J. Y. Chen, L. Song and L. M. Chen, ChemCatChem, 2014, 6, 3174–3181.
24. Y. Zhang, A. Thomas, M. Antonietti and X. Wang, J. Am. Chem. Soc., 2009, 131, 50–51.
25. D. Mitoraj and H. Kisch, Angew. Chem., Int. Ed., 2008, 47, 9975–9978.
26. D. R. Miller, J. Wang and E. G. Gillan, J. Mater. Chem., 2002, 12, 2463–2469.
27. W. Zhu, W. J. Zhang, S. Li, H. Y. Ma, W. Chen and H. J. Pang, Sens. Actuators, B, 2013, 181, 773–781.
28. S. S. Mali, H. J. Kim, J. H. Kim, P. S. Patil and C. K. Hong, Ceram. Interfaces, 2014, 40, 643–649.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra21058e

---