Development of Photoreactive Degradable Branched Polyesters with

Mar 6, 2009 - Department of Applied Chemistry, Graduate School of Engineering, Osaka ..... for pure PLLA were 37 MPa, 5.3%, and 990 MPa, respectively...
0 downloads 0 Views 3MB Size
766

Biomacromolecules 2009, 10, 766–772

Development of Photoreactive Degradable Branched Polyesters with High Thermal and Mechanical Properties Tran Hang Thi,†,‡ Michiya Matsusaki,† and Mitsuru Akashi*,† Department of Applied Chemistry, Graduate School of Engineering, Osaka University, 2-1 Yamada-oka, Suita 565-0871, Japan, and Faculty of Technology of Organic Chemistry, College of Chemistry, Ministry of Industry and Trade, Tien Kien, Lam Thao, Phu Tho, Vietnam Received October 22, 2008; Revised Manuscript Received January 26, 2009

Branched polyesters composed of poly(L-lactide)s (PLLAs) and 3,4-dihydroxycinnamic acid (DHCA) were obtained by the thermal melt-polycondensation of 3,4-diacetoxycinnamic acid (DACA)-terminally conjugated PLLAs (DACA-PLLAs). The chemical structures of the DHCA/PLLA polyesters were confirmed by FT-IR and 1H NMR measurements. All of the polyesters showed high photoreactivities independent of LLA content. Furthermore, the polyesters had high solubility even after UV irradiation. Interestingly, the thermal stability and mechanical properties of PLLA were improved depending on the incorporation ratio of the DHCA moiety into the backbone. The hydrolysis speed of the DHCA/PLLA polyesters decreased upon increasing the composition of LLA, but the pure PLLA presented the highest hydrolysis speed. Furthermore, all of the polyesters showed faster hydrolysis after UV irradiation as compared to the uncross-linked samples. These novel branched DHCA/PLLA polyesters with high thermal stabilities, mechanical properties and photocontrollable degradability may be useful as functional degradable polyester for the environmental and biomedical fields.

Introduction Many investigations on the synthesis, properties, and degradability of biodegradable polymers have been reported over the past decades. In particular, aliphatic polyesters such as PLLA have attracted much attention as biomedical materials and biodegradable plastic materials because they have excellent biocompatibility and suitable physicochemical properties, are available from renewable sources, and degrade completely to nontoxic water and carbon dioxide.1 However, there is still a great need to improve other properties as thermal and mechanical properties for increasing the scopes of biomedical or environmental plastic fields. One of the most effective methods for enhancing the thermal and mechanical properties of PLLA is its copolymerization with other monomers or polymers. Recently, to improve the properties of PLLA, the copolymerization of LLA or PLLA with other monomers or polymers such as 4-hydroxycinnamic acid (4HCA),2 ricinoleic acid,3 glycolic acid,4 polyethyleneglycol,5 and poly -caprolactone6 have been reported. However, although copolymers composed of LLA and aromatic moieties showed high thermal and mechanical properties, their degradabilities and solubility were limited.2,7-11 On the other hand, LLA and aliphatic copolymers showed high solubility and degradabilities, but their thermal and mechanical properties were poor.3-6 Therefore, novel LLA-based copolymers with both high thermal properties and solubility were designed for the biomedical and environmental fields. We have reported coumaric acid derivative homopolymers and copolymers possessing liquid crystal phases, photoreactivities, degradabilities, and cell compatibilities.12-18 The coumaric acid derivative 3,4-dihydroxycinnamic acid (DHCA) has photoreactive, biometabolizable, and nontoxic properties. We also reported that the various properties of brittle 4HCA homopoly* To whom correspondence should be addressed. Tel.: +81-6-6879-7356. Fax: +81-6-6879-7359. E-mail: [email protected]. † Osaka University. ‡ College of Chemistry, Ministry of Industry and Trade.

mer could be improved by copolymerization with DHCA, such as the mechanical and thermal properties, solubility, cell compatibility, and degradability.14 Furthermore, it is known that biodegradable polymers containing unsaturated groups have great potential for various applications.19,20 In a previous study, we successfully performed a terminal conjugation of 3,4-diacetoxycinnamic acid (DACA) to PLLAs and obtained DACA-terminally conjugated PLLAs (DACAPLLAs).15 The thermal properties of these photoreactive DACAPLLAs were significantly improved; in particular, the 10% weight-loss temperature (T10) showed an increase of over 100 °C, as compared to PLLA of the same molecular weight. However, the crystallinities and solubility of the PLLAs were well-maintained after the conjugation of DACA. Because these DACA-PLLAs have reactive groups at both end chains, it is expected that photoreactive, branched polyesters based on PLLA can be obtained by melt-polycondensation. In this paper, we performed the melt-polycondensation of DACA-PLLA at various composition ratios of DACA and PLLA. The obtained branched polyesters presented high solubility even after the photoreaction of the DHCA moiety via a [2 + 2] cycloaddition. The DHCA/PLLA polyesters showed high thermal stabilities depending on the composition ratio of DHCA and the photocontrollable degradabilities. Branched DHCA/ PLLA polyesters will be useful as functional degradable polyesters.

Experimental Section Materials. L-Lactide (LLA; Tokyo Chemical Industry (TCI)) was recrystallized from ethyl acetate and then dried under reduced pressure at room temperature for 24 h. L-Lactic acid (Tokyo Chemical Industry (TCI)) was used as received. Thionyl chloride (SOCl2), 3,4-dihydroxycinnamic acid (DHCA), acetic anhydride (Ac2O), and sodium acetate (AcONa; Wako Pure Chemical Industries, Ltd.) were used without further purification. Pure PLLA (Mn ) 54720, Mw/Mn ) 1.9, determined by gel-permeation chromatography (GPC) using a HLC 8120 GPC

10.1021/bm801203g CCC: $40.75  2009 American Chemical Society Published on Web 03/06/2009

Photoreactive Degradable Branched Polyesters

Biomacromolecules, Vol. 10, No. 4, 2009

767

Scheme 1. Synthesis of the DHCA/PLLA Polyesters

system with an R-M column (TOSOH Co., Ltd.) in tetrahydrofuran (THF) with polystyrene standards) was purchased from Polysciences, Inc. and used without further purification as a comparative experimental reagent. Synthesis of DHCA/PLLA Polyesters. The synthesis methods for DACA-PLLAs have been reported previously.15 DHCA/PLLA polyester with LLA/DHCA at a ratio of 51/1 unit/unit was prepared as follows. A total of 9 g of DACA-PLLA49 and 90 mg of AcONa (1 wt % of DACA-PLLA49) as a transesterification catalyst were placed into a round-bottom flask. To remove the moisture and residual air from the flask, a nitrogen purge was performed five times at room temperature, and the flask was depressurized for 1 h. The flask was then immersed in an electrostatically preheated silicon oil bath at 160 °C. The polycondensation was performed at 160 °C for 4 h in vacuo. Because the DHCA has a cinnamoyl group, to avoid any light response, the polymerization was carried out under darkened conditions. The molten mixture gradually became more viscous upon increasing reaction time. After the reaction, the product was removed from the flask, dissolved in dichloromethane (DCM), and purified by dialysis (using Mw ) 10.000 of dialysis membrane) in DCM over 4 days. The product was then reprecipitated over methanol and dried under reduced pressure at room temperature for 36 h. The yield was calculated at 85 wt %. The other DHCA/PLLA polyesters were prepared by an analogous procedure. The structures of the DHCA/PLLA polyesters were determined by their FT-IR spectra using a Perkin-Elmer Spectrum 100 FT-IR spectrometer (Japan) and their 1H NMR spectra using a JNM-GSX-400 spectrometer (400 MHz; JEOL, Japan) in chloroform-d. The average molecular weights of the DHCA/PLLA polyesters were determined by GPC using a HLC 8120 GPC system with an R-M column (TOSOH Co., Ltd.) in THF and then calibrated with polystyrene standards with a flow rate of 0.6 mL min-1 at 40 °C. Thermal Properties. The thermal properties were analyzed by differential scanning calorimetry (DSC; EXSTAR6100, Seiko Instruments Inc.) and thermogravimetric analysis (TGA; SSC/5200 SII Seiko Instruments Inc.). Both the heating and the cooling rates were 10 °C min-1, and the temperature ranged from 20 to 200 °C for the DSC measurements. The glass transition temperature (Tg) value, crystallizing temperature (Tc), and melting temperature (Tm) were obtained from the DSC curves of the second heating cycle. The thermal degradation behavior of the samples was observed from the TGA curves by heating from 100 to 500 °C at a rate of 20 °C min-1 under a nitrogen atmosphere with a flow rate about 200 mL min-1. Crystallinity. Samples at a concentration of 30 mg mL-1 in chloroform were cast onto glass slides with a volume of 0.1 mL. They were then dried at room temperature for 24 h and measured before and after annealing at 100 °C for 1 h by a wide-angle X-ray diffraction (WAXD) measurement (X-ray diffractometer (RINT UltraX18), equipped with a scintillation counter) using Cu KR radiation (40 kV, 200 mA; wavelength ) 1.5418 Å) and observed with a crossed-polarizing microscope. Photoreactivity. DHCA/PLLA polyesters with LLA/DHCA ratios of 16, 35, 51, 80, and 116 were dissolved in DMF to concentrations of 0.1, 0.23, 0.31, 0.53, and 0.64 mg mL-1, respectively, with the same absorbance, that is, the same numbers as the DHCA moiety. A glassfiltered high-pressure Hg Lamp (λ > 280 nm, 56 mW cm-2) using a

Supercure-352S-UV Lightsource (SAN-EI ELECTRIC Co., Ltd.) was irradiated. The time course of the photoreaction conversion was monitored by UV-visible absorption spectroscopy of the polyester solutions using a HITACHI U3010 spectrophotometer. Mechanical Properties. Film Preparation. A total of 100 mg of each polyester was dissolved in DCM to a concentration of 15 mg mL-1. The solutions were then cast on 5 cm diameter Teflon dishes, and the solvent was allowed to evaporate at room temperature overnight. All obtained films were further dried under reduced pressure at room temperature for 2 days to eliminate any residual solvent. The thickness of the films was about 30 µm. Tensile Test. Tensile specimens were cut from the cast films with dimensions of 30 × 5 × 0.03 mm. The tensile strength, breaking elongation, and tensile modulus with or without UV irradiation at λ > 280 nm for 2 h were determined using a tensile tester (AGS-H/EZT Test Series, Shima Ritsu Co. Ltd., Japan). The maximum load was 50 N, and the cross-head speed was 1 mm min-1. Degradation Studies. The accelerated hydrolysis test of each cast film (10 × 25 × 0.08 mm) with or without UV irradiation was performed in 45 mL (1.5 mL mg-1) of potassium chloride-sodium hydroxide buffered solution (pH ) 12) at 60 °C for predetermined periods of time. The hydrolytically degraded films were washed thoroughly with distilled water at room temperature, followed by drying under reduced pressure for 2 days. The experimental weight-loss values represent the averages of measurements from two replicate specimens. The percentage weight remaining of the hydrolytically degraded films (Wremaining) was calculated using the film weights before (Wbefore) and after (Wafter) hydrolytic degradation

Wremaining(wt%) ) 100Wafter/Wbefore Morphology (SEM). The morphological evaluation of the DHCA/ PLLA polyester films before and after hydrolysis was performed using a JEOL JSM-6701F scanning electron microscope (JEOL, Japan). Contact Angle. The contact angle measurements were performed on the cast films using a DropMaster 100 (Kyowa Interface Science Co., Ltd.) with distilled water droplets (0.5 µL size). Several drops were placed on the film surface, and the contact angle was measured for each spot after contacted for 5 s. The average value was obtained thereafter.

Results and Discussion Synthesis of DHCA/PLLA Polyesters. DACA-PLLAs were synthesized as described in a previous paper.15 The DHCA/ PLLA polyesters were obtained by the thermal melt-polycondensation of DACA-PLLAs using sodium acetate as a transesterification catalyst at 160-170 °C for 4 h in vacuo (Scheme 1). The viscosity of the DHCA/PLLA polyesters in the flask increased with reaction time. After 4 h of reaction, an ochre colored powder was obtained at high yields (Table 1), and the color became deeper in proportion to the increasing DHCA content. The chemical structures of the DHCA/PLLA polyesters were analyzed by FT-IR spectroscopy. All FT-IR spectra of the

768

Biomacromolecules, Vol. 10, No. 4, 2009

Thi et al.

Table 1. Synthetic Conditions and Molecular Weights of DHCA/PLLA Polyestersa samples

DACA-PLLAb (g)

temperature (°C)

LLA/DHCAc (unit/unit)

Mnd (g/mol)

Mwd (g/mol)

Mw/Mnd

yielde (wt %)

nf

DHCA/PLLA16 DHCA/PLLA35 DHCA/PLLA51 DHCA/PLLA80 DHCA/PLLA116

6 10 9 9 7

160 160 160 170 170

16/1 35/1 51/1 80/1 116/1

18490 34100 45750 61110 62720

41410 66380 118670 180080 212000

2.2 1.9 2.6 2.9 3.3

87 85 85 86 91

13 11 12 10 8

a The polymerization of DHCA/PLLA polyesters was carried out in the presence of sodium acetate as a transesterification catalyst (1 wt % of DACAPLLA) with mechanical stirring for 4 h. b Synthetic method of DACA-PLLA was reported at previous study. c The ratio of LLA/DHCA was estimated by 1H NMR. d The molecular weight and distribution were estimated by GPC in THF with polystyrene standards. e The yield is result after the purification. f Number-average repeating unit of DACA-PLLA was determined by the ratio of Mn of DHCA/PLLA polyester and corresponding DACA-PLLA.

Figure 1. Representative 1H NMR spectrum of DHCA/PLLA16 polyester in CDCl3.

DHCA/PLLA polyesters exhibited a characteristic carbonyl (CdO) stretching band of the ester group at 1750 cm-1, a CdC stretching band of the cinnamoyl group at 1630 cm-1, and a methyl (CH3) stretching band from PLLA at 1450 cm-1. The intensity ratio of the CdC stretching peak of the cinnamoyl group to the CH3 stretching peak of PLLA increased with increasing DHCA content in the DHCA/PLLA polyesters. The presence of the CdC stretching peak suggested remaining CdC bond from the cinnamoyl, even after the high-temperature polymerization.22 The chemical structures of the obtained DHCA/PLLA polyesters were also confirmed by 1H NMR measurements (Figure 1). The 1H NMR spectra of all DHCA/ PLLA polyesters showed multiple peaks at 6.2-8.0 ppm assigned to phenylenevinylene protons: one peak at 2.3 ppm assigned to the methyl proton of the acetyl group, one peak at 5.2 ppm assigned to the C-H proton of PLLA, and one peak at 1.6 ppm assigned to the methyl proton of PLLA, respectively. The composition ratios of L-lactic acid (LLA) and DHCA in the DHCA/PLLA polyesters were calculated by the integral of the peak area ratio of the vinylene proton in DHCA at 6.2 ppm and the C-H proton in PLLA at 5.2 ppm (Table 1). The calculated LLA unit from the 1H NMR spectra agreed well with that in the starting material, DACA-PLLA, indicating the stability of DACA-PLLA under the polymerization temperatures. The ratio of the integrated value of the acetyl and methyl protons in the DHCA/PLLA polyesters was smaller than that in DACA-PLLAs, suggesting that the DHCA/PLLA polyesters were successfully polymerized. The average molecular weights of the DHCA/PLLA polyesters were determined by GPC in THF, relatived to the polystyrene standards (Table 1). All polyesters showed a single GPC peak at a high molecular weight (Mn ) 18190-62720),

indicating a successful polymerization. The polymerization degree of DACA-PLLAs was approximately 10, as calculated from the Mn of the DHCA/PLLA polyester and DACA-PLLA with the same content of DHCA. Solubility. The solubility of the DHCA/PLLA polyesters was evaluated in different solvents at the concentration of 1 mg mL-1. The DHCA/PLLA polyesters showed high solubility in various organic solvents; in particular, the polyesters dissolved easily in chlorinated solvents such as chloroform or dichloromethane (DCM), and the obtained solutions were clear. They also dissolved in tetrahydrofuran (THF) and aprotic amidic solvents, and their solubility was higher than DACA-PLLAs with the same DACA content. Moreover, DHCA/PLLA16, in which the composition ratio of LLA/DHCA was 16/1, could dissolve in acetone. It is known that the solubility of a branched polymer is higher than that of the corresponding linear polymer.21 Therefore, the higher solubility of the DHCA/PLLA polyesters is probably due to their branched structure. Thermal Properties. It has been reported that there is an improvement in the thermal properties of PLLA by copolymerization with aromatic moieties.2,7-11,15 In this study, the thermal properties of the DHCA/PLLA polyesters were analyzed by DSC and TGA measurements (Figure 2). Although the Tg and Tc of pure PLLA were 47 and 98 °C, respectively, the Tg and Tc of the DHCA/PLLA polyesters ranged from 56 to 66 °C and from 110 to 137 °C, respectively, and these values increased with an increasing composition ratio of DHCA (Table 2). The elevation of the Tg value is probably due to the incorporation of a grid aromatic structure into the polymer chain, and the hindrance to chain movement. The 10% weight-loss temperature (T10) of the pure PLLA was 291 °C, but the DHCA/

Photoreactive Degradable Branched Polyesters

Biomacromolecules, Vol. 10, No. 4, 2009

769

Figure 2. (a) Representative DSC curves upon a second heating of the DHCA/PLLA polyesters and pure PLLA at 10 °C min-1, and (b) TG curves of pure PLLA and DHCA/PLLA35 at 20 °C min-1. Table 2. Thermal Properties of DHCA/PLLA Polyesters samples pure PLLA DHCA/PLLA16 DHCA/PLLA35 DHCA/PLLA51 DHCA/PLLA80 DHCA/PLLA116

LLA/DHCA (unit/unit) Tga (°C) Tma (°C) Tca (°C) T10b (°C) 100/0 16/1 35/1 51/1 80/1 116/1

47 66 58 60 57 56

171 130 152 155 162 163

98 NDc 131 137 124 110

291 333 349 332 332 324

a The Tg, Tm, and Tc were measured by DSC upon a second heating (10 °C min-1), the “Tc” means the crystallizing temperature. b The 10% weight-loss temperatures, T10, were measured by TGA under nitrogen (20 °C min°1). c ND ) not determined.

PLLA polyesters showed a higher T10 of about 332 to 349 °C (Table 2). These results indicated that the thermal properties of PLLA were improved via the incorporation of an aromatic DHCA moiety into the PLLA backbone. Furthermore, these thermal properties of the DHCA/PLLA polyesters were completely maintained even after the [2 + 2] cycloaddition formation by UV irradiation. Crystallinity. The crystallinities of the DHCA/PLLA polyesters were analyzed by WAXD measurements (Figure 3). The results from all DHCA/PLLA polyesters without annealing showed amorphous phases. When they were annealed at 100 °C for 1 h and gradually cooled at room temperature, some peaks were observed at 2θ ) 14.7, 16.6, 18.9, and 22.4° (θ ) diffraction angle), corresponding to the spacings of 5.9, 5.3, 4.6, and 3.9 Å, respectively, except for the DHCA/PLLA16 polyester. This is consistent with the peaks at 15, 16, 18.5, and 22.5° reported by Ikada et al.23 Furthermore, when the composition ratio of LLA/DHCA was in the range from 35 to 80 units, the exothermic peak and a small endothermic peak were observed prior to the melting peak in DSC measurements, indicating the formation of an R + R′-phase with a 103 helical chain conformation.24,25 The DSC measurements of the DHCA/ PLLA116 polyester showed a similar pattern to that of pure PLLA, suggesting an R-phase with a 103 helical chain conformation. DACA-PLLA16 showed high crystallinity after annealing,15 but the DHCA/PLLA16 polyester showed an amorphous conformation (Figure 3c), possibly due to the highly branched

Figure 3. Representative WAXD patterns and crossed polarizing microscopic images of (a) pure PLLA, (b) DHCA/PLLA35, and (c) DHCA/PLLA16 after annealing at 100 °C for 1 h and gradually cooling at room temperature.

structure inhibiting the molecular packing of the short PLLA chain. We also investigated the crystal morphologies of the DHCA/PLLA polyesters by observation using crossed-polarizing microscopy. These results indicated that the crystal domains of the DHCA/PLLA polyesters were similar to those of pure PLLA. These results suggested that the crystallinity of PLLA was maintained after incorporation of the DHCA moiety into the PLLA chain at over 35 units. Photoreactivities. The cinnamoyl group is well-known to undergo [2 + 2] cycloaddition, leading to the formation of a cyclobutane ring upon UV irradiation (Figure 4d).12-15,26-32 In a previous report, we confirmed that DACA-PLLAs showed high photoreactivities, and their photoreactivities were independent of the chain length of the PLLA.15 Therefore, in this study, the DHCA/PLLA polyesters were expected to have photoreactivities. The time-dependent UV absorption changes of the DHCA/PLLA35 polyester dissolved in DMF are shown in Figure 4a. The maximal absorption peak at λmax ) 294 nm

770

Biomacromolecules, Vol. 10, No. 4, 2009

Thi et al.

Figure 4. (a) UV absorption change of DHCA/PLLA35 polyester in DMF during UV irradiation at λ > 280 nm for 100 min, (b) maximal absorption (λmax ) 294 nm) change of various DHCA/PLLA polyesters, (c) FT-IR spectra of DACA/PLLA35 polyester cast film before and after UV irradiation at λ > 280 nm for 2 h, and (d) photoreaction scheme of the DHCA moiety undergoing UV irradiation at λ > 280 nm. Table 3. Mechanical Properties of DHCA/PLLA Polyesters samples

LLA/DHCA (unit/unit)

tensile strength (MPa)

breaking elongation (%)

tensile modulus (MPa)

pure PLLA DHCA/PLLA16 DHCA/PLLA35 DHCA/PLLA51 DHCA/PLLA80 DHCA/PLLA116

100/0 16/1 35/1 51/1 80/1 116/1

37 ( 3 NDa 23 ( 3 (37 ( 1)b 38 ( 2 (43 ( 2) 42 ( 2 (43 ( 1) 42 ( 3 (47 ( 1)

5.3 ( 0.4 NDa 2.7 ( 0.4 (4.9 ( 0.1) 3.9 ( 0.5 (5.2 ( 0.9) 4.1 ( 0.2 (5.2 ( 0.4) 4.1 ( 0.2 (5.2 ( 0.4)

990 ( 173 NDa 990 ( 135 (900 ( 121) 980 ( 105 (910 ( 66) 1070 ( 140 (1010 ( 112) 1130 ( 184 (1020 ( 1)

a

ND ) not determined.

b

The data in parentheses were obtained after UV irradiated at λ > 280 nm for 2 h.

decreased with increasing UV irradiation time at λ > 280 nm, and the other DHCA/PLLA polyesters also showed the same behaviors (Figure 4b). All DHCA/PLLA polyesters showed an approximately 80% conversion of the [2 + 2] cycloaddition after UV irradiation at λ > 280 nm for 100 min, indicating that their photoreactivities were independent of the chain length of PLLA and the branch density. The photoreaction of DHCA/ PLLA was also investigated by FT-IR measurements. Figure 4c shows the FT-IR spectra of the DHCA/PLLA35 polyester cast films before and after UV irradiation at λ > 280 nm for 2 h. The intensity of the CdC stretching band of the cinnamoyl group at 1630 cm-1 decreased after UV irradiation, indicating that the [2 + 2] cycloaddition had occurred. The high solubility of the DHCA/PLLA polyesters was maintained, even after UV irradiation. Mechanical Properties. The mechanical properties of the DHCA/PLLA polyesters were investigated with a tensile tester. Table 3 summarizes the mechanical properties of the various DHCA/PLLA polyesters with or without UV irradiation at λ > 280 nm for 2 h. The tensile strength and other data were not obtained from the DHCA/PLLA16 polyester because it was very brittle. The tensile strength, breaking elongation, and tensile modulus of the other DHCA/PLLA polyesters without UV irradiation ranged from 23 to 42 MPa, from 2.7 to 4.1%, and from 990 to 1130 MPa, respectively. Incidentally, the values for pure PLLA were 37 MPa, 5.3%, and 990 MPa, respectively. We evaluated the mechanical properties of the DHCA/PLLA polyesters after UV irradiation, and the obtained tensile strength

increased to 37 × 47 MPa. The breaking elongation also showed an increase, but the tensile modulus showed a slight decrement, probably due to inter- or intramolecular photocross-linkings of the DHCA moieties in the branched polyesters following UV irradiation. These results suggested that the mechanical properties of the DHCA/PLLA polyesters were superior to pure PLLA, and they can be controllable by UV irradiation. Degradation Behavior. Accelerated hydrolysis experiments on the DHCA/PLLA polyester films were performed in potassium chloride-sodium hydroxide buffered solution at 60 °C (pH ) 12). Figure 5 shows the weight remaining percentage of pure PLLA and the various DHCA/PLLA polyesters with or without UV irradiation at λ > 280 nm for 2 h. The hydrolysis rate of the DHCA/PLLA polyesters increased with increasing DHCA content but was slower than pure PLLA. To determine the effect of the DHCA moiety on the hydrolysis rate, we evaluated the surface contact angle of all DHCA/PLLA polyesters cast films, but larger differences were not observed (Figure 6). Nagata et al. reported that the degradation rate of copolyesters composed of 4HCA and PLLA following UV irradiation was slower than that without UV irradiation.2 However, in our previous study, we found that the degradation rates of 4HCA/ DHCA copolymers were enhanced by UV irradiation at λ > 280 nm.14 In this study, the same phenomenon was observed, as shown in Figure 5b. The DHCA/PLLA polyester films following UV irradiation at λ > 280 nm for 2 h presented faster hydrolytic degradation as compared to the films without UV irradiation, and all films were degraded completely within 15

Photoreactive Degradable Branched Polyesters

Biomacromolecules, Vol. 10, No. 4, 2009

771

Figure 5. Degradation behavior of DHCA/PLLA polyesters. The weight remaining percentage of DHCA/PLLA polyesters cast films (ratio of LLA/DHCA: (9) pure PLLA, (3) 16, (O) 35, (4) 51, (b) 80, and (0) 116) in a potassium chloride-sodium hydroxide buffered solution at 60 °C (pH ) 12) (a) without or (b) with UV irradiation at λ > 280 nm for 2 h.

Figure 6. Contact angle images of pure PLLA and various DHCA-PLLA polyesters (a) without or (b) with UV irradiation at λ > 280 nm for 2 h.

Figure 7. Degradation behavior in SEM images of DHCA/PLLA35 polyester cast films (a) without or (b) with UV irradiation at λ > 280 nm for 2 h.

days. It is likely that cyclobutane formation can attenuate the ester carbonyl conjugation with the phenylenevinylene, making it more polar and easier to degrade.14 The SEM analysis provided further information on the morphology of the obtained DHCA/PLLA35 polyester films with or without UV irradiation during hydrolysis (Figure 7). The DHCA/PLLA polyester films displayed a smooth surface independent of UV irradiation. However, the hydrolyzed surfaces with UV irradiation seemed to be more porous as compared to the surfaces without UV irradiation.

Conclusion Branched PLLA/DHCA polyesters were successfully synthesized by the thermal melt-polycondensation of DACA-PLLAs using sodium acetate as a transesterification catalyst, with high yields and high molecular weights. All DHCA/PLLA polyesters showed high solubility even after UV irradiation. DACA/ PLLA16 showed an amorphous property, however, the higher LLA/DHCA ratios over 35 showed high crystallinity, the same as pure PLLA. Interestingly, the thermal stability and mechanical properties of PLLA were improved depending on the incorpora-

772

Biomacromolecules, Vol. 10, No. 4, 2009

tion ratio of the DHCA moiety into the backbone. The hydrolysis speed of the DHCA/PLLA polyesters decreased with an increasing composition of LLA. The hydrolysis speed of the DHCA/PLLA polyesters could be controllable by UV crosslinking of DHCA units. The copolymerization with DHCA will be useful as one of the improving methods for mechanical and thermal properties of PLLA, and application of obtained branched DHCA/PLLA is expected in biomedical and environmental fields. Acknowledgment. This research was supported by a Osaka University 21st Century COE Program “Center for Integrated Cell and Tissue Regulation”. We acknowledge Prof. H. Uyama of the Department of Applied Chemistry, Graduate of Engineering, Osaka University, for the TGA measurements.

References and Notes (1) Kaplan, D. L. Biopolymers from Renewable Resources; Springer Verlag: New York, 1998; p 350. (2) Nagata, M.; Sato, Y. Polym. Int. 2005, 54, 386–391. (3) Slivniak, R.; Domb, A. J. Macromolecules 2005, 38, 5545–5553. (4) Ikada, Y.; Tsuji, H. Macromol. Rapid Commun. 2000, 21, 117–132. (5) Li, F.; Li, S.; Ghzaoui, A. E.; Nouailhas, H.; Zhuo, R. Langmuir 2007, 23, 2778–2783. (6) Jeon, O.; Lee, S. H.; Kim, S. H.; Lee, Y. M.; Kim, Y. H. Macromolecules 2003, 36, 5585–5592. (7) Chen, Y.; Wombacher, R.; Wendorff, J. H.; Visjager, J.; Smith, P.; Greiner, A. Biomacromolecules 2003, 4, 974–980. (8) Chen, Y.; Wombacher, R.; Wendorff, J. H.; Greiner, A. Chem. Mater. 2003, 15, 694–698. (9) Li, B.-H.; Yang, M.-C. Polym. AdV. Technol. 2006, 17, 439–443. (10) Chen, Y.; Wombacher, R.; Wendorff, J. H.; Visjager, J.; Smith, P.; Greiner, A. Biomacromolecules 2003, 4, 974–980. (11) Haderlein, G.; Petersen, H.; Schmidt, C.; Wendorff, J. H.; Schaper, A.; Jone, D. B.; Visjager, J.; Smith, P.; Greiner, A. Macromol. Chem. Phys. 1999, 200, 2080–2087. (12) Matsusaki, M.; Kishida, A.; Stainton, N.; Ansell, C. W. G.; Akashi, M. J. Appl. Polym. Sci. 2001, 82, 2357–2364.

Thi et al. (13) Kaneko, T.; Matsusaki, M.; Tran, H. T.; Akashi, M. Macromol. Rapid Commun. 2004, 25, 673–677. (14) Kaneko, T.; Tran, H. T.; Shi, D. J.; Akashi, M. Nat. Mater. 2006, 5, 966–970. (15) Tran, H. T.; Matsusaki, M.; Shi, D. J.; Akashi, M. Chem. Commun. 2008, 3918–3920. (16) Matsusaki, M.; Tran, H. T.; Kaneko, T.; Akashi, M. Biomaterials 2005, 26, 6263–6270. (17) Kaneko, T.; Tran, H. T.; Matsusaki, M.; Akashi, M. Chem. Mater. 2006, 18, 6220–6226. (18) Tran, H. T.; Matsusaki, M.; Shi, D. J.; Kaneko, T.; Akashi, M. J. Biomater. Sci., Polym. Ed. 2008, 19, 75–85. (19) Gimenez, V.; Reina, J. A.; Mantecon, A.; Cadiz, V. Polymer 1999, 40, 2759–2767. (20) Jin, H. J. J. Polym. Sci., Part B: Polym. Phys. 2000, 38, 2240–2246. (21) Yamanaka, K.; Jikei, M.; Kakimoto, M. Macromolecules 2000, 33, 1111–1114. (22) Kimura, K.; Inoue, H.; Kohama, S.-I.; Yamashita, Y.; Sakaguchi, Y. Macromolecules 2003, 36, 7721–7729. (23) Ikada, Y.; Jamshidi, K.; Tsuji, H.; Hyon, S. H. Macromolecules 1987, 20, 904–906. (24) Zhang, J.; Duan, Y.; Sato, H.; Tsuji, H.; Noda, I.; Yan, S.; Ozaki, Y. Macromolecules 2005, 38, 8012–8021. (25) Zhang, J.; Tashiro, K.; Tsuji, H.; Domb, A. J. Macromolecules 2008, 41, 1352–1357. (26) Kawatsuki, N.; Matsuyoshi, K.; Hayashi, M.; Takatsuka, H.; Yamamoto, T.; Sangen, O. Chem. Mater. 2000, 12, 1549–1555. (27) Graley, M.; Reiser, A.; Roberts, A. J.; Phillips, D. Macromolecules 1981, 14, 1752–1757. (28) Egerton, P. L.; Trigg, J.; Hyde, E. M.; Reiser, A. Macromolecules 1981, 14, 100–104. (29) Haddleton, D. M.; Creed, D.; Griffin, A. C. Macromol. Chem. Rapid Commun. 1989, 10, 391–396. (30) Egerton, P. L.; Pitts, E.; Reiser, A. Macromolecules 1995, 28, 1214– 1220. (31) Gupta, P.; Trenor, S. R.; Long, T. E.; Wilkes, G. L. Macromolecules 2004, 37, 9211–9218. (32) Chae, B.; Lee, S. W.; Ree, M.; Jung, Y. M.; Kim, S. B. Langmuir 2003, 19, 687–695.

BM801203G