DBU-catalyzed CO₂ fixation in polypropargylamines under solvent-free conditions

Introduction

CO₂ is a useful and attractive C1 resource in polymer chemistry, given its direct applicability to polymer synthesis and modification. The use of CO₂ as a sustainable precursor of valuable polymers has aroused much interest, from both environmental and industrial perspectives. Thus far, research efforts have focused on the direct (co)polymerization of CO₂, the polymerization of monomers synthesized from CO₂, and the incorporation of gaseous CO₂ by postmodification approaches [1,2,3,4]. Among the most environmentally benign attempts is metal- and solvent-free CO₂ fixation in polymeric substrates under moderate conditions. However, this approach is hindered by the relatively low reactivity of CO₂ and the difficulty in performing polymer reactions in solvent-free conditions. Endo and coworkers reported an efficient base-catalyzed gas–solid phase reaction that incorporates CO₂ into oxirane-derived polymers [5,6,7,8], but current reports on this topic are limited. We have reported highly efficient CO₂ fixation in polypropargylamines with a 2-pyridyl group using 1,8-diazabicyclo [5.4.0]undec-7-ene (DBU) as a base catalyst [9]. Within a few minutes, the reaction yielded the corresponding polyoxazolidinone under atmospheric CO₂ and room temperature conditions. Owing to the highly reactive substrate, the reaction occurred even under conditions with very low CO₂ concentrations in air. Herein, we report the efficient CO₂ fixation in polypropargylamines under solvent-free conditions. Two types of polymers with a propargylamine-repeating unit are synthesized: one with a main chain of repeating units and the other with side chains of repeating units. The side chain-type polypropargylamines are newly designed for the sake of simplification of copolymerization with other monomers in order to incorporate functional groups. Both types of polymers could be used for solvent-free CO₂ fixation. For side chain-type polymers, the effect of the incorporation of a tertiary amine moiety into the polymer chain was evaluated.

Experimental procedure

Materials and measurements

Commercially available solvents and reagents were used as received. (6-Bromo-pyridin-2-yl)methanol and N-benzyl-1,1-dimethylpropargylamine were synthesized as described in the literature [9]. ¹H and ¹³C NMR spectra were recorded by a Bruker Biospin AVANCE DPX-300 spectrometer and a Bruker AVANCEIIIHD500 using deuterated solvents. Spectra were calibrated using residual undeuterated solvent, tetramethylsilane as the internal standard. The IR spectra were recorded on a JASCO FT/IR-230 spectrometer. High-resolution mass spectrometry–electrospray ionization (HRMS–ESI) data were collected using a JEOL JMS-700 mass spectrometer at the Center for Advanced Materials Analysis, Tokyo Institute of Technology and are available on request. Size exclusion chromatography (SEC) was carried out at 30 °C in dimethylformamide (DMF) (5 mM LiBr, 0.85 mL/min) using a JASCO PU-2080 system equipped with a set of TOSOH TSKgel G2500H and G4000H columns. The number average molecular weight (M_n), weight average molecular weight (M_w), and polydispersity index (PDI) of the obtained polymers were calculated based on a polystyrene calibration. The 5% and 10% weight decomposition temperatures T_d5 andT_d10, at which the polymers lose 5% and 10% of their weight, respectively, were determined by thermogravimetric analysis (TGA). The TGA was carried out on a Shimadzu TGA-50 instrument under N₂ atmosphere (flow rate 50 mL/min) at a heating rate of 10 °C/min. The glass transition temperatures (T_gs) of the polymers were determined by differential scanning calorimetry (DSC) using a Shimadzu DSC-60 instrument in N₂ atmosphere (flow rate 50 mL/min) with liquid N₂ as a refrigerant. The DSC heating rate was 10 °C/min.

Synthesis of monomer 2

A solution of methacryloyl chloride (1.33 g, 12.7 mmol) in CH₂Cl₂ (5.0 mL) was slowly added to a solution of (6-bromo-pyridin-2-yl)methanol (2.00 g, 10.6 mmol) and triethylamine (1.80 mL, 12.7 mmol) in CH₂Cl₂ (20 mL) at 0 °C and stirred at room temperature for 18 h. The resulting mixture was filtered to remove the insoluble part, and the filtrate was washed with 5% NaOH aq., sat. NH₄Cl aq., and water. The organic layer was dried over anhydrous MgSO₄, and filtered and evaporated at reduced pressure. The residue was purified by silica gel column chromatography using n-hexane/ethyl acetate = 20/1 (v/v) as the eluent to give A (2.19 g, 80%). M.p. 32.4–34.0 °C; ¹H NMR (500 MHz, CDCl₃) δ 7.57 (t, J = 7.8 Hz, 1 H), 7.43 (dd, J = 7.9, 0.6 Hz, 1 H), 7.33 (dd, J = 7.5, 0.8 Hz, 1 H), 6.23 (q, J = 1.1 Hz, 1 H), 5.66 (q, J = 1.6 H Hz, 1 H), 5.28 (s, 2 H), 2.00 (q, J = 0.8 Hz, 3 H) ppm; ¹³C NMR (126 MHz, CDCl₃) δ 166.69, 157.50, 141.55, 139.03, 135.75, 127.16, 126.45, 120.11, 66.05, 18.32 ppm; HRMS-ESI (m/z) [M HNa]⁺ calculated for [C₁₀H₁₀BrNO₂Na]⁺ 277.9787; found: 277.9781; IR (neat) 2921 (C-H), 1719 (C 77C), 1156 (C-N) cm^–1.

N-Benzyl-1,1-dimethylpropargylamine (0.31 g, 1.76 mmol), Pd(PPh₃)Cl₂ (11 mg, 0.016 mmol), CuI (6 mg, 0.03 mmol), and Et₃N (890 μL, 6.4 mmol) were added to a THF solution of A (0.21 g 0.80 mmol) under Ar atmosphere. The mixture was stirred at 50 °C for 2 h and diluted with CHCl₃. The organic layer was washed with NH₄Cl aq. and NaHCO₃ aq., dried over anhydrous Na₂SO₄, and filtered and evaporated under reduced pressure. The residue was purified by silica gel column chromatography eluted with EtOAc/Hex = 1/4 (v/v) to give 2 as a yellow solid (0.27 g, 98%). m.p. 33.7-35.7 °C; ¹H NMR (500 MHz, CDCl₃) δ 7.67 (t, J = 7.8 Hz, 1 H), 7.38 (t, J = 6.6 Hz, 3 H), 7.32 (q, J = 7.4 Hz, 3 H), 6.23 (t, J = 1.2 Hz, 1 H), 5.65 (t, J = 1.5 Hz, 1 H), 5.33 (s, 2 H), 3.97 (s, 2 H), 2.01 (q, J = 0.8 Hz, 3 H), 1.54 (s, 6 H) ppm; ¹³C NMR (126 MHz, CDCl₃) δ 166.84, 156.51, 143.05, 140.53, 136.82, 135.91, 128.42, 128.36, 126.93, 126.38, 126.25, 120.20, 95.11, 81.77, 66.79, 50.57, 49.11, 29.39, 18.35 ppm; HRMS-ESI (m/z) [M]⁺ calculated for [C₂₂H₂₅N₂O₂]⁺ 349.1911; found: 349.1905; IR (neat) 3282 (N-H), 2978 (C-H), 2208 (C≡C), 1719 (C ),C), 1156 (C-N) cm^–1.

Radical polymerization of monomer 2

Azobisisobutyronitrile (AIBN; 4.5 mg, 4 mol%) was added to a solution of 2 (238 mg, 0.68 mmol) in toluene (850 μL, 0.8 M) under an Ar atmosphere. The reaction mixture was heated at 70 °C for 6 h and then cooled to room temperature and concentrated under reduced pressure. The residue was redissolved in CHCl₃ (1 mL). The precipitate was collected by pouring CHCl₃ (1 mL) solution into n-hexane (30 mL) and was dried under reduced pressure to give P2AU as a white solid (0.22 g, 90%). ¹H NMR and FT-IR data are described in the ESI.

Radical copolymerization of monomer 2 and 2-(N,N- diethylamino)ethyl methacrylate (DEAEMA)

Compound 2 and DEAEMA were fused by radical copolymerization, varying the feed ratio of the two monomers in a manner similar to the process for the polymerization of 2. The products were P3AU, P4AU, and P5AU. The ¹H NMR and FT-IR data of P3AU, P4AU, and P5AU are described in the ESI.

Solvent-free CO₂ fixation in P1AU

To prepare the polymer film, P1AU solution in CHCl₃ and DBU (200 mol% per repeating unit) was placed in Teflon® Petri dishes (80 mm diameter). The solvent was allowed to evaporate under Ar conditions for 3 days and under vacuum conditions for 1 day. The prepared film was placed into a glass vial. After purging with CO₂, the glass vial was incubated at 60 °C, and the reaction was monitored by ¹H NMR spectroscopy. Once the reaction had finished, the reaction mixture was poured into 5% H₂O/MeOH (v/v). The resulting precipitates were collected by filtration and dissolved in CHCl₃ (1 mL), and the solution was again poured into 5% H₂O/MeOH (v/v). The resulting precipitates were collected by filtration, giving P1OZ at 86% yield.

¹H-NMR (500 MHz, DMSO-d₆) δ 7.78 (br, 1 H), 7.68 (br, 1 H), 7.29–7.21 (m, 2 H), 7.03–6.90 (br, 3 H), 6.28–5.92 (br, 1 H), 5.13 (br, 2 H), 4.42 (br, 2 H), 1.60–1.13 (m, 6 H) ppm; IR (neat) 2971 (C–H), 1780 (C = O), 1585 (aromatic C = C), 1683 (C = C) cm^–1.

Solvent-free CO₂ fixation in P2AU–P5AU

The general method was as follows. A polymer film was prepared by charging a solution of P2AU (40 mg) in CHCl₃ and DBU (20 mol% per repeating unit) in a round-bottomed flask and allowing the solvent to evaporate in vacuo. After purging with CO₂, the flask was heated at 60 °C, and the reaction was monitored by ¹H NMR spectroscopy. Once the reaction had finished, the reaction mixture was dissolved in CHCl₃ (1 mL) and washed with aq. NaHCO₃ (1 mL x 3) and brine (1 mL). The organic layer was dried over Na₂CO₃ and concentrated under reduced pressure to give P2OZ (34 mg).

The characterization of the formed polymers (P2OZ–P5OZ) is described in the ESI.

Results and discussion

Synthesis and characterization of the polymers

The main-chain type polypropargylamine P1AU was synthesized as described in the literature [9], namely, by the Sonogashira–Hagihara coupling of the AB-type monomer (1). The corresponding polymer was obtained at 71% yield (scheme S1). From the SEC result, the number average molecular weight (M_n) and PDI of P1AU were estimated to be 38,600 and 2.21, respectively. The vinyl polymer P2AU was synthesized via the typical free-radical polymerization of the vinyl monomer with the propargylamine moiety (2). The reaction was initiated by AIBN in toluene (Table 1). The preparation of 2 is described in Scheme 1. P2AU with a moderate M_n and PDI (15,500 and 2.71, respectively) was isolated by reprecipitation in n-hexane. All P2AU signals in ¹H NMR were assignable, indicating that the radical polymerization proceeded without damaging the propargylamine scaffold. The random copolymers P3AU–P5AU were prepared by varying the feed ratio of 2 and DEAEMA. Given the high CO₂-capturing ability of its tertiary amine moiety [10, 11], the DEAEMA comonomer was expected to accelerate the rate of CO₂ fixation to the propargylamine moiety. The composition ratios of the comonomer in P3AU to P5AU almost coincided with the feed ratios based on the integral ratios in the ¹H NMR spectra, indicating similar polymerizability of these two monomers. The M_n and thermal properties of P1AU–P5AU are summarized in Table 2. Unexpectedly, the DSC measurements yielded almost identical T_g values for P1AU and P2AU. Each copolymer in the P3AU–P5AU series exhibited a single T_g peak, which shifted to the lower temperature side (33 → 24 → 0 °C) as the DEAEMA content increased. The T_g of a DEAEMA homopolymer prepared by the same procedure (M_n 10,900, PDI 1.98) appeared at –22 °C (see ESI), so these results are entirely reasonable. No obvious melting peak was observed over the measurement range in all cases.

Table 1 Synthesis of P2AU–P5AU

Scheme 1

Synthesis of monomer 2

Table 2 M_n and thermal properties of polymers

Solvent-free CO₂ fixation in P1AU

A P1AU film containing DBU (200 mol%) was prepared by the slow evaporation of CHCl₃ solution under a 0.1 MPa CO₂ atmosphere. At room temperature, which was tried initially, the reaction failed to complete even after prolonged reaction times (25% conversion in 288 h), in spite of the results in DMSO solution (100% conversion in 5 min) [9]. After raising the reaction temperature to 60 °C, which is slightly higher than the T_g of P1AU, the conversion rate was 90% after 120 h (Fig. 1). The decreased benzyl peaks of P1AU (b and c) and the appearance of double-bond peaks at ~5.8–6.2 ppm in the ¹H NMR spectra strongly suggest the formation of the oxazolidinone moiety P1OZ. The ratio of Z- to E-isomers (89:11) almost coincided with that of the solution reaction in DMSO.

Fig. 1

Partial ¹H NMR of P1AU (a) before and (b) after the solvent-free CO₂-fixation reaction at 60 °C for 120 h (500 MHz, 298 K, CDCl₃)

Solvent-free CO₂ fixation in P2AU–P5AU

Each polymer film of the side-chain-type polymers P2AU–P5AU was prepared by removing CHCl₃ from a solution of mixed polymer and DBU in vacuo. Because self-standing films of this type are difficult to form, thin films were prepared on the flask wall. To fabricate homogeneous films, we also reduced the DBU content from 200 mol% to 20 mol%. Note that the T_g of P2AU decreased from 42 to –2 °C upon the addition of DBU, probably because DBU behaved as a plasticizer for the polymer. The CO₂-fixation reaction at 60 °C transformed both the homopolymer P2AU and the copolymers P3AU–P5AU into their corresponding polyoxazolidinones (P2OZ–P5OZ) at high conversion rates (Fig. 2 and ESI). Figure 3 shows the time-conversion curves, and Table 3 summarizes the results of the solvent-free CO₂-fixation reaction. The reaction rate increased at higher DEAEMA compositions, as expected from its basic nature. Namely, the times to near-quantitative conversion were shorter for P3AU (9 h), P4AU (9 h), and P5AU (6 h) than for P2AU (22 h). As shown in the time-conversion curve profile, P2AU did not reach full conversion even after a prolonged reaction time. Note that the CO₂ fixation in P3AU–P5AU did not proceed without DBU, indicating that the tertiary amine moiety of the DEAEMA unit could not catalyze the reaction alone. It can be concluded that the competitive effect of DBU and DEAEMA accelerated the CO₂ fixation even under solvent-free conditions. Z-isomers were preferentially formed in all cases. Because both the DBU contents and preparation procedures differed between P1AU and P2AU–P5AU, comparing the isomer ratios among these polymers is of low significance, although P1AU gave the highest Z-isomer ratio. On the other hand, the Z-selectivities of P2AU–P5AU slightly decreased with increasing DEAEMA content for reasons that are unclear at present.

Fig. 2

Partial ¹H NMR spectra of P2AU (a) before and (b) after the solvent-free CO₂-fixation reaction at 60 °C for 22 h (500 MHz, 298 K, CDCl₃)

Fig. 3

Time-dependent curves of the CO₂-fixation reactions of P2AU–P5AU

Table 3 Results of solvent-free CO₂-fixation of P1AU–P5AU^a

Physical properties of the polymers

The M_n of P1OZ was lower than that of P1AU, whereas P2OZ–P5OZ exhibited M_n values similar to those of P2AU–P5AU (Table 2). The M_n values clearly depend on the position of the reactive group (SEC charts of the polymers are shown in the ESI). This phenomenon can be explained as follows. The formation of a cyclic structure would shrink the main chain to some extent, decreasing the hydrodynamic volume of P1OZ, although main-chain scission cannot be ruled out. On the other hand, P2OZ–P5OZ were produced by modifying only the side chains of the vinyl polymer; the main chain was unchanged. Finally, the thermal stabilities of the polymers were investigated by TGA analysis. The T_d5 values of P1OZ–P5OZ were higher than those of P1AU–P5AU, owing to the thermally stable oxazolidinone moiety formed in the former [12]. Notably, the T_d5 values of the P1OZ and P2OZ homopolymers were 88 °C and 66 °C higher than those of P1AU and P2AU, respectively. Both the T_d5 and T_g of P1OZ–P5OZ increased after CO₂ fixation, probably because the fixation increased the rigidity of the polymer structures. Consequently, the thermal stability was improved through the CO₂-fixation reaction.

Conclusion

We successfully performed solvent-free CO₂ fixation into polymers having 2-pyridyl group-substituted propargylamine moieties in the main and side chains. Achieving polymer reactions under such conditions is notoriously difficult. The reactivity of the polymer films was improved by the inclusion of the pendant tertiary amine moiety in the reaction. Additionally, the CO₂-fixation reaction clearly enhanced the thermal stability of the polymers. Polymer resins hardened by the incorporation of CO₂ can be fabricated by applying similar propargylamine moiety-containing polymers to cross-linked systems [13]. As the polyoxazolidinone backbone can be regarded as a new class of polyurethane and is actually used for thermoplastic materials [14], further studies focusing on the functions of the polyoxazolidinones will provide novel polymer materials.