Highly efficient synthesis of dimethyl carbonate from

Catalysis Communications 75 (2016) 87–91

Contents lists available at ScienceDirect

Catalysis Communications journal homepage: www.elsevier.com/locate/catcom

Highly efficient synthesis of dimethyl carbonate from methanol and carbon dioxide using IL/DBU/SmOCl as a novel ternary catalytic system Avinash A. Chaugule, Harshad A. Bandhal, Ashif H. Tamboli, Wook-Jin Chung, Hern Kim ⁎ Department of Energy Science and Technology, Energy and Environment Fusion Technology Center, Myongji University, Yongin, Gyeonggi-do 17058, South Korea

a r t i c l e

i n f o

Article history: Received 24 September 2015 Received in revised form 1 December 2015 Accepted 12 December 2015 Available online 15 December 2015 Keywords: Ionic liquid Carbon dioxide Samarium oxychloride Dimethyl carbonate

a b s t r a c t Excellent yield of dimethyl carbonate (DMC) was obtained by direct physical or chemical adsorption of carbon dioxide on [EmimOH][NTf 2] ionic liquid (IL) in the presence of samarium oxychloride (SmOCl) and 1,8-diazabicyclo[2.2.2]undec-7-ene (DBU) as a super base. The novel ternary catalyst system consisting of [EmimOH][NTf2], DBU, and SmOCl was found to appreciably convert methanol (13.01%) to DMC with excellent selectivity (99.13%). The adsorption of CO2 on IL in the presence of DBU was analyzed by 13C experiment. Moreover, catalytic reactivity of SmOCl and OH-functional group was proved by a predictable mechanism. Various parameters such as reaction temperature, pressure, reaction time, and reusability of catalyst were investigated to maximize DMC yield. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Carbon dioxide (CO2) is a primary greenhouse gas produced by human activities. Because of its enormous emission, CO2 contributes N60% to global warming [1]. On the contrary, CO2 is an abundant nontoxic and recyclable carbon source that can substitute conventional toxic chemicals such as phosgene, isocyanides, and carbon monoxide [2]. Synthesis of carbonic acid esters has attracted considerable research attention recently. Dimethyl carbonate (DMC) is the lowest homology of this family, which has several industrial applications, including use as methylating [3] and carbonylating agent [4] instead of toxic dimethyl sulfate and phosgene, food-flavoring agent [5], solvent for secondary lithium-ion batteries [6], and reagent for biodiesel production [7]. Phosgenation was the first commercial method for the production of DMC, but due to the extreme toxicity of phosgene, this process has been replaced by processes such as oxidative carbonylation and transesterification. Major drawbacks of oxidative carbonylation are the use of poisonous and corrosive gases and explosion risk associated with the use of CO and O2 mixture. On the contrary, direct synthesis of DMC from CO2 and methanol could be an ideal alternative for DMC production at commercial level. The use of cheap, abundant, and noncorrosive raw materials (CO2 and MeOH) makes it an environmentally benign process [8]. Although a number of catalysts such as Ni(acetate)2, ZrO2, polyphosphoric acid–ZrO2, ZrO2–MgO, CeO2, and Ce–ZrO2 solid solutions have been investigated for this reaction, DMC yield still remains low [9,10]. The use of stoichiometric amounts of dehydrating agents can help increase DMC yield considerably. However, most of ⁎ Corresponding author. E-mail address: [email protected] (H. Kim).

http://dx.doi.org/10.1016/j.catcom.2015.12.009 1566-7367/© 2015 Elsevier B.V. All rights reserved.

these agents are irrecoverable or require a regeneration process. Therefore, their use has some limitations in terms of production cost. In addition, DMC reaction also proceeded smoothly with high yield using catalyst including tin alkoxy compounds such as Bu2Sn (OEt)2, Sn(OMe)4, and Sn(OBu)4 [10]. In recent years, ionic liquids (ILs) have attracted increasing attention as catalyst due to their broad liquid range and excellent thermal and chemical stability [11,12]. Concentration of CO2 in solution increased significantly by adding IL into the reaction mixture. In addition to being a CO2 sorbent, ILs can act as a water scavenger. Therefore, their use as a solvent, promoter, or catalyst has been extensively investigated by several groups. In this study, basic IL is synthesized, whose combination with an organic super base, 1,8-diazabicyclo[2.2.2]undec-7-ene (DBU), can reversibly capture CO2. The effect of addition of metal salts is also investigated. This ternary system consisting of an IL, organic base, and metal oxychloride was found to be highly efficient for catalyzing direct DMC synthesis.

2. Experimental 2.1. Preparation of IL 2.1.1. Preparation of 1-(2-hydroxyethyl)-3-methyl imidazolium chloride [EmimOH][Cl] IL 1-Methyl imidazole (10.0 g 0.12 mol) and 2-chloroethanol (14.7 g, 0.12 mol) were taken in a 100-ml round-bottomed flask and refluxed with 40-ml acetonitrile under continuous stirring for 28 h. Then, the reaction mixture was slowly cooled to −4 °C to afford a white crystalline solid as 1-(2-hydroxyethyl)-3-methyl imidazolium chloride IL.

88

A.A. Chaugule et al. / Catalysis Communications 75 (2016) 87–91

Table 1 Catalytic performance of various ILs for synthesis of DMC. Entry

Catalyst/additive

Methanol conversion [%]

DMC selectivity [%]

Yield [%]

1 2 3 4 5 6

–a DBU [EmimOH][Cl] [EmimOH][NTF2] [EmimOH][NTF2] + DBU [EmimOH][NTF2] + DBU + SmOCl

– 1 14.3 8.1 9.5 13.0

– 2 33.7 60.0 72.2 99.1

– 0 4.8 4.8 6.9 12.8

a Reaction carried out without any catalyst. Reaction conditions: 783 mol of methanol, 6.4 MPa CO2, 4 g IL, 1 g DBU, 4 h.

2.1.2. Preparation of 1-(2-hydroxyethyl)-3-methylimidazolium bis(trifluoromethanesulfonyl) imide [EmimOH][NTf2] IL 1-(2-Hydroxyethyl)-3-methyl imidazolium chloride (6 g, 0.018 mol) was reacted with bis(trifluoromethanesulfonyl)imide lithium salt (10.6 g, 0.018 mol) and 40 ml of acetone under stirring at room temperature for 28 h. Then, the mixture was filtered to remove LiCl salt formed during the reaction. The filtrate was then evaporated under reduced pressure to obtain [HEMIm] [NTf2] IL, which was subjected to silver nitrate test to ensure complete removal of chloride.

spectrum (FTIR, Varian 2000). Thermal decomposition temperature was measured using thermogravimetric analysis (TGA, PerkinElmer TGA-7 instrument). Differential scanning calorimetry (DSC) data were obtained in a sealed aluminum pan with a cooling and heating rate of 10 °C/min on Mettler DSC822. The general structure of the compounds was examined by X-ray diffraction (XRD) using an analytical X'Pert MPD diffractometer with Bragg's angle ranging from 20° to 80°. 2.3. DMC synthesis from methanol and CO2 Methanol (25 ml, 0.625 mol) and [HEMIm][NTf2] IL (4 g, 0.07 mol) with DBU (0.007 mol) and samarium oxychloride (SmOCl) (0.005 mol) were charged into the reactor. The reaction was carried out at 140 °C under continuous stirring at 700 rpm for 4 h. After completion of reaction, the reaction mixture was distilled to separate the IL. The DMC in liquid phase was analyzed by gas chromatography (GC17A, Shimadzu Corporation) using capillary column (Stabilwax, 30m-long, 0.53-mm inner diameter, and 1-μm film thickness) equipped with a flame ionization detector. The amount of the product was determined using the external standard method. Furthermore, DMC was identified by 1 H NMR, FTIR, and gas chromatography–mass spectrometry (GC–MS). 3. Results and discussion

2.2. Characterization 3.1. Catalyst characterization All synthesized ILs were characterized by 1H nuclear magnetic resonance (NMR) and 13C NMR spectroscopy using Bruker 500 MHz and 125 MHz spectrometers, respectively, in CDCl3. Infrared spectra of the prepared ILs were observed by a Fourier transform infrared

3.1.1. TGA of [EmimOH][NTf2] IL TGA is very important for estimating the stability of the IL. The NTf2-containing room temperature ionic liquid has a wider liquid

Scheme 1. Predicted mechanism for synthesis of DMC in the presence of IL, DBU, and SmOCl.

A.A. Chaugule et al. / Catalysis Communications 75 (2016) 87–91 Table 2 Effect of catalytic performance on synthesis of DMC using IL + DBU with various metal oxy salts.

1 2 3 4 5 6 7 8

Catalyst system

Methanol conversion [%]

DMC selectivity [%]

Turnover number [TON]

[EmimOH][NTF2]/DBU/SmNO3 [EmimOH][NTF2]/DBU/SmONO [EmimOH][NTF2]/DBU/SmCl3 [EmimOH][NTF2]/DBU/Sm2O3 [EmimOH][NTF2]/DBU/SmOCl [EmimOH][NTF2]/DBU/LaOCl [EmimOH][NTF2]/DBU/FeOCl SmOCl

8.6 11.9 18.1 9.0 13.0 11.4 10.3 1.2

34.1 59.6 40.4 38.0 99.1 88.0 86.1 1.5

2.1 7.4 7.0 1.5 11.7 9.4 8.5 0.4

Reaction conditions: 783 mol of methanol, 6.4 MPa CO2, 4 g IL, 4 h.

range than other organic compounds [13]. Thermal decomposition profiles of the prepared IL are shown in Fig. S 1 a. It was observed that [EmimOH][NTf2] IL is stable up to 350 °C without any significant weight loss. Moreover, DSC curve of the IL was observed in two cycles that shows no crystallization on cooling from 340 to − 30 °C and preserves good stability on heating from − 30 to 340 °C. This might be due to the presence of high molecular weight NTf2 anion conjugate with imidazolium containing ring and strong intermolecular interaction between them. 3.1.2. XRD characterization of the cocatalyst Fig. S1b illustrates the XRD patterns of Sm2O3, SmOCl, and SmCl3, respectively. The XRD pattern of Sm2O3 shows diffraction peaks at 2θ = 19.97, 28.26, 32.76, 47.06, and 55.90 with crystal planes (JCPDS card no. 00–015-0813). When SmCl3 was calcined at 600 °C, its monoclinic structure was completely converted to a tetragonal structure of SmOCl that is confirmed by XRD patterns with peaks at 13.14, 27.95, 31.98, 34.93, 39.42, 41.78, 45.56, 47.94, 52.21, 53.40, and 58.66 with crystal planes (JCPDS card no. 00–012-0790). This, structural change of SmCl3 to SmOCl might change Lewis acidic and basic nature in crystalline structure. Therefore, acidic and basic properties of SmOCl in combination with IL and DBU probably demonstrate activation of MeOH.

89

fact, no reaction occurred without catalyst (Table 1). [EmimOH][NTf2] ILs were found to have high activity with 4.8% yield and 60% selectivity of DMC, whereas the yield of [EmimOH][Cl] IL was low (Table 1), because anion containing fluorine alkyl (NTf2) group possess lower binding energy of anion–CO2 complex than anion containing Cl [15]. The DBU was also found to be inactive, which might be due to loss of activity by the formation amidinium alkyl carbonate (Scheme 1) species with MeOH and CO2 [16] (Table 1). Subsequently, both DBU and [EmimOH][NTf2] IL had a considerable effect on catalytic activity (Table 1). This increase of DMC yield upon addition of DBU and IL could be due to their ability to capture CO2 reversibly. This immobilization of CO2 enhances the reaction by not only converting CO2 in more reactive molecule, but also increasing the concentration of CO2 in the reaction mixture. Hence, [EmimOH][NTf2]/DBU catalytic system was considered for future catalytic studies. Metal oxy salts are associated with Lewis acidic and basic sites. Hence, their use as a cocatalyst can improve methanol conversion without affecting DMC selectivity. In addition, both Lewis basic (Sm 2 O 3 ) + Lewis acidic (SmCl 3 , Sm (NO 3 ) 3 ) and Lewis acidic + Lewis basic (SmOCl, FeOCl, LaOCl) metal salts with IL/DBU have a considerable effect on DMC yield. These results reveal that catalytic efficiency of metal salts decreases in the following order: SmOCl N LaOCl N FeOCl N SmCl 3 N SmONO N Sm 2 O 3 N SmNO 3 (Table 2). This clearly shows that metal salts with both Lewis acidic and Lewis basic sites exhibit superior activity to those with either acidic or basic properties. Moreover, turnover number of the IL + DBU catalytic system was increased from 6.93 to 11.77 with addition of SmOCl (Table 2). On the contrary, only SmOCl was found to be completely inactive for DMC (Table 2). The presence of IL/DBU system has a potential to form the carbonate species with CO2, which can be considered as an activated form of CO2. This is interesting because IL cation and anion moieties stabilized the intermediate and DBU can help absorb the CO2, thereby serving as excellent promoters for this reaction. Moreover, SmOCl mainly activate methanol through coordination, and the generated methoxide anion can favor the attack on CO2-absorb intermediate 2 (Scheme 1) formed by DBU and IL. Hence, the IL/DBU/SmOCl catalytic system has shown potential to obtain high yield of DMC (Table 2).

3.2. Catalytic performance of DMC synthesis The Lewis basic ILs catalyzed the conversion of carbon dioxide to various cyclic carbonates, such as IL derived from 1,4diazabicyclo[2.2.2]octane (DABCO) and DBU [14]. Easy availability, low cost, and air/water stability have attracted increased attention toward this salt. In this context, DMC was synthesized from MeOH and CO2 by IL/DBU/SmOCl catalytic system. For comparison, [EmimOH]+ cation with Cl− and NTf− 2 anion IL base was tested in this study. In

3.2.1. Effect of temperature, pressure, and time on DMC synthesis In order to investigate the effect of temperature on DMC yield, the reaction was carried out at different temperatures, and the results are shown in Fig. 1a. Both DMC production and selectivity increased with an increase of temperature. When the temperature was increased from 80 to 140 °C, the amount of DMC produced increased from 0.09 to 0.11 mol. On the contrary, when the temperature was further increased to 170 °C, the amount of DMC produced and selectivity toward

Fig. 1. Effect of temperature (a), pressure (b), and reaction time (c) on synthesis of DMC using 617 mmol of methanol, 6.4 MPa CO2, 0.5 g SmOCl, 4 g IL, 1 g DBU, 140 °C, and 4 h.

90


Fig. 2. Effect of ionic liquid (IL) loading (a) and SmOCl loading (b) on synthesis of DMC under optimized conditions.

DMC sharply decreased. Moreover, this temperature (170 °C) accelerated other side reactions, such as formation of dimethoxymethane (DMM) and decomposition of DMC to other gaseous by-products. However, 140 °C was determined as the optimal temperature for the reaction, and all further reactions were carried out at this temperature. Fig. 1b shows the effect of pressure on the synthesis of DMC. It is evident from the figure that the production of DMC increases sharply with the increase of pressure up to 6.5 MPa. Further increase of pressure did not have any pronounced effect on DMC production. Hence, all further experiments were conducted at the optimal pressure of 6.5 MPa. Time profile of IL/DBU/SmOCl-catalyzed reaction of DMC is shown in Fig. 1c. It is evident from the figure that DMC yield increases linearly with time and reaches its maximum value after 4 h. Further progress of the reaction decreased DMC yield from 0.11 to 0.09 mol after 8 h [17]. These results indicate that DMC yield rapidly increased with time up to 4 h, and decomposition of DMC to other gaseous products occurred on further progression of the reaction, which might be due to side reactions. 3.2.2. Effect of amount of IL and SmOCl Fig. 2a shows the effect of different amounts of IL on the synthesis of DMC. DMC yield depends on the composition of IL and the presence or absence of cocatalyst. When both cocatalyst and IL are absent, the yield of DMC was zero (Fig. 2). Addition of IL increases both yield and selectivity of DMC. Moreover, addition of 4 g of IL increased DMC yield and selectivity to 0.11 mol and 99.12%, respectively. Subsequently, addition

of 0.5 g of SmOCl resulted in an excellent yield of DMC (Fig. 2b). In conclusion, addition of IL catalyst (4 g) and SmOCl (0.5 g) as cocatalyst exhibited high performance at 140 °C and 6.4 MPa. 3.3. Reaction mechanism In this study, absorption of CO2 can occur by reaction between either DBU, methanol, and CO2 [18] or DBU, IL, and CO2 [19]. Thus, DBU and [EmimOH][NTf2] react first to form zwitterionic intermediate 1 (Scheme 1). This intermediate then reacts with CO2 to form amidinium alkyl carbonate salt 2 [20]. Formation of both 1 and 2 was verified by 13C NMR spectroscopy (Fig. S2). The chemical shift of C7 carbon of pure DBU was observed at 162.5 ppm. After addition of [EmimOH][NTf2] IL in DBU, the chemical shift changes to 163.84 ppm due to development of partial positive charge on C7 carbon, which indicates formation of intermediate 1 (Fig. S2). In addition, when CO2 gas is bubbled through this mixture, a new peak appears at 160 ppm, which could be attributed to carbonate carbon of intermediate 2. SmOCl provides a synergistic combination of Lewis acidic chloride and Lewis basic oxide [20] that can activate methanol. First, methanol coordinates with Lewis acidic samarium and makes hydroxyl hydrogen acidic. Subsequent abstraction of this hydrogen (H) by Lewis basic oxygen SmOCl will generate methoxide ion. Nucleophilic attack of this methoxide on intermediate 2 results in the formation of intermediate 3, which undergoes attack from another methoxide ion to form intermediate 4. Finally, intermediate 4 undergoes an intramolecular decomposition to give products and regenerate the zwitterionic intermediate 1. 3.4. Reusability Finally, methanol and DMC were completely removed under reduced pressure. Then, fresh methanol was added to the remaining mixture of IL, DBU, and SmOCl and reused for the next cycle under the same reaction conditions. The reusability results are shown in Fig. 3, which demonstrates that the activity of catalysts was almost constant even at third cycle. Thus, the catalyst system (IL/DBU/SmOCl) can be used thrice without any predominant change of DMC selectivity. Moreover, after the third cycle, XRD analysis of SmOCl showed that, except (001) and (003), all peaks repeated exactly. Hence, there was no significant change in the crystal plane of SmOCl and, therefore, its stability was satisfied (Fig. S5). 4. Conclusions

Fig. 3. Reusability of catalytic system (IL/DBU/SmOCl) for synthesis of DMC under optimized conditions.

In summary, a new catalyst system for direct DMC synthesis from methanol and CO2 was developed. The ternary catalytic system [EmimOH][NTf2]/DBU/SmOCl, which facilitates the reaction, proceeds


through CO2 absorption by OH-functional group and results in high yield of DMC. SmOCl also has a synergetic effect on catalyzing the direct DMC synthesis reaction. Moreover, catalysts (IL/DBU/SmOCl) can be easily recovered and reused without pronounced loss in catalytic activity. Acknowledgments This study was supported by the National Research Foundation of Korea (NRF) — grants funded by the Ministry of Science, ICT and Future Planning (2014R1A2A2A01004352) and the Ministry of Education (2009-0093816), Republic of Korea. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.catcom.2015.12.009. References [1] Y. Cao, H. Cheng, L. Ma, F. Liu, Z. Liu, Catal. Surv. Jpn. 16 (2012) 134.

[2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20]

91

J.C. Choi, H. Yasuda, T. Sakakura, Chem. Rev. 107 (2007) 2387. F. Crisante, M.C. Ginnasi, R. Bernini, Molecules 16 (2011) 1425. P. Tundo, M. Selva, Acc. Chem. Res. 35 (2002) 716. V.C. Srivastava, I.M. Mishra, P. Kumar, Catal. Commun. 60 (2015) 31. Y. Talyosefa, D. Aurbacha, B. Markovskya, E. Markevicha, E. Zinigrada, L. Asrafa, J.S. Gnanaraja, H.J. Kim, Electrochim. Acta 50 (2004) 254. V. Bevoni, D. Fabbri, M. Notari, F. Rivetti, Fuel 86 (2007) 690. B.A.V. Santos, C.S.M. Pereira, V.M.T.M. Silva, J.M. Loureiro, A.E. Rodrigues, Appl. Catal. A Gen. 455 (2013) 226. S. Bruno, S. Viviana, L. Jose, R. Alirio, ChemBioEng Revs. 1 (2014) 214. D. Ballivet-Tkatchenko, R. Burgat, S. Chambrey, L. Plasseraud, P. Richard, J. Organomet. Chem. 691 (2006) 1498. A.H. Tamboli, A.A. Chaugule, F.A. Sheikh, W.J. Chung, H. Kim, Chin. J. Catal. (2015)http://dx.doi.org/10.1016/S1872-2067(15)60848–8. A.A. Chaugule, A.H. Tamboli, F.A. Sheikh, W.J. Chung, H. Kim, J. Mol. Liq. 208 (2015) 321. N.V. Plechkova, K.R. Seddon, Chem. Soc. Rev. 37 (2008) 150. Z.Z. Yang, L.N. He, C.X. Miao, S. Chanfreau, Adv. Synth. Catal. 352 (2010) 2233. S.Y. Sudha, A. Khanna, Eng. Technol. (2009). C. Wang, S.M. Mahurin, H. Luo, G.A. Baker, H. Li, S. Dai, Green Chem. 12 (2010) 874. C. Jin, Q. Cai, B. Lu, H. Tangbo, Y. Shan, Catal. Lett. 103 (2005) 228. M.C. Ozturk, C.S. Ume, E. Alper, Chem. Eng. Technol. 35 (2012) 2098. J. Sun, B. Lu, X. Wang, X. Li, J. Zhao, Q. Cai, Fuel Process. Technol. 115 (2013) 237. H. Yasuda, L.N. He, T. Sakakura, J. Catal. 209 (2002) 550.