Synthesis and characterization of amino-functionalized poly(propylene carbonate)

Abstract

The first synthesis of amino-functionalized poly(propylene carbonate) (PPC) by terpolymerization of carbon dioxide (CO₂), propylene oxide (PO), and N,N-dibenzyl amino glycidol (DBAG) following the removal of benzyl protecting groups. The copolymerization conditions were investigated in detail. PPC with varying monomer DBAG contents (0–3.3 mol%) was obtained in high yield. The benzyl groups of the terpolymers could be completely removed to provide amino-functionalized PPC without backbone degradation. Thermal properties and contact angles of the functionalized PPCs were measured, showing the dependence of glass transition temperature (T_g) on molecular weight and the expected increase in hydrophilicity with increasing content of amino entities.

---

Carbon dioxide (CO₂) is the main gas leading to the greenhouse effect, and it is also a non-toxic, cheap, renewable and abundant C₁ resource. Nowadays use of CO₂ as the raw material for synthetic chemicals and materials is gradually becoming the focus of attention in the promotion of the rational use of resources.^1–4 The alternating copolymerization of CO₂ and epoxides, which was first reported by Inoue and co-workers in 1960s,⁵ has been considered as one of the most promising processes to produce a variety of aliphatic polycarbonates.⁶ Among these, poly(propylene carbonate) (PPC) derived from CO₂ and propylene oxide (PO) is known as a cheap and biodegradable polymer material due to its good properties such as compatibility, translucence, innocuousness etc.⁷ PPC has great potential application in the development of plastics, elastomers, fibers, adhesives, and so on.⁸ However, the practical application of PPC in the biomedical field has been limited by the lack of functional groups in the backbone.

It has been reported that linear aliphatic functional polycarbonate exhibit great potential for medical devices, drug delivery systems and in tissue engineering.^9,10 Therefore, many approaches have been developed to the synthesis of functional PPC, such as copolymerization of CO₂ and functional epoxides,¹¹ terpolymerization of CO₂, PO and functional epoxides.^12,13 In fact, copolymerization of CO₂ and functional epoxides major depends on the activities of epoxides and catalyst, leading to the increase of cost. For example, Frey reported that the copolymerization of carbon dioxide and glycidyl ether to prepare the functional aliphatic polycarbonates, but molecular weight of the resulting copolymers was low.¹⁴ In contrast, terpolymerization of CO₂, PO and functional epoxides is an efficient method to prepare functional PPC. The preparation of functional aliphatic polycarbonates from CO₂ PO and epoxides with functional groups randomly distributed along the PPC backbone requires a suitable functional comonomer to tolerate the polymerization conditions. Therefore, this strategy often requires two steps including the incorporation of the third monomer into PPC chains and the post-treatment of terpolymer. It is reported that “click chemistry” or the deprotection of functional groups have been used for post-polymerization reactions. Frey et al.¹⁵ reported the terpolymerization of CO₂, PO and 1,2-epoxy-5-hexene following with thiol–ene reaction using mercaptoethanol and thioglycolic acid to obtain hydroxyl and carboxyl functional PPC. Theato et al.¹⁶ reported the terpolymerization of CO₂, PO and 2-[[(2-nitrophenyl)methoxy]-methyl] oxirane, and then by ultraviolet (UV) light irradiation for removing the protection group to obtain hydroxyl functional PPC. Frey et al.¹⁷ reported the synthesis of hydroxy functional polycarbonates by direct random copolymerization of CO₂ with 1,2-isopropylidene glyceryl glycidyl ether as a protected bishydroxy-functional glycidyl ether. Furthermore, glycidyl methyl ether has also been employed in a terpolymerization to tailor both functionality and solubility. Subsequent acidic hydrolysis resulted in stable aliphatic polycarbonates with 1,2-diol side chains without degradation of the polymer backbone.

Although several attempts have been achieved to incorporated functional groups into the backbone of PPC, there has been no report for the synthesis of amino-functionalized PPC. In fact, the amino group is a very interesting functionality, which is utilized in many organic reactions as a consequence of its versatile reactivity, especially in drug delivery and tissue engineering.^18–20 Therefore, in the current work, we created the amino functional PPC by the terpolymerization of CO₂, PO and N,N-dibenzyl amino glycidol (DBAG) following the reaction to remove the benzyl group of terpolymers. The results showed that the third monomer DBAG can not impede the copolymerization conditions, the benzyl group of terpolymers can be completely removed without damaging the backbone structure. The copolymerization conditions and properties of the amino functional aliphatic polycarbonates were studied in this work.

The synthetic strategy developed to obtain the amino-functionalized PPC is shown in Scheme 1. Comonomer DBAG can be prepared from dibenzylamine and epichlorohydrin in mild conditions (Scheme S1 in ESI†).²¹ Moreover, it has been reported that the benzyl group of DBAG can be completely removed to produce amino group.²² Thus, amino-functionalized PPC prepared from the terpolymerization of CO₂, PO and DBAG catalyzed by zinc glutarate (ZnGA) following the removal of benzyl group from polymer backbone.

Scheme 1 Synthesis of amino-functionalized poly(propylene carbonate).

The copolymerization conditions of CO₂, PO and DBAG was detailed investigated. As shown in Table S1,† the optimum copolymerization conditions were 80 °C, 20 h and 5.0 MPa CO₂ pressure. Under these conditions, the terpolymerization of CO₂, PO and DBAG was achieved in high yield. As shown in Table 1, the yield of the copolymerization was between 5365 to 7091 g mol⁻¹ Zn. The content of DBAG in terpolymers increased from 1.3 to 3.3 mol%. However, M_n of the terpolymers (PPC–DBAGs) derived from CO₂, PO and DBAG decreased from 41 000 to 19 000 g mol⁻¹ corresponding to DBAG/PO molar feed ratios varying from 0 to 15. It is concluded that the activity of DBAG copolymerized with CO₂ was lower than that of PO. So there has no polymer obtained when the DBAG/PO molar feed ratio increasing to 20.

Table 1 Characterization data of copolymer samples^a

Sample	DBAG/PO feed (mol%)	Yield (g mol^−1 of Zn)	DBAG^b in polymer (mol%)	Carbonate linkages^c (%)	Mn^d PPC–DBAG (g mol ^−1)	Mw/Mn^d PPC–DBAG	Mn PPC–NH2 (g mol ^−1)	Mw/Mn PPC–NH2
a Copolymerization conditions: CO2 pressure 5.0 MPa, temperature 80 °C, reaction time 20 h.b Comonomer DBAG content, calculated from ^1H NMR spectra of the PPC–DBAG polymer samples.c By ^1H NMR spectroscopy.d By gel permeation chromatography (GPC) calibrated to polystyrene standards in THF at room temperature.e No polymer obtained.
P1	0	6195	0	94%	41 000	3.2	39 500	3.1
P2	2.5	7091	1.3	98%	35 600	3.1	31 300	3.1
P3	5	6235	1.9	95%	29 100	2.8	26 100	2.6
P4	10	5998	2.5	93%	25 000	2.5	21 600	2.3
P5	15	5365	3.3	92%	19 000	2.0	14 900	2.1
P6^e	20

---

The structure of PPC–DBAGs was confirmed by ¹H NMR spectroscopy (Fig. 1 and S3 in ESI†). Compared with PPC, the absorption peaks at 1.3, 4.2 and 5.0 ppm were assigned to CH₃, CH₂ and CH in the carbonate unit, respectively. Agreement of the DBAG fraction in the copolymers with the composition of the epoxide monomer feed is confirmed by ¹H NMR spectroscopy from the comparison of the polycarbonate backbone signals a, b, c and d (3.9, 3.7, 5.2, and 2.4 ppm), the absorption peaks at 7.2–7.5 ppm were assigned to the phenyl group of DBAG, indicating that DBAG were incorporated into the PPC backbone successfully.

Fig. 1 ¹H NMR spectra of P5 PPC–DBAG and P5 PPC–NH₂.

It has been reported that the benzyl group of DBAG could be transformed into amino group by using ceric ammonium nitrate (CAN) at room temperature.²² Therefore, amino-functionalized PPC (PPC–NH₂) were easily prepared by removing the benzyl protecting group of PPC–DBAG. The resulting PPC–NH₂ polymers were characterized by ¹H NMR spectroscopy with respect to its composition (Fig. 1 and S4 in ESI†). From ¹H NMR spectroscopy of PPC–NH₂, the absorption peaks at 3.7, 3.9 and 7.2–7.5 ppm assigned to the benzyl group of DBAG have disappeared. Meanwhile, the FT-IR spectra (Fig. 2) directly proved that PPC–NH₂ was successfully obtained. Compared with PPC–DBAG the new characteristic absorption at 3365 cm⁻¹ was assigned to the amino stretching vibration. The molecular weights determined by GPC of the PPC–NH₂ samples are very comparable to corresponding PPC–DBAG samples (Table 1). It may thus be anticipated that main chain degradation is not remarkably taking place during the removal of benzyl groups from the backbone of PPC–DBAG. Note that the molecular weight decreased after the deprotection due to the loss of the benzyl protecting groups.

Fig. 2 FT-IR spectra of P5 PPC–DBAG and P5 PPC–NH₂.

The glass transition temperature (T_g) of PPC–DBAG and PPC–NH₂ samples are listed in Table 2. It can be seen that P2 PPC–DBAG exhibits the highest T_g of 37.6 °C owing to the incorporation of DBAG into the backbone of PPC. However, T_g of PPC–DBAG decreased with increasing DBAG content, which ascribes to the decrease of the molecular weight of PPC–DBAG. T_g of PPC–NH₂ polymers are slightly lower than those of PPC–DBAG due to the removal of benzyl protecting group.

Table 2 Glass transition temperature and contact angle of polymer samples

Sample based on	Tg^a (°C) PPC–DBAG	Tg (°C) PPC–NH2	CA^b (deg) PPC–DBAG	CA (deg) PPC–NH2
a Glass transition temperature (Tg) determined by DSC.b Contact angle.
P1	37.0	37.1	70.7	70.1
P2	37.6	36.6	71.1	66.7
P3	37.4	36.3	72.4	64.1
P4	37.0	35.8	73.8	62.1
P5	35.3	34.6	76.1	60.2

---

Consistently, the hydrophilicity of the respective polymers P1–P5 PPC–DBAG and PPC–NH₂ are different. The change in polarity is for instance expressed in the contact angle of polymer thin films to water (Table 2). Contact angles of PPC–DBAG polymers are in the range of 70.7–76.1°, corresponding to DBAG/PO molar feed ratios varying from 0 to 15. The contact angle of P1–P5 PPC–DBAG is showing an upward tendency may be attributed to the increasing DBAG content, indicating that the hydrophilic gradually reduced. Meanwhile, the PPC–NH₂ polymers have a significantly decreased contact angle from 70.1° to 60.2°. The higher the content of comonomer units in the PPC–NH₂, the lower the contact angle is (Fig. 3).

Fig. 3 Contact angle analysis of PPC–DBAG and PPC–NH₂ samples.

In conclusion, a new type of amino-functionalized, aliphatic polycarbonate random copolymers from CO₂, PO, and a protected glycidyl ether derivative (DBAG) has been synthesized. The protecting groups of DBAG can be completely removed by CAN with no impact on the backbone of PPC. The amino group content of PPC–NH₂ polymers could be adjusted by varying the molar ratio feed of DBAG comonomer. The composition, glass transition temperature and hydrophilicity of terpolymers have been investigated. The resulting PPC–NH₂ polymers with high hydrophilic are expected to be used in wide application fields and as a general precursor for the post-functionalized PPCs.

Acknowledgements

We are grateful to the project of Gansu province department of education (2013A-009), Lanzhou Sci Techn Bureau (2014-2-4), NWNU-LKQN-12-4 and PCSIRT (IRT1177) for financial support.

References

1. M. Cokoja, C. Bruckmeier and B. Rieger, Angew. Chem., Int. Ed., 2011, 50, 8510–8517.
2. T. Sakakura, J. C. Choi and H. Yasuda, Chem. Rev., 2007, 107, 2365–2387.
3. E. J. Beckman, Science, 1999, 283, 946–947.
4. T. Sakakura and K. Kohno, Chem. Commun., 2009, 1312–1330.
5. S. Inoue, H. Konuma and T. Tsuruta, J. Polym. Sci., Part B: Polym. Lett., 1969, 7, 287–292.
6. D. J. Darensbourg, Chem. Rev., 2007, 107, 2388–2410.
7. G. A. Luinstra and E. Borchardt, Adv. Polym. Sci., 2012, 245, 29–48.
8. G. A. Luinstra, Polym. Rev., 2008, 48, 192–219.
9. J. Feng, R. X. Zhuo and X. Z. Zhang, Prog. Polym. Sci., 2012, 37, 211–236.
10. H. Tian, Z. Tang and X. Zhuang, Prog. Polym. Sci., 2012, 37, 237–280.
11. J. Hilf and H. Frey, Macromol. Rapid Commun., 2013, 34, 1395–1400.
12. J. Geschwind, F. Wurm and H. Frey, Macromol. Chem. Phys., 2013, 214, 892–901.
13. J. Hilf, A. Phillips and H. Frey, Polym. Chem., 2014, 5, 814–818.
14. J. Geschwind and H. Frey, Macromolecules, 2013, 46, 3280–3287.
15. J. Geschwind, F. Wurm and H. Frey, Macromol. Chem. Phys., 2013, 214, 892–901.
16. X. Wu, H. Zhao, B. Nörnberg, P. Theato and G. A. Luinstra, Macromolecules, 2014, 47, 492–497.
17. J. Geschwind and H. Frey, Macromol. Rapid Commun., 2013, 34, 150–155.
18. D. M. Ryan and B. L. Nilsson, Polym. Chem., 2012, 3, 18–33.
19. W. Tang and S. C. Ng, Nat. Protoc., 2008, 3, 691–697.
20. J. Ding, F. Shi and C. Xiao, Polym. Chem., 2011, 2, 2857–2864.
21. B. Obermeier, F. Wurm and H. Frey, Macromolecules, 2010, 43, 2244–2251.
22. S. D. Bull, S. G. Davies and P. M. Kelly, J. Chem. Soc., Perkin Trans., 2001, 23, 3106–3111.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra02854j

---