- Email: [email protected]
Synthesis of oligo(carbon dioxide)
Journal of CO₂ Utilization 27 (2018) 42–47
Contents lists available at ScienceDirect
Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou
Synthesis of oligo(carbon dioxide)
T
⁎
Yinghao Fu, Congming Xiao
College of Material Science and Engineering of Huaqiao University, Quanzhou, 362021, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Carbon dioxide Oligomerization Synergetic catalysis Mild conditions
For the first time, we reported the preparation of the oligomer of carbon dioxide. The oligomerization of carbon dioxide was carried out with the aid of the synergetic catalysis of 1,8-diazabicyclo(5, 4, 0)undec-7-ene (DBU) and quaternary salt of water-soluble chitosan (QWSC), which was conducted at 75 °C for 24 h by using petroleum ether as dispersing medium. The product was separated and purified through repeated dissolution-precipitation by using dichloromethane, petroleum ether and distilled water as solvents and precipitant respectively. Fourier transform infrared spectroscopy (FTIR), fluorescence spectroscopy, carbon nuclear magnetic resonance (13CNMR), and MALDI-TOF spectrometry (MS) was applied to characterize the product. The results confirmed that the obtained product was the expected oligomer of CO2, linear H-(O-CO)17-H and cyclic (O-CO)13 containing DBU-like group.
1. Introduction
2. Experimental section
Recently, much attention has been paid to employ carbon dioxide as C1 source to synthesize chemical substances [1–3]. To this end, various catalysts, active reagents and energy-consuming conditions are the common adopted means. So far, most of the reported CO2-based polymers are the copolymers derived from carbon dioxide and epoxide and prepared in the presence of catalysts [4]. In fact, carbon dioxide is an unsaturated compound. It may polymerize under suitable conditions. Iota and his cooperators have reported a quartzlike solid generated from CO2. However, its preparation conditions are as extreme as the pressure and temperature are 40 GPa and 1800 K respectively [5]. Rodig et al. have synthesized a cyclic trimer of CO2 through four-step synthesis, but its starting material is not carbon dioxide at all. Besides, the trimer only exists at the temperature lower than −40 °C [6]. To our best knowledge, no other research about the oligomerization or polymerization of CO2 has been reported until now. CO2 is so inert that it is hard to be polymerized. As well known, suitable catalyst is useful to overcome the barrier. For instance, Tsutsumi et al. have applied a bifunctional catalyst to activate the coupling of carbon dioxide with epoxide. As a result, the reaction can be performed under 1 MPa [7]. Herein, we present a facile approach to carry out the oligomerization of carbon dioxide, which is attributed to taking advantage of a synergetic catalyst.
2.1. Materials
⁎
Water-soluble chitosan was obtained through physically treating chitosan [8]. Quaternary salt of water-soluble chitosan (QWSC) was prepared by reacting 1% WSC aqueous solution with 20 mL 2,3-epoxypropyltrimethylammonium chloride (weight ratio between WSC and salt is 1:3) at 75 °C for 48 h. QWSC was harvested by precipitating the reaction mixture from ethanol, filtered and dried. The quaternary-group amount of QWSC equaled to 2.84 mmol/g, which was determined by conductometric titration. Carbon dioxide gas (99.9%), 1,8-diazabicyclo (5, 4, 0)undec-7-ene (DBU, analytical grade), petroleum ether (60–90 °C, 90–120 °C, analytical grade) and dichloromethane (analytical grade) were all purchased domestically and used as received. 2.2. Synthesis of oligomer of CO2 The oligomer was prepared in the following process. Five millilitres DBU, 0.1 mL water and 50 mL petroleum ether (90–120 °C) was mixed in a three-neck bottle with magnetic stirring. CO2 gas was introduced and the mixture was kept at room temperature for 2 h. Another 1 mL DBU and 0.5 g QWSC was added and the bottle was filled with the gas again. Then, the mixture was kept at 75 °C under stirring for 24 h. The solid phase of the mixture was separated and washed with petroleum ether (60–90 °C) to remove most of DBU. The raw product was dissolved with 10 mL dichloromethane from the solid phase. Most
Corresponding author. E-mail address: [email protected] (C. Xiao).
https://doi.org/10.1016/j.jcou.2018.06.021 Received 24 March 2018; Received in revised form 12 May 2018; Accepted 23 June 2018 2212-9820/ © 2018 Elsevier Ltd. All rights reserved.
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Scheme 1. Designed polymerization route of carbon dioxide.
2.3. Characterizations
of dichloromethane and a little petroleum ether were removed by distilling at 60 °C. The remained petroleum ether was removed and blew with CO2 gas to examine whether it became turbid or not. The raw product was dissolved with 5 mL dichloromethane, precipitated from 20 mL petroleum ether (60–90 °C) and separated out as mentioned above. This purification process was repeated for 5 times. In order to completely purify the product, the obtained raw product was dissolved in dichloromethane and extracted with distilled water. The dichloromethane-phase was subjected to distilling for removal of the solvent, the purified product was dried in vacuum and its yield was 98.47 mg. The control experiment was performed by replacing CO2 gas with N2 under the same conditions.
Fluorescence profiles of the solutions of the product and DBU were recorded using a HITACHI F-7000 FL Spectrophotometer. The excitation wavelength was 300–600 nm and the emission wavelength was between 300 and 800 nm at a scan speed of 1200 nm/min. Fourier transform infrared (FTIR) spectroscopy of the product was recorded using a Nicolet iS50 FTIR spectrometer. MALDI-TOF mass spectrometry of the product was analyzed with a Bruker AutoHex III instrument in reflecting positive ion mode by using alpha-cyano-4-hydroxycinnamic acid (CHCA) as matrix. The outlet and detection voltages were 19 and 1.7 kV respectively, and the wavelength of the nitrogen laser was 43
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Fig. 1. Thermal analysis profiles of the product derived from CO2 (left: DSC, right: TGA).
Fig. 4. UV spectra of upper liquors of the purification process.
Fig. 2. FTIR spectra of the initial and used QWSC.
3. Results and discussion It is reported that the bifunctional catalyst [7] and polysaccharidesbased catalytic systems containing DBU [9,10] are especially efficient for the conversion of carbon dioxide. Therefore, we consider that a catalytic system composed of polysaccharide derivative and DBU may make the polymerization of CO2 possible. QWSC is a quaternary ammonium salt. There are many hydroxyl groups on its backbone. The OH group can be transferred into anionic moiety with the aid of super base DBU, which makes QWSC amphoteric. The amphoteric QWSC acts as a bifunctional catalyst to activate carbon dioxide to become active species containing O−. The active species are anchored on the surface of solid QWSC. Meanwhile, the super base DBU may combine CO2 to form active intermediate bearing C+. The cationic intermediates are attracted toward the anionic ones on the surface of solid QWSC. As a result, the activated CO2 molecules are covalently jointed together (Scheme 1). QWSC and DBU play the role of synergetic catalysis, which enables the polymerization of carbon dioxide can be carried out under mild condition. It is worth mentioning that the preparation conditions are quite mild. The polymerization is conducted under normal atmosphere at 75 °C by utilizing CO2 gas as starting material. Although DBU is possible to react with CO2 gas to form DBUH+HCO3− in the presence of small amount water [11] at ambient temperature, the salt will decompose at the temperature higher than 50 °C [12]. Another possible side reaction is the physical combination of CO2 toward QWSC, which will not generate any by-product. As expected, the polymerization of carbon dioxide is carried out to provide a brown waxy solid. We observe that the obtained compound is able to exist under normal atmosphere pressure and temperature. By contrast, nothing is obtained from the control experiment performed by replacing CO2 gas with N2 under the
Fig. 3. Effect of reuse times on the catalytic efficiency of QWSC.
337 nm. Carbon nuclear magnetic resonance (13CNMR) of the product was measured with a Bruker AVANCE III500 NMR spectrometer by using CDCl3 as solvent. The upper liquors obtained from the purification process were scanned with a UV2450 UV-visible spectrophotometer by using distilled water as solvent. Differential scanning calorimetry (DSC) analysis of the product was carried out with a Netzsch DSC 200 F3 analyzer. The sample was heated from −20 to 150 °C at a rate of 10 °C/min to record DSC curve under N2 atmosphere. Thermogravimetric analyses (TGA) of the product were performed with a SHIMADZU TGA-50H thermoanalyzer, which was conducted over the temperature range from 20 to 900 °C with a programmed temperature increment of 10 °C/min under argon and air atmospheres respectively.
44
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Fig. 5. Fluorescence patterns of the product derived from CO2 (left, concentration: 0.5 mg/mL) and DBU (right, concentration: 5 mg/mL).
Fig. 8. Structure of the product derived from CO2.
point of DBU is as high as 261 °C. The boiling point of petroleum ether applied as reaction medium is 90–120 °C. Evidently, both DBU and petroleum ether will not decompose at 75 °C as well. Secondly, the product is carefully purified. QWSC is easily to be removed by filtering. The solvents can be removed through distilling. DBU is removed via repeated dissolution-precipitation. DBU dissolves in petroleum ether slightly, and the product is insoluble in petroleum ether. Both of them dissolve in dichloromethane. Thus, DBU can be removed through repeated dissolution-precipitation by using dichloromethane as solvent and petroleum ether as precipitant respectively. DBU may become DBU-CO2 adduct or DBUH+HCO3− when it contacts with the gas. The formed adduct or salt is insoluble in petroleum ether. This is taken advantage to judge the degree of purification. After five dissolution-precipitation cycles, the upper liquor keeps clear when CO2 gas is introduced. UV absorption peaks of the corresponding upper liquors exhibit at the same wavelength. Their intensities become weaker and weaker, and the absorption peak disappears at last (Fig. 4). In addition, the characteristic peaks of DBU and the product appear at 420 and 440 nm on their fluorescence profiles respectively. Moreover, the peak of the product is wider than that of DBU, and the intensity of the peak of the product is much greater than that of DBU though the concentration of the product is only one tenth of that of DBU (Fig. 5).
Fig. 6. FTIR spectra of DBU and the product derived from CO2.
same conditions. DSC analysis shows that only one wide and weak peak is appeared around 88.7 °C (Fig. 1, left). This endothermic peak is attributed to the melting point of the solid sample. As the melting peak of a compound of low molecular weight is usually a sharp one, the analysis result suggests the sample is an oligomer. Moreover, the melting point is much higher than the room temperature, which confirms that the product is stable at ambient temperature. TGA profiles of the sample measured under air and argon atmospheres are almost the same. Both of them show three decomposition stages. The initial decomposition temperature of the sample exhibits at 110 °C (Fig. 1, right). The thermal decomposition results indicate what we synthesized is quite stable. The solid is what we want, not something else. Firstly, no decomposition of the agents happens. It is reported that the quaternary chitosan salt does not decompose before 285 °C [13]. FTIR spectra of the reused QWSC samples are almost the same to that of initial QWSC, which demonstrates the chemical structure of QWSC keeps the same during the polymerization (Fig. 2). It is found that QWSC still remains its catalytic function in the presence of DBU after three-time application (Fig. 3). These indicate that the QWSC does not decompose. The boiling
Fig. 7. Full (left) and local (right) mass spectra of the product derived from CO2. 45
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Fig. 9. Carbon NMR profile of the product derived from CO2.
800 are two groups of multiplet (Fig. 7, right). Two strong peaks at m/ z 750.9 and 760.9 indicate that the product contains two components and their m/z are about 751 and 761 respectively. No band appeared at m/z 153, which means no free DBU exists in the product. As mentioned above, we think that the product can be represented as xe (OeCO)ney. Accordingly, the structural formula of the product can be obtained through calculating and rewritten as linear He (OeCO)17-H [(751-1-1-1)/44 = 17] and cyclic a-(OeCO)13 [(761-39-152-0)/ 44≈13], where a is DBU-like moiety (Fig. 8). In other words, the product is an oligomer with OeCO as repeat units. The chemical composition of the repeat unit is the same to carbon dioxide. The reaction enthalpy of the oligomerization of carbon dioxide, n O]C] O → e(OeCO)ne, can be estimated from the enthalpy change between the product and the monomer. Thus, ⊿H=⊿Hp−⊿Hm, where ⊿Hp and ⊿Hm are the enthalpies of formation of the product and CO2 respectively. The enthalpy ⊿Hp = 367 (1 – e−0.5756n) – 721 (kJ/mol) [14], and it equals to −354.21 and −354.02 kJ/mol when n = 13 and 17, respectively. According to literature [15], ⊿Hm= −393.52 kJ/ mol. Then, ⊿H= −354.21 + 393.52 = 39.31 kJ/mol or ⊿H= −354.02 + 393.52 = 39.50 kJ/mol. If the degree of polymerization n is high enough (n→∞), ⊿Hp will be −354.00 kJ/mol and ⊿H= −354.00 + 393.52 = 39.52 kJ/mol. The evaluation of the reaction enthalpy confirms the product is the oligomer of carbon dioxide once more. The structure of the product is further verified with 13C NMR. As shown in Fig. 9, the characteristic chemical shifts corresponding to OeCO are exhibited at 176.53, 176.3, 176.08 and 170.88 ppm on the NMR pattern of the product. The other peaks are attributed to the end group, DBU-like moiety. There are no signs on the omitted part of the NMR pattern. The assignments of the signs are described on Fig. 8. As shown, CO2 molecules have become the oligomer and its repeat unit is OeCO.
These results confirm that no free DBU is remained in the product. In order to remove any water-soluble by-product, the obtained product is dissolved in dichloromethane and extracted with distilled water. The pure product is harvested from the dichloromethane-phase via distilling. Besides the product, nothing else is remained after such these purification steps. In other words, the product is indeed a compound derived from CO2. As shown in Fig. 5, its solution exhibits fluorescent characteristic. The fluorescence absorption of the product is attributed to the p-π conjugation of repeated oxy-carbonyl (eOeCOe) units. In order to acquire the structural information of the pure product, FTIR, CNMR and MS are employed. There are many adsorption peaks on the FTIR spectra of the product. The peaks that appeared at 2931, 2857 and those between 1500–640 cm−1 are quite like the ones of DBU. But the spectra of the product are indeed different from that of DBU (Fig. 6). Two characteristic bands that exhibited around 3335 cm−1 and −1 1737–1618 cm provide us useful information. The wide peak around 3335 cm−1 suggests that the product contains carboxyl, hydroxyl and/ or NeH groups. It is obvious that the characteristic peak appeared around 1618 cm−1 is much wider than the one at 1608 cm−1 on the spectrum of DBU, which is attributed to the repeated eOeCOe moieties since the p-π conjugations between the pi electrons of C]O group and the nonbonding electrons of oxygen (COeO:) can lower the stretching frequency of carbonyl group. In addition, the one shown at 1147cm−1 belongs to CeO bonds, while that at 1557 cm−1 may be assigned to eCOeNe group. These analyses suggest that the product is composed of eOeCOe units, COOH group and DBU-like moiety. The molecular weight of the product is determined with mass spectroscopy. There are three peaks on the spectra (Fig. 7, left). Actually, the peak appears at m/z 39.22 is quite weak, and it may be assigned to potassium ion. The much stronger signals closing to m/z 46
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
4. Conclusions
B. Rieger, Chem. Eur. J. 17 (2011) 8858–8869. [2] M. He, Y. Sun, B. Han, Angew. Chem. Int. Ed. 52 (2013) 9620–9633. [3] E. Vessally, M. Babazadeh, A. Hosseinian, S. Arshadi, L. Edjlali, J. CO2 Util. 21 (2017) 491–502. [4] S.J. Poland, D.J. Darensbourg, Green Chem. 19 (2017) 4990–5011. [5] V. Iota, C.S. Yoo, H. Cynn, Science 283 (1999) 1510–1513. [6] M.J. Rodig, A.W. Snow, P. Scholl, S. Rea, J. Org. Chem. 81 (2016) 5354–5361. [7] Y. Tsutsumi, K. Yamakawa, M. Yoshida, T. Ema, T. Sakai, Org. Lett. 12 (2010) 5728–5731. [8] Y.H. Fu, C.M. Xiao, Int. J. Biol. Macromol. 103 (2017) 575–580. [9] J. Sun, W.G. Cheng, Z.F. Yang, J.Q. Wang, T.T. Xu, J.Y. Xin, S.J. Zhang, Green Chem. 16 (2014) 3071–3078. [10] A.H. Tamboli, A.A. Chaugule, H. Kim, Fuel 184 (2016) 233–241. [11] D.J. Heldebrant, P.G. Jessop, C.A. Thomas, C.A. Eckert, C.L. Liotta, J. Org. Chem. 70 (2005) 5335–5338. [12] T. Mizuno, T. Nakai, M. Mihara, Heteroat. Chem. 23 (2012) 276–280. [13] V.A. Spinelli, M.C.M. Laranjeira, V.T. Fávere, React. Funct. Polym. 61 (2004) 347–352. [14] E. Lewars, J. Mol. Struct. (Theochem) 363 (1996) l–15. [15] B.I. Dunlap, I.V. Schweigert, A.P. Purdy, A.W. Snow, A. Hu, J. Chem. Phys. 138 (2013) 134304.
In this article, we present the preparation of oligo(carbon dioxide) by using CO2 gas as the monomer. High temperature and pressure are unnecessary for the synthesis process. The approach is facile and feasible. The obtained oligomer is stable at ambient temperature. It contains two components, and their structures have been confirmed. We hope what we presented here is a good beginning for the conversion of carbon dioxide, and more great progresses to be made thereafter. Acknowledgement The authors thank the Natural Science Foundation of Fujian Province of China for funding (grant no. 2015J01201). References [1] M.W. Lehenmeier, C. Bruckmeier, S. Klaus, J.E. Dengler, P. Deglmann, A.-K. Ott,
47
Contents lists available at ScienceDirect
Journal of CO2 Utilization journal homepage: www.elsevier.com/locate/jcou
Synthesis of oligo(carbon dioxide)
T
⁎
Yinghao Fu, Congming Xiao
College of Material Science and Engineering of Huaqiao University, Quanzhou, 362021, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Carbon dioxide Oligomerization Synergetic catalysis Mild conditions
For the first time, we reported the preparation of the oligomer of carbon dioxide. The oligomerization of carbon dioxide was carried out with the aid of the synergetic catalysis of 1,8-diazabicyclo(5, 4, 0)undec-7-ene (DBU) and quaternary salt of water-soluble chitosan (QWSC), which was conducted at 75 °C for 24 h by using petroleum ether as dispersing medium. The product was separated and purified through repeated dissolution-precipitation by using dichloromethane, petroleum ether and distilled water as solvents and precipitant respectively. Fourier transform infrared spectroscopy (FTIR), fluorescence spectroscopy, carbon nuclear magnetic resonance (13CNMR), and MALDI-TOF spectrometry (MS) was applied to characterize the product. The results confirmed that the obtained product was the expected oligomer of CO2, linear H-(O-CO)17-H and cyclic (O-CO)13 containing DBU-like group.
1. Introduction
2. Experimental section
Recently, much attention has been paid to employ carbon dioxide as C1 source to synthesize chemical substances [1–3]. To this end, various catalysts, active reagents and energy-consuming conditions are the common adopted means. So far, most of the reported CO2-based polymers are the copolymers derived from carbon dioxide and epoxide and prepared in the presence of catalysts [4]. In fact, carbon dioxide is an unsaturated compound. It may polymerize under suitable conditions. Iota and his cooperators have reported a quartzlike solid generated from CO2. However, its preparation conditions are as extreme as the pressure and temperature are 40 GPa and 1800 K respectively [5]. Rodig et al. have synthesized a cyclic trimer of CO2 through four-step synthesis, but its starting material is not carbon dioxide at all. Besides, the trimer only exists at the temperature lower than −40 °C [6]. To our best knowledge, no other research about the oligomerization or polymerization of CO2 has been reported until now. CO2 is so inert that it is hard to be polymerized. As well known, suitable catalyst is useful to overcome the barrier. For instance, Tsutsumi et al. have applied a bifunctional catalyst to activate the coupling of carbon dioxide with epoxide. As a result, the reaction can be performed under 1 MPa [7]. Herein, we present a facile approach to carry out the oligomerization of carbon dioxide, which is attributed to taking advantage of a synergetic catalyst.
2.1. Materials
⁎
Water-soluble chitosan was obtained through physically treating chitosan [8]. Quaternary salt of water-soluble chitosan (QWSC) was prepared by reacting 1% WSC aqueous solution with 20 mL 2,3-epoxypropyltrimethylammonium chloride (weight ratio between WSC and salt is 1:3) at 75 °C for 48 h. QWSC was harvested by precipitating the reaction mixture from ethanol, filtered and dried. The quaternary-group amount of QWSC equaled to 2.84 mmol/g, which was determined by conductometric titration. Carbon dioxide gas (99.9%), 1,8-diazabicyclo (5, 4, 0)undec-7-ene (DBU, analytical grade), petroleum ether (60–90 °C, 90–120 °C, analytical grade) and dichloromethane (analytical grade) were all purchased domestically and used as received. 2.2. Synthesis of oligomer of CO2 The oligomer was prepared in the following process. Five millilitres DBU, 0.1 mL water and 50 mL petroleum ether (90–120 °C) was mixed in a three-neck bottle with magnetic stirring. CO2 gas was introduced and the mixture was kept at room temperature for 2 h. Another 1 mL DBU and 0.5 g QWSC was added and the bottle was filled with the gas again. Then, the mixture was kept at 75 °C under stirring for 24 h. The solid phase of the mixture was separated and washed with petroleum ether (60–90 °C) to remove most of DBU. The raw product was dissolved with 10 mL dichloromethane from the solid phase. Most
Corresponding author. E-mail address: [email protected] (C. Xiao).
https://doi.org/10.1016/j.jcou.2018.06.021 Received 24 March 2018; Received in revised form 12 May 2018; Accepted 23 June 2018 2212-9820/ © 2018 Elsevier Ltd. All rights reserved.
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Scheme 1. Designed polymerization route of carbon dioxide.
2.3. Characterizations
of dichloromethane and a little petroleum ether were removed by distilling at 60 °C. The remained petroleum ether was removed and blew with CO2 gas to examine whether it became turbid or not. The raw product was dissolved with 5 mL dichloromethane, precipitated from 20 mL petroleum ether (60–90 °C) and separated out as mentioned above. This purification process was repeated for 5 times. In order to completely purify the product, the obtained raw product was dissolved in dichloromethane and extracted with distilled water. The dichloromethane-phase was subjected to distilling for removal of the solvent, the purified product was dried in vacuum and its yield was 98.47 mg. The control experiment was performed by replacing CO2 gas with N2 under the same conditions.
Fluorescence profiles of the solutions of the product and DBU were recorded using a HITACHI F-7000 FL Spectrophotometer. The excitation wavelength was 300–600 nm and the emission wavelength was between 300 and 800 nm at a scan speed of 1200 nm/min. Fourier transform infrared (FTIR) spectroscopy of the product was recorded using a Nicolet iS50 FTIR spectrometer. MALDI-TOF mass spectrometry of the product was analyzed with a Bruker AutoHex III instrument in reflecting positive ion mode by using alpha-cyano-4-hydroxycinnamic acid (CHCA) as matrix. The outlet and detection voltages were 19 and 1.7 kV respectively, and the wavelength of the nitrogen laser was 43
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Fig. 1. Thermal analysis profiles of the product derived from CO2 (left: DSC, right: TGA).
Fig. 4. UV spectra of upper liquors of the purification process.
Fig. 2. FTIR spectra of the initial and used QWSC.
3. Results and discussion It is reported that the bifunctional catalyst [7] and polysaccharidesbased catalytic systems containing DBU [9,10] are especially efficient for the conversion of carbon dioxide. Therefore, we consider that a catalytic system composed of polysaccharide derivative and DBU may make the polymerization of CO2 possible. QWSC is a quaternary ammonium salt. There are many hydroxyl groups on its backbone. The OH group can be transferred into anionic moiety with the aid of super base DBU, which makes QWSC amphoteric. The amphoteric QWSC acts as a bifunctional catalyst to activate carbon dioxide to become active species containing O−. The active species are anchored on the surface of solid QWSC. Meanwhile, the super base DBU may combine CO2 to form active intermediate bearing C+. The cationic intermediates are attracted toward the anionic ones on the surface of solid QWSC. As a result, the activated CO2 molecules are covalently jointed together (Scheme 1). QWSC and DBU play the role of synergetic catalysis, which enables the polymerization of carbon dioxide can be carried out under mild condition. It is worth mentioning that the preparation conditions are quite mild. The polymerization is conducted under normal atmosphere at 75 °C by utilizing CO2 gas as starting material. Although DBU is possible to react with CO2 gas to form DBUH+HCO3− in the presence of small amount water [11] at ambient temperature, the salt will decompose at the temperature higher than 50 °C [12]. Another possible side reaction is the physical combination of CO2 toward QWSC, which will not generate any by-product. As expected, the polymerization of carbon dioxide is carried out to provide a brown waxy solid. We observe that the obtained compound is able to exist under normal atmosphere pressure and temperature. By contrast, nothing is obtained from the control experiment performed by replacing CO2 gas with N2 under the
Fig. 3. Effect of reuse times on the catalytic efficiency of QWSC.
337 nm. Carbon nuclear magnetic resonance (13CNMR) of the product was measured with a Bruker AVANCE III500 NMR spectrometer by using CDCl3 as solvent. The upper liquors obtained from the purification process were scanned with a UV2450 UV-visible spectrophotometer by using distilled water as solvent. Differential scanning calorimetry (DSC) analysis of the product was carried out with a Netzsch DSC 200 F3 analyzer. The sample was heated from −20 to 150 °C at a rate of 10 °C/min to record DSC curve under N2 atmosphere. Thermogravimetric analyses (TGA) of the product were performed with a SHIMADZU TGA-50H thermoanalyzer, which was conducted over the temperature range from 20 to 900 °C with a programmed temperature increment of 10 °C/min under argon and air atmospheres respectively.
44
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Fig. 5. Fluorescence patterns of the product derived from CO2 (left, concentration: 0.5 mg/mL) and DBU (right, concentration: 5 mg/mL).
Fig. 8. Structure of the product derived from CO2.
point of DBU is as high as 261 °C. The boiling point of petroleum ether applied as reaction medium is 90–120 °C. Evidently, both DBU and petroleum ether will not decompose at 75 °C as well. Secondly, the product is carefully purified. QWSC is easily to be removed by filtering. The solvents can be removed through distilling. DBU is removed via repeated dissolution-precipitation. DBU dissolves in petroleum ether slightly, and the product is insoluble in petroleum ether. Both of them dissolve in dichloromethane. Thus, DBU can be removed through repeated dissolution-precipitation by using dichloromethane as solvent and petroleum ether as precipitant respectively. DBU may become DBU-CO2 adduct or DBUH+HCO3− when it contacts with the gas. The formed adduct or salt is insoluble in petroleum ether. This is taken advantage to judge the degree of purification. After five dissolution-precipitation cycles, the upper liquor keeps clear when CO2 gas is introduced. UV absorption peaks of the corresponding upper liquors exhibit at the same wavelength. Their intensities become weaker and weaker, and the absorption peak disappears at last (Fig. 4). In addition, the characteristic peaks of DBU and the product appear at 420 and 440 nm on their fluorescence profiles respectively. Moreover, the peak of the product is wider than that of DBU, and the intensity of the peak of the product is much greater than that of DBU though the concentration of the product is only one tenth of that of DBU (Fig. 5).
Fig. 6. FTIR spectra of DBU and the product derived from CO2.
same conditions. DSC analysis shows that only one wide and weak peak is appeared around 88.7 °C (Fig. 1, left). This endothermic peak is attributed to the melting point of the solid sample. As the melting peak of a compound of low molecular weight is usually a sharp one, the analysis result suggests the sample is an oligomer. Moreover, the melting point is much higher than the room temperature, which confirms that the product is stable at ambient temperature. TGA profiles of the sample measured under air and argon atmospheres are almost the same. Both of them show three decomposition stages. The initial decomposition temperature of the sample exhibits at 110 °C (Fig. 1, right). The thermal decomposition results indicate what we synthesized is quite stable. The solid is what we want, not something else. Firstly, no decomposition of the agents happens. It is reported that the quaternary chitosan salt does not decompose before 285 °C [13]. FTIR spectra of the reused QWSC samples are almost the same to that of initial QWSC, which demonstrates the chemical structure of QWSC keeps the same during the polymerization (Fig. 2). It is found that QWSC still remains its catalytic function in the presence of DBU after three-time application (Fig. 3). These indicate that the QWSC does not decompose. The boiling
Fig. 7. Full (left) and local (right) mass spectra of the product derived from CO2. 45
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
Fig. 9. Carbon NMR profile of the product derived from CO2.
800 are two groups of multiplet (Fig. 7, right). Two strong peaks at m/ z 750.9 and 760.9 indicate that the product contains two components and their m/z are about 751 and 761 respectively. No band appeared at m/z 153, which means no free DBU exists in the product. As mentioned above, we think that the product can be represented as xe (OeCO)ney. Accordingly, the structural formula of the product can be obtained through calculating and rewritten as linear He (OeCO)17-H [(751-1-1-1)/44 = 17] and cyclic a-(OeCO)13 [(761-39-152-0)/ 44≈13], where a is DBU-like moiety (Fig. 8). In other words, the product is an oligomer with OeCO as repeat units. The chemical composition of the repeat unit is the same to carbon dioxide. The reaction enthalpy of the oligomerization of carbon dioxide, n O]C] O → e(OeCO)ne, can be estimated from the enthalpy change between the product and the monomer. Thus, ⊿H=⊿Hp−⊿Hm, where ⊿Hp and ⊿Hm are the enthalpies of formation of the product and CO2 respectively. The enthalpy ⊿Hp = 367 (1 – e−0.5756n) – 721 (kJ/mol) [14], and it equals to −354.21 and −354.02 kJ/mol when n = 13 and 17, respectively. According to literature [15], ⊿Hm= −393.52 kJ/ mol. Then, ⊿H= −354.21 + 393.52 = 39.31 kJ/mol or ⊿H= −354.02 + 393.52 = 39.50 kJ/mol. If the degree of polymerization n is high enough (n→∞), ⊿Hp will be −354.00 kJ/mol and ⊿H= −354.00 + 393.52 = 39.52 kJ/mol. The evaluation of the reaction enthalpy confirms the product is the oligomer of carbon dioxide once more. The structure of the product is further verified with 13C NMR. As shown in Fig. 9, the characteristic chemical shifts corresponding to OeCO are exhibited at 176.53, 176.3, 176.08 and 170.88 ppm on the NMR pattern of the product. The other peaks are attributed to the end group, DBU-like moiety. There are no signs on the omitted part of the NMR pattern. The assignments of the signs are described on Fig. 8. As shown, CO2 molecules have become the oligomer and its repeat unit is OeCO.
These results confirm that no free DBU is remained in the product. In order to remove any water-soluble by-product, the obtained product is dissolved in dichloromethane and extracted with distilled water. The pure product is harvested from the dichloromethane-phase via distilling. Besides the product, nothing else is remained after such these purification steps. In other words, the product is indeed a compound derived from CO2. As shown in Fig. 5, its solution exhibits fluorescent characteristic. The fluorescence absorption of the product is attributed to the p-π conjugation of repeated oxy-carbonyl (eOeCOe) units. In order to acquire the structural information of the pure product, FTIR, CNMR and MS are employed. There are many adsorption peaks on the FTIR spectra of the product. The peaks that appeared at 2931, 2857 and those between 1500–640 cm−1 are quite like the ones of DBU. But the spectra of the product are indeed different from that of DBU (Fig. 6). Two characteristic bands that exhibited around 3335 cm−1 and −1 1737–1618 cm provide us useful information. The wide peak around 3335 cm−1 suggests that the product contains carboxyl, hydroxyl and/ or NeH groups. It is obvious that the characteristic peak appeared around 1618 cm−1 is much wider than the one at 1608 cm−1 on the spectrum of DBU, which is attributed to the repeated eOeCOe moieties since the p-π conjugations between the pi electrons of C]O group and the nonbonding electrons of oxygen (COeO:) can lower the stretching frequency of carbonyl group. In addition, the one shown at 1147cm−1 belongs to CeO bonds, while that at 1557 cm−1 may be assigned to eCOeNe group. These analyses suggest that the product is composed of eOeCOe units, COOH group and DBU-like moiety. The molecular weight of the product is determined with mass spectroscopy. There are three peaks on the spectra (Fig. 7, left). Actually, the peak appears at m/z 39.22 is quite weak, and it may be assigned to potassium ion. The much stronger signals closing to m/z 46
Journal of CO₂ Utilization 27 (2018) 42–47
Y. Fu, C. Xiao
4. Conclusions
B. Rieger, Chem. Eur. J. 17 (2011) 8858–8869. [2] M. He, Y. Sun, B. Han, Angew. Chem. Int. Ed. 52 (2013) 9620–9633. [3] E. Vessally, M. Babazadeh, A. Hosseinian, S. Arshadi, L. Edjlali, J. CO2 Util. 21 (2017) 491–502. [4] S.J. Poland, D.J. Darensbourg, Green Chem. 19 (2017) 4990–5011. [5] V. Iota, C.S. Yoo, H. Cynn, Science 283 (1999) 1510–1513. [6] M.J. Rodig, A.W. Snow, P. Scholl, S. Rea, J. Org. Chem. 81 (2016) 5354–5361. [7] Y. Tsutsumi, K. Yamakawa, M. Yoshida, T. Ema, T. Sakai, Org. Lett. 12 (2010) 5728–5731. [8] Y.H. Fu, C.M. Xiao, Int. J. Biol. Macromol. 103 (2017) 575–580. [9] J. Sun, W.G. Cheng, Z.F. Yang, J.Q. Wang, T.T. Xu, J.Y. Xin, S.J. Zhang, Green Chem. 16 (2014) 3071–3078. [10] A.H. Tamboli, A.A. Chaugule, H. Kim, Fuel 184 (2016) 233–241. [11] D.J. Heldebrant, P.G. Jessop, C.A. Thomas, C.A. Eckert, C.L. Liotta, J. Org. Chem. 70 (2005) 5335–5338. [12] T. Mizuno, T. Nakai, M. Mihara, Heteroat. Chem. 23 (2012) 276–280. [13] V.A. Spinelli, M.C.M. Laranjeira, V.T. Fávere, React. Funct. Polym. 61 (2004) 347–352. [14] E. Lewars, J. Mol. Struct. (Theochem) 363 (1996) l–15. [15] B.I. Dunlap, I.V. Schweigert, A.P. Purdy, A.W. Snow, A. Hu, J. Chem. Phys. 138 (2013) 134304.
In this article, we present the preparation of oligo(carbon dioxide) by using CO2 gas as the monomer. High temperature and pressure are unnecessary for the synthesis process. The approach is facile and feasible. The obtained oligomer is stable at ambient temperature. It contains two components, and their structures have been confirmed. We hope what we presented here is a good beginning for the conversion of carbon dioxide, and more great progresses to be made thereafter. Acknowledgement The authors thank the Natural Science Foundation of Fujian Province of China for funding (grant no. 2015J01201). References [1] M.W. Lehenmeier, C. Bruckmeier, S. Klaus, J.E. Dengler, P. Deglmann, A.-K. Ott,
47