View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/C4RA11126E (Paper) RSC Adv., 2014, 4, 57210-57217

Highly stereoselective bimetallic complexes for lactide and ε-caprolactone polymerization†

Xuan Pang, Ranlong Duan, Xiang Li, Zhiqiang Sun, Han Zhang, Xianhong Wang and Xuesi Chen*
Key Laboratory of Polymer Ecomaterials, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun 130022, P. R. China. E-mail: xschen@ciac.ac.cn

Received 24th September 2014 , Accepted 16th October 2014

First published on 17th October 2014

Abstract

A series of Schiff-base compounds containing bimetallic ligands and two aluminum centers have been synthesized and characterized by 1H, 13C NMR and elemental analysis. These compounds could be successfully used as catalysts for the polymerization of rac-lactide and ε-caprolactone. The polylactides (PLAs) prepared with these compounds were highly isotactic enriched (Pm = 0.97). The nature and steric hindrance of the ligands coordinated to the central metal ions remarkably influenced the polymer properties. Kinetic studies revealed that the polymerization was first order with respect to each of these compounds and lactide/caprolactone monomers. All the polymerizations were living, with good molecular weight control and relatively narrow molar mass distributions.

Introduction

Because fossil resources are inexorably being depleted, research has been focused on sustainable solutions. As an alternative to petrochemical-based polymers, biodegradable polymers have attracted increasing interest in recent years. Synthetic polyesters, such as poly(lactic acid) (PLA) and poly(ε-caprolactone) (PCL), are well-known biodegradable and biocompatible materials, and they can degrade via hydrolytic cleavage of the ester bonds of the polymer backbone.1–3 PLA and PCL are generally synthesized by the ring-opening polymerization (ROP) of lactide (LA) and ε-caprolactone (ε-CL).4–6 Because of the presence of two chiral centers in the lactide monomer, different lactide stereoisomers are distinguished, namely, (S,S)-LA (L-LA), (R,R)-LA (D-LA), and (R,S)-LA (meso-LA). The stereochemistry of the monomeric units in the polymer chains plays an important role in the mechanical, physical, and degradation properties of PLA materials.7,8 Highly stereoregular (S,S)-PLA and (R,R)-PLA are crystalline polymers, with a high melting temperature and good mechanical strength.7 Atactic PLAs with a random placement of S- and R-LA units in the polymer chains are amorphous and brittle materials. Because of the crystallinity of stereoregular PLAs, these materials slowly degrade in a physiological environment, whereas atactic PLA materials degrade considerably faster.8 The synthesis of stereoregular PLA materials starting from a racemic mixture of (S,S)- and (R,R)-LA, referred to as rac-LA, can be traced back to 1996 by the pioneering studies of Spassky and coworkers.9 They showed that an enantiomeric pure Schiff base aluminum alkoxide catalyst-initiator preferably polymerized (S,S)-LA over (R,R)-LA from rac-LA, leading to an isotactic type PLA material with a gradient of S- and R-LA units in the polymer chains. A few studies have attempted to elucidate the relationship between the monometallic aluminum salen Schiff base complexes and the stereoselectivity via the enantiomorphic site control mechanism or chain-end control mechanism.10–18

However, few studies about bimetallic Schiff base aluminum complexes have been reported, despite their great potential structure varieties.19–24 Recently, our research lab has successfully developed some bimetallic aluminum salen complexes; they were used as catalysts to prepare isotactic enriched PLA from rac-LA.25–27 To the best of our knowledge, no systematic studies about the effects of phenolate substituents and the bridging part of the bimetallic Schiff base ligand have yet been reported.

Intrigued by the successful use of those complexes in polymerization catalysis, we are very interested to further investigate the highly stereoselective complexes based on flexible skeletons. To further expand our previous investigation on bimetallic aluminum salen complexes, we synthesized a series of bimetallic Schiff base ligands and their aluminum complexes with flexible bridging part (Scheme 1). Compared with our recently reported work, the research described in this work is focused on our most recent progress of bimetallic aluminum catalyst systems with a flexible bridging part for the controlled and very highly stereoselective ROP of LA and ε-CL.


image file: c4ra11126e-s1.tif
Scheme 1 Synthetic pathway for the preparation of ligands and complexes.

Experimental section

General

All the experiments were carried out under a dry nitrogen atmosphere in a glovebox. Starting materials for the synthesis of ligands, rac-LA and ε-caprolactone were purchased from Aldrich Inc. Ethyl acetate and 2-propanol were distilled from CaH2 under the protection of argon. rac-Lactide was purified by recrystallization from ethyl acetate and dried under vacuum at room temperature before use. NMR spectra were recorded on a Bruker AV 400 M. Chemical shifts were given in parts per million from tetramethylsilane. Gel permeation chromatography (GPC) measurements were conducted with a Waters 515 GPC with CHCl3 as the eluent (flow rate: 1 mL min−1 at 35 °C). The molecular weights were calibrated against polystyrene (PS) standards.

Synthesis of compounds I and II

Compounds I and II were prepared by a modified method according to the reported procedure.30,31 Compound I: A stirred mixture of dimethyl malonate and paraformaldehyde at 60 °C was treated with 10 wt% KOH in MeOH (10 mL) with stirring for 14 h. Then, the mixture was cooled to RT and shaken with H2O (100 mL). The lower organic phase layer was separated and excess dimethyl malonate was removed by distillation at reduced pressure. After cooling, the residue was treated with conc. aq. NH3 (300 mL) and the resulting biphasic mixture was vigorously stirred for 24 h. The resulting precipitated product was removed by filtration and successively washed with EtOAc-MeOH (4[thin space (1/6-em)]:[thin space (1/6-em)]1, 250 mL), acetone (100 mL), and then dried in an oven (120 °C, 15 h) to afford compound I, tetracarboxamidopropane.

Compound II: BH3 (1 M, 200 mL in THF) was slowly added over 1 h to a THF (40 mL) suspension of compound I (5.0 g, 23.1 mmol). The reaction vessel was kept in an ice bath during the addition and once the addition was complete, the mixture was refluxed overnight. The reaction mixture was cooled to ambient temperature and the excess of borane was destroyed by the cautious addition of H2O (20 mL). The solution was evaporated to dryness and slowly treated with 5 M HCl (100 mL). The resulting mixture was refluxed for 1 h and then heated to dryness. The obtained white residue was dissolved in H2O (50 mL) and 5 M NaOH (40 mL) was added. The solution was stirred for 30 min and evaporated to dryness. Compound II was directly used without purification.

Synthesis of ligands 1–4

A solution of compound II (0.5 g) in ethanol (40 mL) was added dropwise to a stirred solution of 3,5-di-substituted salicylaldehyde in ethanol. The reaction mixture was refluxed for 10 h before cooling to room temperature. The precipitate was collected by vacuum filtration and recrystallization in C2H5OH/CHCl3.

1: 1H NMR (400 MHz, CDCl3): δ = 13.34(s OH 4H), 8.36(s NCH 4H), 7.31, 7.19, 6.96, 6.86(m ArH 16H), 3.68(m NCH2CH 8H), 2.32(m CH(CH2)3 2H), 1.58(t CHCH2CH 2H). 13C NMR (100 MHz, CDCl3): δ = 166.42(NCH), all benzene ring: 160.83, 132.52, 131.61, 118.84, 117.07; 61.34(CHCH2N), 38.29(CH(CH2)3), 31.62(CHCH2CH). Elem. Anal.: calcd. C 72.90, H 6.29, N 9.72; Found C 72.88, H 6.32, N 9.71.

2: 1H NMR (400 MHz, CDCl3): δ = 13.28(s OH 4H), 8.30(s NCH 4H), 7.03(d ArH 4H), 6.85(d ArH 4H), 3.70(m NCH2CH 8H), 2.33(m CH(CH2)3 2H), 2.27(s ArCH3 24H), 1.60(t CHCH2CH 2H). 13C NMR (100 MHz,CDCl3): δ = 166.50(NCH), all benzene ring: 157.21, 134.22, 129.18, 127.20, 125.65, 117.78; 61.30(CHCH2N), 38.15(CH(CH2)3), 31.45(CHCH2CH), 20.40, 15.51(ArCH3). Elem. Anal.: calcd. C 74.97, H 7.61, N 8.13; Found C 74.99, H 7.62, N 8.15.

3: 1H NMR (400 MHz, CDCl3): δ = 13.79(s OH 4H), 8.42(s NCH 4H), 7.39(d ArH 4H), 7.10(d ArH 4H), 3.65(m NCH2CH 8H), 2.37(m CH(CH2)3 2H), 1.63(t CHCH2CH 2H), 1.45(s C(CH3)3 36H), 1.30(s C(CH3)3 36H). 13C NMR (100 MHz, CDCl3): δ = 167.34(NCH), all benzene ring: 158.27, 140.25, 136.83, 127.16, 126.16, 118.04; 61.13(CHCH2N), 37.99(CH(CH2)3), 35.22, 34.30(ArC(CH3)3), 32.26(CHCH2CH), 31.66, 29.62(C(CH3)3). Elem. Anal.: calcd. C 78.47, H 9.83, N 5.46; Found C 78.45, H 9.82, N 5.42.

4: 1H NMR (400 MHz, CDCl3): δ = 13.50(s OH 4H), 8.32(s NCH 4H), 7.40, 7.05, 6.84(m ArH 12H), 3.65(m NCH2CH 8H), 2.38(m CH(CH2)3 2H), 1.52(t CHCH2CH 2H), 0.92(s SiC(CH3)3 36H), 0.33(s Si(CH3)2 24H). 13C NMR (100 MHz, CDCl3): δ = 166.65(NCH), all benzene ring: 166.18, 139.57, 133.04, 125.08, 118.29, 117.88; 61.58(CHCH2N), 38.32(CH(CH2)3), 31.86(CHCH2CH), 27.27(SiC(CH3)3), 17.83(SiC(CH3)3), −4.56(Si(CH3)2). Elem. Anal.: calcd. C 68.55, H 8.97, N 5.42; Found C 68.57, H 8.95, N 5.41.

Synthesis of complexes 1a–4a

AlEt3 (0.2 mmol) in toluene (5 mL) was added to the stirred 1 mL toluene solution of ligand precursors 1–4 (0.1 mmol) at RT. The reaction was maintained at 80 °C for 12 h, and the reaction mixture was then slowly cooled to RT. The toluene was removed under vacuum.

1a: 1H NMR (400 MHz, CDCl3): δ = 8.19(s NCH 4H), 7.47, 7.18, 6.92, 6.65(m ArH 16H), 3.53(m NCH2CH 8H), 2.29(m CH(CH2)3 2H), 1.28(t CHCH2CH 2H), 1.02(m AlCH2CH3 6H), −0.08(m AlCH2CH3 4H). 13C NMR (100 MHz, CDCl3): δ = 173.18(NCH), all benzene ring: 169.79, 165.12, 138.20, 135.00, 132.87, 121.94, 118.08, 116.64; 58.34(CHCH2N), 36.28(CH(CH2)3), 29.82(CHCH2CH), 9.00(AlCH2CH3), 0.62(AlCH2CH3). Elem. Anal.: calcd. C 68.41, H 6.18, N 8.18; Found C 68.45, H 6.17, N 8.16.

2a: 1H NMR (400 MHz, CDCl3): δ = 7.94(s NCH 4H), 7.20(d ArH 4H), 6.87(d ArH 4H), 3.54(m NCH2CH 8H), 2.31(m CH(CH2)3 2H), 2.22(s ArCH3 24H), 1.30(t CHCH2CH 2H), 0.98(m AlCH2CH3 6H), −0.09(m AlCH2CH3 4H). 13C NMR (100 MHz, CDCl3): δ = 173.00(NCH), all benzene ring: 161.73, 139.70, 137.42, 132.02, 129.62, 125.44, 117.37; 63.40(CHCH2N), 39.84(CH(CH2)3), 29.84(CHCH2CH), 20.25, 16.09(ArCH3), 9.00(AlCH2CH3), 0.62(AlCH2CH3). Elem. Anal.: calcd. C 70.83, H 7.34, N 7.03; Found C 70.79, H 7.32, N 7.01.

3a: 1H NMR (400 MHz, CDCl3): δ = 8.23(s NCH 4H), 7.61(d ArH 4H), 7.07(d ArH 4H), 3.54(m NCH2CH 8H), 2.60(m CH(CH2)3 2H), 1.59(m CHCH2CH 2H), 1.47(s C(CH3)3 36H), 1.34(s C(CH3)3 36H), 1.04(m AlCH2CH3 6H), −0.05(m AlCH2CH3 4H). 13C NMR (100 MHz, CDCl3): δ = 174.07(NCH), all benzene ring: 167.35, 162.49, 140.69, 139.51, 133.08, 129.10, 118.23; 58.01(CHCH2N), 36.85(CH(CH2)3), 35.44, 34.19(ArC(CH3)3), 35.23(CHCH2CH), 31.38, 29.46(C(CH3)3), 9.10(AlCH2CH3), 0.79(AlCH2CH3). Elem. Anal.: calcd. C 75.23, H 9.43, N 4.94; Found C 75.25, H 9.39, N 4.96.

4a: 1H NMR (400 MHz, CDCl3): δ = 8.16(s NCH 4H), 7.63, 7.43, 6.87(m ArH 12H), 3.54(m NCH2CH 8H), 2.57(m CH(CH2)3 2H), 1.49(t CHCH2CH 2H), 1.04(m AlCH2CH3 6H), 0.95(s SiC(CH3)3 36H), 0.32(s Si(CH3)2 24H), −0.08(m AlCH2CH3 4H). 13C NMR (100 MHz, CDCl3): δ = 173.60(NCH), all benzene ring: 166.65, 145.65, 139.57, 136.57, 133.05, 117.78; 58.40(CHCH2N), 36.48(CH(CH2)3), 29.86(CHCH2CH), 27.33(SiC(CH3)3), 17.55(SiC(CH3)3), 9.02(AlCH2CH3), 0.70(AlCH2CH3), −4.55(Si(CH3)2). Elem. Anal.: calcd. C 65.36, H 9.71, N 3.50; Found C 65.34, H 9.68, N 3.47.

Synthesis of complexes 1b–4b

The treatment of the appropriate aluminum ethyls 1a–4a with stoichiometric 2-propanol in toluene resulted in the formation of the desired aluminum isopropoxides 1b–4b along with the release of C2H6 at 80 °C for 10 h. The reaction mixture was then slowly cooled to RT. The toluene was removed under vacuum.

1b: 1H NMR (400 MHz, CDCl3): δ = 8.14(s NCH 4H), 7.44, 7.23, 6.70(m ArH 16H), 4.29(m NCH2CH 8H), 3.39(m AlOCH(CH3)2), 2.41(m CH(CH2)3 2H), 1.39(t CHCH2CH 2H), 0.93(m AlOCH(CH3)2 12H). 13C NMR (100 MHz, CDCl3): δ = 170.68(NCH), all benzene ring: 167.68, 138.37, 135.69, 133.26, 121.77, 116.49; 66.94(CHCH2N), 63.09(AlOCH(CH3)2), 35.62(CH(CH2)3), 29.85(CHCH2CH), 27.85(AlOCH(CH3)2). Elem. Anal.: calcd. C 66.12, H 6.23, N 7.52; Found C 66.15, H 6.21, N 7.51.

2b: 1H NMR (400 MHz, CDCl3): δ = 8.00(s NCH 4H), 7.21(d ArH 4H), 6.79(d ArH 4H), 4.18(m NCH2CH 8H), 3.36(m AlOCH(CH3)2), 2.30(m CH(CH2)3 2H), 2.23(s ArCH3 24H), 1.35(t CHCH2CH 2H), 0.85(m AlOCH(CH3)2 12H). 13C NMR (100 MHz, CDCl3): δ = 169.97(NCH), all benzene ring: 163.18, 139.07, 137.45, 130.40, 129.63, 124.60, 117.08; 67.36(CHCH2N), 63.65(AlOCH(CH3)2), 34.13(CH(CH2)3), 29.84(CHCH2CH), 27.62(AlOCH(CH3)2), 20.37, 16.01(ArCH3). Elem. Anal.: calcd. C 68.67, H 7.29, N 6.54; Found C 68.66, H 7.32, N 6.53.

3b: 1H NMR (400 MHz, CDCl3): δ = 8.18(s NCH 4H), 7.46(d ArH 4H), 7.00(d ArH 4H), 3.94(m NCH2CH 8H), 3.36(m AlOCH(CH3)2), 2.31(m CH(CH2)3 2H), 1.50(m CHCH2CH 2H), 1.46(s C(CH3)3 36H), 1.32(s C(CH3)3 36H), 0.98(m AlOCH(CH3)2 12H). 13C NMR (100 MHz, CDCl3): δ = 170.46(NCH), all benzene ring: 163.80, 140.67, 138.01, 130.74, 128.37, 127.09, 118.08; 62.54(CHCH2N), 61.86(AlOCH(CH3)2), 36.12(CH(CH2)3), 35.58, 34.09(ArC(CH3)3), 29.45(CHCH2CH), 31.51, 29.87(C(CH3)3), 28.12(AlOCH(CH3)2). Elem. Anal.: calcd. C 73.45, H 9.29, N 4.69; Found C 73.46, H 9.32, N 4.71.

4b: 1H NMR (400 MHz, CDCl3): δ = 8.16(s NCH 4H), 7.53, 7.20, 6.73(m ArH 12H), 3.63(m NCH2CH 8H), 3.37(m AlOCH(CH3)2), 2.24(m CH(CH2)3 2H), 1.29(t CHCH2CH 2H), 0.94(s SiC(CH3)3 36H), 0.78(m AlOCH(CH3)2 12H), 0.32(s Si(CH3)2 24H). 13C NMR (100 MHz, CDCl3): δ = 173.59(NCH), all benzene ring: 163.80, 140.67, 138.01, 130.74, 128.37, 127.09, 118.08; 63.09(AlOCH(CH3)2), 58.39(CHCH2N), 37.14(CH(CH2)3), 29.86(CHCH2CH), 27.33(SiC(CH3)3), 26.95(AlOCH(CH3)2), 17.78(SiC(CH3)3), −4.55(Si(CH3)2). Elem. Anal.: calcd. C 64.44, H 9.60, N 3.38; Found C 64.41, H 9.62, N 3.41.

Results and discussion

Ligand synthesis

The synthetic pathway of the ligand family is described in Scheme 1. Ligand precursors 1–4 were obtained by condensation reaction between the substituted salicylaldehyde and the corresponding tetramine by a four-step synthesis.30,31 The analytical results were in agreement with their respective formulas. For example, 1H NMR spectrum showed signals at about δ 8.36 and 3.68 ppm, which were attributed to the N[double bond, length as m-dash]CH proton and NCH2CH protons of the ligands, respectively. The intensity ratio of the signals at 8.36 and 3.68 ppm was 1[thin space (1/6-em)]:[thin space (1/6-em)]2. This was consistent with the structures of ligand 1.

Complexes formation and characterization

Complexes 1a–4a were obtained via the reaction of ligands 1–4 with stoichiometric AlEt3 (Scheme 1). For example, 1H NMR spectrum showed signals at about δ 1.02 and −0.08 ppm, which were attributed to the methyl protons and methylene protons of the aluminum ethyl group, respectively. The N[double bond, length as m-dash]CH proton displayed a signal at 8.19 ppm. The intensity ratio of the three signals at 1.02, −0.08 and 8.19 ppm was 3[thin space (1/6-em)]:[thin space (1/6-em)]2[thin space (1/6-em)]:[thin space (1/6-em)]1, which confirmed the structure of 1a. To further mimic the structures of the actual initiators, isopropoxides 1b–4b were prepared by in situ alcoholysis. The treatment of 1a–4a with stoichiometric 2-propanol resulted in the formation of 1b–4b (Scheme 1). The AlOCH(CH3)2 protons displayed a multiplet at 0.93 ppm, while the N[double bond, length as m-dash]CH proton showed a multiplet at 8.14 ppm. The intensity ratio of the two signals was 6[thin space (1/6-em)]:[thin space (1/6-em)]1, which confirmed the structures of 1b.

Ring-opening polymerization of rac-LA

Complexes 1a–4a toward the ROP of rac-LA were investigated in the presence of 2-propanol, and the representative results aresummarized in Table 1 and 2. The polymerizations processes and levels of conversion were monitored by 1H NMR via the determination of the samples withdrawn from the reaction mixtures.
Table 1 Polymerization data of rac-LA with complexes 1a–4a and 1b–4ba
Entry Complex Temp (°C) t (h) [LA]0/[Cat] Convb (%) Mn(calcd) × 10−3c Mn (NMR) × 10−3d Mn(GPC) × 10−3e PDIe Pmf
a All polymerizations were carried out in toluene solution, [LA]0 = 0.5 mol L−1.b Measured by 1HNMR.c Calculated from the molecular weight of LA × [LA/2]0/[Cat] × conversion + Mw(iPrOH).d Obtained from 1H NMR analysis.e Obtained from GPC analysis and calibrated against polystyrene standard. The true value of Mn could be calculated according to the formula Mn = 0.58MnGPC.28f Pm.29
1 1a 70 4.5 200 94 13.6 13.9 14.1 1.14 0.67
2 2a 70 5 200 90 13.0 13.4 14.2 1.09 0.73
3 3a 70 9 100 92 6.7 6.1 6.4 1.12 0.90
4 3a 70 10 150 85 9.2 9.7 9.4 1.09 0.91
5 3a 70 16 200 87 12.6 11.8 12.4 1.04 0.90
6 3a 70 29 300 90 19.5 20.2 20.4 1.05 0.91
7 4a 70 52 200 75 10.9 10.5 10.8 1.11 0.97
8 3a 80 16 200 95 13.7 13.1 13.4 1.18 0.85
9 3a 90 9 200 92 13.3 14.3 14.8 1.22 0.81
10 3a 100 6 200 90 13.0 14.5 15.1 1.20 0.77
11 3a 110 5.5 200 94 13.6 14.9 15.5 1.24 0.74
12 1b 70 3 200 88 12.7 12.4 12.5 1.10 0.68
13 1b 70 5 400 82 23.7 24.8 25.2 1.09 0.68
14 2b 70 4 200 87 12.6 12.9 12.4 1.07 0.72
15 3b 70 12 200 81 11.7 11.4 11.2 1.11 0.89
16 3b 80 9.5 200 84 12.2 12.8 12.7 1.20 0.85
17 3b 90 5.5 200 79 11.4 11.9 12.1 1.22 0.80
18 4b 70 35 200 60 8.7 8.3 8.1 1.12 0.97

Table 2 Kinetic results of rac-LA polymerization
Entry Complex Temp/°C [LA]0/[Cat] kapp/s−1 × 10−4 kp/L mol−1 s−1 × 10−2
1 1a 70 200 1.83 7.33
2 2a 70 200 1.38 5.50
3 3a 70 100 0.75 1.50
4 3a 70 150 0.52 1.55
5 3a 70 200 0.38 1.53
6 3a 70 300 0.25 1.50
7 3a 80 200 0.53 2.17
8 3a 90 200 0.83 3.33
9 3a 100 200 1.13 4.50
10 3a 110 200 1.53 6.17
11 4a 70 200 0.08 0.33
12 1b 70 200 1.85 7.33
13 1b 70 400 0.93 7.50
14 2b 70 200 1.33 5.33
15 3b 70 200 0.37 1.48
16 3b 80 200 0.55 2.17
17 3b 90 200 1.01 3.67
18 4b 70 200 0.08 0.33

Kinetics

1H NMR spectroscopy and gel permeation chromatography (GPC) were used to determine the Mn values of the PLA obtained. All the resulting polymers had number-averaged molecular weights (from the 1H NMR) close to the theoretical ones (calculated from the monomer-to-catalyst molar ratio). The molecular weight distributions of the PLAs produced were all narrow, ranging from 1.04 to 1.14. The data from the conversions versus time were collected (Fig. 1). The conversions linearly increased with polymerization time. First-order kinetics in the monomer was observed in each case [eqn (1)], and kapp was the apparent polymerization rate constant.
 
−d[LA]/dt = kapp[LA] (1)
image file: c4ra11126e-f1.tif
Fig. 1 Kinetic plots of the rac-lactide conversion vs. the reaction time, [M]0/[cat] = 200. (a) complex 1a; (b) complex 2a; (c) complex 3a; (d) complex 4a.

When the monomer to catalyst ratio was fixed, 1a had the highest conversion at a certain time, as shown in Fig. 1.

When the complex was selected, for example, complex 3a, the case with the lower monomer to catalyst ratio had the higher conversion at a certain reaction time (see ESI†). A linear relationship between number-average molecular weight (Mn) and monomer conversion was observed (Fig. 2). To investigate the order in catalyst, kapp was plotted versus the concentration of complex 3a (Fig. 3, Table 2, entries 3–6). From this plot, kapp linearly increased with complex 3a concentrations, indicating that the order in complex 3a was first-order as well. Therefore, the polymerization of rac-LA using 3a followed an overall kinetic equation of the form as given by eqn (2), kp was the polymerization rate constants, kp = kapp/[3a].

 
−d[LA]/dt = kp[LA][3a] (2)


image file: c4ra11126e-f2.tif
Fig. 2 Plot of PLA Mn (■) and polydispersity (▲) as a function of rac-lactide conversion using (a) complex 1a, [M]0/[cat] = 200; (b) complex 3a, [M]0/[cat] = 200; (c) complex 3a, [M]0/[cat] = 300.

image file: c4ra11126e-f3.tif
Fig. 3 kapp vs. the concentration of 3a for the rac-LA polymerization.

Kinetic data showed that the activities of 1a–4a could be remarkably influenced by the substituent group on the phenolic ring. Complex 1a had the highest activity (kp values: 0.0733 L mol−1 s−1 for 1a, 0.055 L mol−1 s−1 for 2a, 0.0153 L mol−1 s−1 for 3a and 0.0033 L mol−1 s−1 for 4a, Table 2, entries 1, 2, 5, and 11).

The bulky substituents with more steric hindrance would have a negative effect on the approach of the lactide to the active species, as a result they would slow down the polymerization rate.10,32,33 The influence of temperature on the polymerization of rac-LA was investigated (Fig. 4). The increasing temperature had a positive effect because the polymerization rate increased with the increasing temperature. The apparent polymerization rate constant (kapp) and polymerization rate constants (kp) values are given in Table 2. For example, an increase in the temperature from 70 °C to 110 °C led to an increase in kp value (from 0.0153 L mol−1 s−1 to 0.0617 L mol−1 s−1, Table 2, entry 5 and 10) when using complex 3a. The activation energy of the polymerization was calculated by fitting kp values determined at different temperatures to the Arrhenius equation (kp = Ae−Ea/RT); as shown in Table 2, an activation energy Ea of 37.7 kJ mol−1 was deduced by plotting lnkp versus 1/T (Fig. 5), which was lower than that obtained using the conventional tin(II) octanoate (70.9 ± 1.5 KJ mol−1).34


image file: c4ra11126e-f4.tif
Fig. 4 Kinetics of the rac-LA polymerization using 3a at the reaction temperatures of (a) 110 °C; (b) 100 °C; (c) 90 °C; (d) 80 °C; (e) 70 °C, [M]0/[Cat] = 200.

image file: c4ra11126e-f5.tif
Fig. 5 Plot of LnKp vs. 1/T for the polymerization of rac-lactide with 3a.

Complex structure and stereoselective polymerization

To investigate the microstructures of the PLAs during the polymerization process, the microstructures of the resultant PLAs were determined from the methine region of the homonuclear decoupled 1H NMR spectra. The substituted salicylaldehydes of the ligand had a remarkable effect on the performance of complexes 1a–4a.

The Pm value was 0.67 for 1a, 0.73 for 2a and 0.90 for 3a. It is worth noting that Pm increased to 0.97 when the bulky substituent TBDMS (Si(CH2)2C(CH3)3) was introduced in 4a (Table 1, entries 1, 2, 3, and 7). As in this research, TBDMS would lead to a 33% increase in stereoselectivity compared with the less bulky group H (as in complex 1a). This was consistent with a previous study that the enhancement of stereoselectivity requires bulky substituents at the ortho positions for the stereoselective polymerization adopting chain end control mechanism.10 To date, this was the highest stereoselectivity (Pm = 0.97) using bimetallic Schiff base catalyst for rac-LA polymerization. One of the homonuclear decoupled 1H NMR spectra using complex 4a is shown in Fig. 6.


image file: c4ra11126e-f6.tif
Fig. 6 Homonuclear decoupled 1H NMR spectra using 4a.

The stereoselectivity of complexes 1a–4a decreased with the increasing temperature, as the temperature changed from 70 °C to 110 °C, using 3a led to a reduction in Pm value of 18%, (Pm value from 0.90 at 70 °C to 0.74 at 110 °C, Table 1, entries 3 and 11).

To better understand the real actual initiators, we synthesized isopropoxides complexes 1b–4b (Scheme 1). Complexes 1b–4b toward the ROP of rac-LA were also studied. Kinetic data showed that 1b–4b had almost the same performance compared with the corresponding aluminum ethyl analogies 1a–4a.

It is believed that the polymerization consists of two steps while using in situ alcoholysis. In the first step, 1a–4a reacted with 2-propanol with a rate constant krea to produce 1b–4b as the actual initiators. Then, 1b–4b initiated the LA polymerization with a polymerization rate constant kp. It is postulated that 1a–4a converted to isopropoxy species before the polymerization occurred. If krea ≫ kp, then the observed polymerization rate was mainly determined by and presumably equal to kp. Otherwise, if krea ≪ kp, the observed polymerization rate would be influenced by krea. Our recent monometallic aluminum-based study showed that the bis(pyrrolidene)-based aluminum isopropoxide 2b exhibited a considerably higher activity compared with the in situ-formed one from 2a/2-propanol.23 However, in this research, 1b–4b had similar activities compared with their corresponding analogies 1a–4a (Table 1, entries 12–18 and Table 2, entries 12–18). Therefore, it was postulated that in this work, values of kp were considerably lower than that of krea because of the fast alcoholysis reaction.

Ring-opening polymerization of ε-CL

Because PCLs have received significant attention for use as both implantable biomaterials and as a drug delivery system, complexes 1a–4a were explored to further expand their application in the ROP of ε-CL. Polymerization data are given in Table 3.
Table 3 Polymerization data of ε-CL with complexes 1a–4aa
Entry Complex t (min) [CL]0/[Cat] Convb (%) Mn(calcd) × 10−3c Mn(NMR) × 10−3d Mn(GPC) × 10−3e PDIe kapp/s−1 × 10−4 kp/L mol−1 s−1 × 10−2
a All polymerizations were carried out in toluene solution at 40 °C, [CL]0 = 0.5 mol L−1.b Measured by 1HNMR.c Calculated from the molecular weight of CL × [CL/2]0/[Cat] × conversion + Mw(iPrOH).d Obtained from 1H NMR analysis.e Obtained from GPC analysis and calibrated against polystyrene standard. The true value of Mn could be calculated according to the formula Mn = 0.56MnGPC.35
1 1a 150 200 98 11.2 11.5 11.9 1.19 5.0 20.0
2 1a 275 400 95 21.7 20.4 20.8 1.17 5.2 20.7
3 1a 390 600 92 31.6 30.1 29.8 1.25 5.1 20.3
4 2a 215 200 97 11.1 10.7 10.5 1.22 3.5 14.0
5 2a 295 300 95 16.3 16.0 16.1 1.24 3.4 13.7
6 3a 410 150 94 8.1 7.6 7.5 1.18 1.5 4.3
7 3a 690 200 95 10.9 11.4 11.2 1.22 0.9 3.7
8 3a 725 300 92 15.8 15.5 15.3 1.09 0.6 3.7
9 3a 1230 400 90 20.6 21.2 21.8 1.24 0.5 4.0
10 4a 840 200 92 10.6 10.1 10.3 1.08 0.7 2.7
11 4a 1415 400 95 21.7 21.0 20.7 1.11 0.4 3.0
12 1b 120 200 94 10.8 10.3 10.7 1.14 4.8 19.3
13 2b 195 200 95 10.9 10.4 11.0 1.25 3.5 14.1
14 3b 645 200 94 10.8 11.4 11.9 1.14 0.9 3.7
15 4b 805 200 90 10.3 11.0 11.2 1.11 0.7 2.8

The conversion data as a function of time were collected (Fig. 7). The conversions of ε-CL monomer linearly increased with the polymerization time. First-order kinetics in monomer was observed in each case. Similar results were revealed as in the rac-LA polymerization that when the monomer to catalyst ratio was fixed, 1a had the highest conversion at a certain time (Fig. 7). As mentioned in the rac-LA polymerization, the bulky substituents with more steric hindrance would have a negative effect on the approach of ε-CL to the active species; as a result, they would slow down the polymerization rate (Table 3, entries 1, 4, 7, and 10). Taking complex 3a as an example, the reaction with the lower monomer to catalyst ratio had the higher conversion at a certain reaction time (see ESI†).


image file: c4ra11126e-f7.tif
Fig. 7 Kinetic plots of the ε-CL conversion vs. the reaction time, [M]0/[cat] = 200. (a) complex 1a; (b) complex 2a; (c) complex 3a; (d) complex 4a.

First-order kinetics in the catalyst can also be observed from Fig. 8 because the kapp linearly increases with the catalyst concentration. The number-average molecular weight (Mn) also followed a linear relationship in the monomer conversion. The molecular control and the low polydispersity indicated that the polymerization had a characteristic of controlled propagation (see ESI†). Similar to the rac-LA polymerization, the kinetic data showed that 1b–4b had almost the same performance compared with the corresponding aluminum ethyl analogies 1a–4a in the ROP of ε-CL (Table 3, entries 12–15).


image file: c4ra11126e-f8.tif
Fig. 8 kapp vs. the concentration of 3a for the ε-CL polymerization.

Conclusions

A series of Schiff-base compounds containing bimetallic ligands and two aluminum centers have been synthesized. These complexes were applied as catalysts in rac-lactide and ε-CL polymerization. The activities and stereoselectivities of complexes were thoroughly investigated. The polylactide (PLA) prepared with these compounds were highly isotactic enriched (Pm = 0.97). Kinetic studies showed that the polymerizations were living with narrow molar mass distributions. Polymerization was of first order with respect to each of these compounds and the lactide/caprolactone monomers. To further mimic and confirm the structures of the actual initiators, isopropoxides 1b–4b were prepared via in situ alcoholysis. 1b–4b had almost similar activity compared with 1a–4a. This indicated that the values of the polymerization rate constant (kp) were considerably lower than that of the rate constant to produce the isopropoxide (krea) because of the fast alcoholysis reaction. Following research will focus on the relationship of the activity and stereoselectivity of bimetallic complexes and the application of highly stereoselective PLA materials.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (nos. 51173183, 51103058, 51203155 and 51321062) and the Ministry of Science and Technology of China (no. 2011AA02A202).

Notes and references

  1. K. E. Uhrich, S. M. Cannizzaro, R. S. Langer and K. M. Shakesheff, Chem. Rev., 1999, 99, 3181–3198 CrossRef CAS PubMed.
  2. D. J. Mooney, G. Organ, J. P. Vacanti and R. Langer, Cell Transplant., 1994, 3, 203–210 CAS.
  3. J. L. Eguiburu, M. J. Fernandez-Berridi, F. P. Cossio and J. San Roman, Macromolecules, 1999, 32, 8252–8258 CrossRef CAS.
  4. J. Liu and H. Ma, Dalton Trans., 2014, 43, 9098–9110 RSC.
  5. W.-L. Kong and Z.-X. Wang, Dalton Trans., 2014, 43, 9126–9135 RSC.
  6. V. Balasanthiran, M. H. Chisholm, K. Choojun and C. B. Durr, Dalton Trans., 2014, 43, 2781–2788 RSC.
  7. S. A. Krouse, R. R. Schrock and R. E. Cohen, Macromolecules, 1987, 20, 903–904 CrossRef CAS.
  8. A. Belbella, C. Vauthier, H. Fessi, J. P. Devissaguet and F. Puisieux, Int. J. Pharm., 1996, 129, 95–102 CrossRef CAS.
  9. N. Spassky, M. Wisniewski, C. Pluta and A. LeBorgne, Macromol. Chem. Phys., 1996, 197, 2627–2637 CrossRef CAS.
  10. N. Nomura, R. Ishii, M. Akakura and K. Aoi, J. Am. Chem. Soc., 2002, 124, 5938–5939 CrossRef CAS PubMed.
  11. N. Nomura, R. Ishii, Y. Yamamoto and T. Kondo, Chem.–Eur. J., 2007, 13, 4433–4451 CrossRef CAS PubMed.
  12. H. Du, X. Pang, H. Yu, X. Zhuang, X. Chen, D. Cui, X. Wang and X. Jing, Macromolecules, 2007, 40, 1904–1913 CrossRef CAS.
  13. X. Pang, H. Du, X. Chen, X. Wang and X. Jing, Chem.–Eur. J., 2008, 14, 3126–3136 CrossRef CAS PubMed.
  14. Z. H. Tang, X. S. Chen, X. Pang, Y. K. Yang, X. F. Zhang and X. B. Jing, Biomacromolecules, 2004, 5, 965–970 CrossRef CAS PubMed.
  15. T. M. Ovitt and G. W. Coates, J. Am. Chem. Soc., 2002, 124, 1316–1326 CrossRef CAS PubMed.
  16. Z. Y. Zhong, P. J. Dijkstra and J. Feijen, Angew. Chem., Int. Ed., 2002, 41, 4510–4513 CrossRef CAS.
  17. Z. Y. Zhong, P. J. Dijkstra and J. Feijen, J. Am. Chem. Soc., 2003, 125, 11291–11298 CrossRef CAS PubMed.
  18. K. Majerska and A. Duda, J. Am. Chem. Soc., 2004, 126, 1026–1027 CrossRef CAS PubMed.
  19. A. W. Kleij, Chem.–Eur. J., 2008, 14, 10520–10529 CrossRef CAS PubMed.
  20. S. J. Wezenberg and A. W. Kleij, Angew. Chem., Int. Ed., 2008, 47, 2354–2364 CrossRef CAS PubMed.
  21. D. J. Darensbourg and O. Karroonnirun, Organometallics, 2010, 29, 5627–5634 CrossRef CAS.
  22. N. Maudoux, T. Roisnel, V. Dorcet, J. F. Carpentier and Y. Sarazin, Chem.–Eur. J., 2014, 20, 6131–6147 CrossRef CAS PubMed.
  23. M. Normand, T. Roisnel, J. F. Carpentier and E. Kirillov, Chem. Commun., 2013, 49, 11692–11694 RSC.
  24. I. Bratko and M. Gomez, Dalton Trans., 2013, 42, 10664–10681 RSC.
  25. Z. Qu, R. Duan, X. Pang, B. Gao, X. Li, Z. Tang, X. Wang and X. Chen, J. Polym. Sci., Part A: Polym. Chem., 2014, 52, 1344–1352 CrossRef CAS.
  26. X. Pang, R. Duan, X. Li, B. Gao, Z. Sun, X. Wang and X. Chen, RSC Adv., 2014, 4, 22561–22566 RSC.
  27. X. Pang, R. Duan, X. Li and X. Chen, Polym. Chem., 2014, 5, 3894–3900 RSC.
  28. J. Baran, A. Duda, A. Kowalski, R. Szymanski and S. Penczek, Macromol. Rapid Commun., 1997, 18, 325–333 CrossRef CAS.
  29. Pm is the probability of meso linkages, according to [mmm] = Pm2 + (1 − Pm)Pm/2, [mmr] = [rmm] = (1 − Pm)Pm/2, [rmr] = (1 − Pm)2/2, [mrm] = [(1 − Pm)2 + Pm(1 − Pm)]/2.
  30. P. R. Blakemore, C. Kilner, N. R. Norcross and P. C. Astles, Org. Lett., 2005, 7, 4721–4724 CrossRef CAS PubMed.
  31. M. T. Raisanen, H. Korpi, M. R. Sundberg, A. Savin, M. Leskela and T. Repo, Inorg. Chim. Acta, 2013, 394, 203–209 CrossRef CAS PubMed.
  32. A. Amgoune, C. M. Thomas, T. Roisnel and J. F. Carpentier, Chem.–Eur. J., 2006, 12, 169–179 CrossRef CAS PubMed.
  33. B. M. Chamberlain, M. Cheng, D. R. Moore, T. M. Ovitt, E. B. Lobkovsky and G. W. Coates, J. Am. Chem. Soc., 2001, 123, 3229–3238 CrossRef CAS PubMed.
  34. D. R. Witzke, R. Narayan and J. J. Kolstad, Macromolecules, 1997, 30, 7075–7085 CrossRef CAS.
  35. C. J. Whiteoak, N. Kielland, V. Laserna, E. C. Escudero-Adan, E. Martin and A. W. Kleij, J. Am. Chem. Soc., 2013, 135, 1228–1231 CrossRef CAS PubMed.

Footnote

† Electronic supplementary information (ESI) available: Calculation of the entropy and enthalpy difference between homo-propagation and cross-propagation. See DOI: 10.1039/c4ra11126e
This journal is © The Royal Society of Chemistry 2014
Click here to see how this site uses Cookies. View our privacy policy here.