Thermal and mechanical properties of chemical crosslinked polylactide (PLA)

Thermal and mechanical properties of chemical crosslinked polylactide (PLA)

Polymer Testing 27 (2008) 957–963 Contents lists available at ScienceDirect Polymer Testing journal homepage: www.elsevier.com/locate/polytest Mate...

698KB Sizes 0 Downloads 27 Views

Polymer Testing 27 (2008) 957–963

Contents lists available at ScienceDirect

Polymer Testing journal homepage: www.elsevier.com/locate/polytest

Material Properties

Thermal and mechanical properties of chemical crosslinked polylactide (PLA) Sen-lin Yang, Zhi-Hua Wu*, Wei Yang, Ming-Bo Yang College of Polymer Science and Engineering, State Key Laboratory of Polymer Materials Engineering, Sichuan University, Chengdu, 610065, Sichuan, People’s Republic of China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 June 2008 Accepted 18 August 2008

To improve the thermal stability and mechanical properties of PLA, crosslinking was introduced via chemical treatment of the melt by adding small amounts of crosslinking agent triallyl isocyanurate (TAIC) and dicumyl peroxide (DCP). A series of crosslinked PLA materials with different gel fraction and crosslink density were prepared. The crosslinked PLA samples were characterized by fourier transform infra-red spectrometry (FTIR). The thermal and mechanical properties of samples were also investigated by means of differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), tensile testing and dynamic mechanical analysis (DMA). The results showed that the crosslinking of PLA started at a low content of either TAIC or DCP, resulting in a decrease of crystallinity and a significant improvement of the thermal degradation initiation and completion temperatures, which indicated better thermal stability than neat PLA. Crosslinking was also responsible for the improved tensile modulus and tensile strength. Ó 2008 Elsevier Ltd. All rights reserved.

Keywords: Polylactide (PLA) Chemical crosslinking Mechanical properties Peroxide TAIC Thermal stability

1. Introduction Polylactide (PLA) is a typical biodegradable polyester obtained by synthesis of lactic acid (or lactide), which can be produced from renewable resources such as corn or sugarcane [1–3]. PLA is an enantiomeric polyester including poly(L-lactic acid)(PLLA) and poly(D-lactic acid)(PDLA). The chiral center in the structure allows varied enantiomeric compositions of PLA. With good biodegradability and good processability, PLA was regarded as one of the most promising biodegradable polymers and was expected to substitute some of the non-biodegradable engineering plastics [4,5]. However, the poor heat stability and mechanical properties limited its applications [6–8]. Many technologies, such as annealing, adding nucleating agents [9–12], forming composites with fiber or nanoparticles [13–16], chain extending [17,18] and introducing crosslinking structures were proved effective for enhancing the heat stability or mechanical properties of PLA materials. * Corresponding author. Tel.: þ86 28 85461716; fax: þ86 28 87578533. E-mail address: [email protected] (Z.-H. Wu). 0142-9418/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymertesting.2008.08.009

The crosslinking structures of PLA can be formed by irradiation. g-Irradiation and electron beam irradiation were widely applied to crosslinking of PLA in the presence of a small amount of crosslinking agent triallyl isocyanurate (TAIC) [19–21]. For example, Mitomo et al. reported that PLLA/(3%TAIC) irradiated at 50 kGy reached a 84% gel fraction and showed typical heat stability above the glasstransition temperature after annealing at 90  C for 1 h [22]. Nagasawa et al. further proved that by irradiation with an electron beam, PLLA/3%TAIC showed excellent heat stability which was demonstrated by the retention of original shape at or even higher than 200  C [23]. Nevertheless, in irradiation crosslinking of PLA materials, the radiation apparatus is expensive and the PLA products must be in the form of thin plates so as to get enough energy from the radiation to initiate the crosslinking reactions, which significantly limits the practical application of this method. Chemical crosslinking is another possible way to introduce crosslinking structures in PLA. Some chemical reactions between the crosslinking agent and the polymer chains can be initiated by chemical treatments without irradiation, and modified PLA materials with different gel fraction and

958

S.-lin Yang et al. / Polymer Testing 27 (2008) 957–963

crosslinking density for further processing of PLA products can be obtained. For example, peroxide crosslinking of PLA with dicumyl peroxide (DCP) can form a gel structure. Nijenhuis et al. reported that at high peroxide (DCP) concentrations (13–25 wt%) and high curing temperatures (192  C), a gel fraction of 100% could be determined gravimetrically. However, the high peroxide concentrations resulted in a steep drop of the tensile strength [24]. In addition, a large amount of chloroform solvent was required, and a further separate purification step was necessary. To the best of our knowledge, there are still very few reports on the thermal or mechanical properties of chemical crosslinked PLA materials, so further and deeper studies on the chemical crosslinking of PLA materials are to be expected. In this article we reported our efforts on the chemical crosslinking of a commercial PLA in the presence of small amounts of TAIC and DCP, aiming at improving the thermal stability and mechanical properties. This method is economically advantageous because it was carried out in the melt state with only low amounts of TAIC and DCP, and no extra purification step and special apparatus were necessary. 2. Experimental part 2.1. Materials A commercially available polylactide (PLA) (NatureWorks, PLA 3001D) with a melt flow index (MFI) range from 10 to 30 g/10 min (190  C, 2.16 kg) and a density of 1.24 g/cm3, was purchased from NatureWorks LLC. The triallyl isocyanurate (TAIC) and dicumyl peroxide (DCP) were supplied by Chen Guan Co. (Sichuan, China). 2.2. Preparation of samples PLA samples containing different concentrations of TAIC (0.15 wt%, 0.5 wt%, 1.0 wt%, 3.0 wt%) and DCP (0.2 wt%, 0.5 wt%, 1.0 wt%, 1.5 wt%) were mixed in a Haake melt mixer at 50 rpm, 180  C for 10 min. Then, the samples were hot-pressed at 190  C for 3 min followed by cold-pressing at room temperature (about 20  C) for 3 min to form the sheets with thickness of 0.8 mm. For comparison, PLA without any crosslinking agent was treated with the same procedure. The samples with 0.15 wt% TAIC and 0.5 wt% DCP (PLA-1), 0.5 wt% TAIC and 0.5 wt% DCP (PLA-2), 1.0 wt% TAIC and 1.0 wt% DCP (PLA-3), 3.0 wt% TAIC and 1.0 wt% DCP (PLA-4) were chosen for further characterization, i.e. FTIR, DSC, TGA, DMA and tensile test. 2.3. Infra-red analysis IR spectra of pure PLA and the crosslinked PLA samples were recorded with a FTIR spectrometer (Nicolet-560, America). Before testing, the dry gel component of chemical crosslinked PLA samples was mixed with KBr powder and cold-pressed into a suitable disk for FTIR measurement.

 Gel fraction ð%Þ ¼ Wg =Wo  100

(1)

where Wo is the original weight (dry) of the crosslinked PLA, Wg is the weight remaining (dry gel component) of the crosslinked PLA after being dissolved in chloroform at room temperature for 48 h. Degree of swelling (volume ratio of absorbed solvent to dry gel sample) is calculated using the following Eq. (2):

Degree of swelling ðqÞ ¼



Ws  W g



  Wg rP =rCHCl3 (2)

where Wg is the weight of dry gel component in the crosslinked PLA sample, Ws is the weight of gel component swollen at room temperature for 48 h in chloroform rP and rCHCl3 are densities of PLA and chloroform, respectively. 2.5. Thermal analysis The thermal properties of PLA samples (about 6 mg) were measured by DSC (Netzsch DSC-204F1) using aluminium oxide as the standard. The melting point (Tm), glass-transition temperature (Tg), enthalpy of cold crystallization (DHc) and enthalpy of melting (DHm) of each sample were measured from 15  C to 200  C under nitrogen at a heating rate of 10  C/min. The thermal stability of the samples (about 7 mg) was investigated with a TGA (TA-Q600) under nitrogen from room temperature (about 20  C) to 500  C at a heating rate of 10  C/min. 2.6. Mechanical properties The tensile properties of the samples were measured in accordance with ISO 527 at room temperature (about 20  C) using a tensile tester (Instron-4302). Dynamic mechanical properties were investigated using a dynamic mechanical analyzer DMA (TA-Q800). Samples in the form of strips (20 mm  6 mm  0.8 mm) were measured in tensile mode at a constant frequency of 1.0 Hz as a function of temperature from 20  C to 200  C at a heating rate of 3  C/min under nitrogen flow. 2.7. Morphology observation The fracture surfaces of pure PLA, PLA-3 and PLA-4 samples were studied with a JEOL JSM-5900LV scanning electron microscope (SEM) under an acceleration voltage of 20 kV. Prior to the SEM examination, samples were submerged in liquid nitrogen and broken to expose the internal structure for SEM studies, and all the surfaces were sputtered with gold. 3. Results and discussion 3.1. Chemical crosslinking of PLA

2.4. Measurement of gel fraction and degree of swelling Gel fraction was measured by the weight remaining after dissolving the sample in chloroform using the following Eq. (1):

TAIC, whose structural formula is given in Scheme 1, has been proved to be one of the most effective crosslinking agents for PLA. The double bonds in TAIC can be easily

S.-lin Yang et al. / Polymer Testing 27 (2008) 957–963

TAIC) and 1627 cm1 (C]C of TAIC), respectively, which can be observed in the spectra of PLA-2, PLA-3 and PLA-4. The absorbance intensities of these two new peaks increase with increasing TAIC content, which clearly confirms the chemical crosslinking reaction between PLA and TAIC. Once the DCP was added to the PLA/TAIC melt blends, DCP decomposed into the RO radicals [27]. Then, the RO radicals abstracted hydrogen from PLA, which created the radicals (i.e.—Co) in PLA chains [28–30]. At the same time, the double bonds of allyl groups in TAIC were broken, and two kinds of radicals (number (1) was the usual case, number (2) was the rare case) might be activated [22]. Consequently, the radicals from TAIC combined with —Co groups of PLA, and the chemical crosslinking structure between PLA molecules could take place as shown in Scheme 2. The gel fraction of the chemical crosslinked PLA samples containing different amounts of TAIC as a function of DCP content are shown in Fig. 2. It clearly shows that the crosslinking of PLA starts at a very low concentration of TAIC (0.15 wt%) and DCP (0.2 wt%). The gel fraction of crosslinked PLA increases with increase of TAIC and DCP contents. The crosslink density of PLA samples is indicated by the degree of swelling (DS) and a low degree of swelling implies a high crosslink density. It can be observed in Fig. 3 that the DS decreases with the increase of TAIC and DCP contents, which is also consistent with the increase of the gel fraction.

O CH2

HC H2C O

N C

C

N

N

C

CH2

HC

CH2

O

CH2 HC

CH2

Scheme 1. Structural formula of TAIC.

Transmittance( % )

PLA

PLA-2

PLA-3

3200

1180

1756 1691 1627

PLA-4

3600

959

1800 1600 1400 1200 1000 800

Wavenumbers(cm-1) Fig. 1. FTIR spectra of PLA and chemical crosslinked PLA samples.

broken to produce monomer radicals and then combine with the polymer radicals to form the crosslinking network. Fig. 1 shows the FTIR spectra of the pure and chemical crosslinked PLA samples. The peaks at about 1756 cm1 and 1180 cm1 which belong to the C]O stretching and C–O–C stretching of PLA, respectively, [25,26] are clearly visible in all the IR spectra. After the chemical reaction with TAIC in the presence of DCP, two new peaks appear at about 1691 cm1 (C]O of

3.2. Thermal properties The DSC results of PLA and chemical crosslinked PLA samples are shown in Fig. 4 and Table 1. The glass-transition, cold crystallization and melting can be clearly observed in the curves. It is noted that a double-peak in the cold crystallization of PLA-2 is observed, which may have

O CH2

HC

H2C N O

C

C N

O N CH2 C

HC

CH2

C · CH2 CH CH2 N N CH2 CH (1) (2) C C O O N OR H2C C CH2 H

OR

O

CH2 HC

CH2

H O H O O C C O C C CH3 CH3 n

OR

H O O O C C O C C CH3 CH3

CH3 O

O

H O

C H

H2C N O

O CH3

C N

N CH2 C

CH2

O C C O C C CH3

C

n

CH2

n CH3

C C O C C O

CH2

OR

O H C H

n

CH2

O C H

CH2

Scheme 2. One possibility of reaction scheme for chemical crosslinking of TAIC between two PLA molecules.

960

S.-lin Yang et al. / Polymer Testing 27 (2008) 957–963

90

PLA 60 45 0 wt% TAIC 0.2 wt% TAIC 0.5 wt% TAIC 1.0 wt% TAIC 3.0 wt% TAIC

30 15 0 0.00

0.25

0.50

0.75

1.00

1.25

Endotherm

Gel Fraction (%)

75

PLA-4

1.50

0

resulted from the introduction of two kinds of chain structure, i.e. typically crosslinked PLA (59.14% of Gel fraction) chains with 18.38 of DS and non-crosslinked PLA chains. Such two kinds of molecular chains would reach their maximum rates of cold crystallization at two different temperatures. The peak at the lower temperature corresponds to the non-crosslinked PLA chains with better chain mobility, in favor of chain segments packing into crystalline structure at a lower energy level. On the other hand, the peak at higher temperature corresponds to the typically crosslinked PLA chains and the chain mobility is retarded by the crosslinking structure. If the proportion of crosslinking structure and non-crosslinking structure could not be balanced, or the difference between two kinds of chain structure was not significant, the double-peak would not appear. The cold crystallization of crosslinked samples tends to shift to higher temperature and seems to disappear with increasing TAIC content. This is also caused by the large molecular chain networks with a high crosslink density

0.2 wt% TAIC 0.5 wt% TAIC 1.0 wt% TAIC 3.0 wt% TAIC

Degree of Swelling

30

20

40

60

80

100 120 140 160 180 200

Temperature(°C)

Fig. 2. Gel fraction of chemical crosslinked PLA as a function of DCP concentration.

35

PLA-2 PLA-3

DCP (wt%)

40

PLA-1

25

Fig. 4. The DSC heating curves of PLA and chemical crosslinked PLA samples.

which inhibit chain segment motion for crystallization. As the crosslinking structures interfered with the crystallization process, many imperfect crystallites were formed [21]. Therefore, the cold crystallization of the slightly crosslinked samples (PLA-1, PLA-2) starts at a lower temperature than that of pure PLA, and the crystallinity decreases. The Tm also shifts to lower temperature with the increase of gel fraction and crosslink density, as indicated in Table 1. These results are in good agreement with the DSC heating curves of radiation crosslinked PLA [19–21]. The thermal stability of the crosslinked PLA samples determined by thermogravimetric analysis in nitrogen is shown in Fig. 5, and the thermal degradation curve of pure PLA is also shown for comparison. The thermal degradation of all the samples experiences a one-stage weight loss. The onset temperature of thermal degradation of pure PLA is approximately 280  C, and the degradation completes at about 350  C. The introduction of chemical crosslinking improves the thermal stability of the PLA with an increase of both onset thermal degradation temperature and complete degradation temperature. Especially for PLA-4, the onset degradation and the complete degradation temperatures are improved to approximately 310  C and 375  C, respectively. The characteristic thermal degradation temperatures, including T0.1 and T0.5, defined as the temperature at which

20

Table 1 Thermal properties and the crystallinity of PLA and chemical crosslinked PLA

15

Samples

Tg ( C)

DHc (J/g)

Tm ( C)

DHm (J/g)

c (%)

PLA PLA-1 PLA-2 PLA-3 PLA-4

61.2 61.2 61.9 60.2 59.9

28.17 25.01 26.56 23.35 12.18

170.5 167.5 166.1 156.7 153.8

43.23 35.39 31.78 28.47 14.03

32.02 26.62 23.54 21.09 10.39

10 5 0 0.00

0.25

0.50

0.75

1.00

1.25

1.50

DCP (wt%) Fig. 3. Degree of swelling of chemical crosslinked PLA as a function of DCP concentration.

Abbreviations: Tgdglass-transition temperature determined from the inflection point of the heat flow curve; DHcdenthalpy of the cold crystallization; Tmdtemperature of the melting peak; DHmdthe melting *  100%; DHm * ¼ 135 J/g, the enthalpy; cdcrystallinity; c ¼ DHm/DHm melting enthalpy of 100% crystalline PLA [31].

S.-lin Yang et al. / Polymer Testing 27 (2008) 957–963

100

PLA

Table 3 Mechanical properties of PLA and chemical crosslinked PLA samples

PLA-2

Samples

Tensile strength (MPa)

Tensile modulus (GPa)

Elongation at break (%)

PLA PLA-1 PLA-2 PLA-3 PLA-4

65.78  0.39 73.56  0.51 75.24  0.63 67.86  0.48 65.17  1.14

1.68  0.07 1.74  0.07 1.87  0.08 1.95  0.04 1.99  0.15

8.91  0.44 7.48  0.39 7.40  0.32 3.87  0.21 3.21  0.19

PLA-3 PLA-4

60

40

20

0 200

250

300

350

400

450

500

Temperature(°C) Fig. 5. TGA curves of PLA and chemical crosslinked PLA samples.

10% and 50% mass loss occurs respectively, are summarized in Table 2. It can be easily seen that T0.1 and T0.5 increase with the increase of gel fraction and crosslink density, indicating improvement of thermal stability of PLA. This may also be due to the large molecular chain networks with high crosslink density in the crosslinked PLA which retard degradation.

3.3. Mechanical properties The tensile properties of pure PLA and chemical crosslinked PLA samples are shown in Table 3 and Fig. 6. The introduction of crosslink structure into PLA results in the increase of the tensile modulus and the decrease of elongation at break. This may be attributed to the crosslinking structure which stiffens the PLA material. However, the increase of gel fraction and crosslink density made the crosslinked PLA more brittle. At the same time, it is also noticed that the tensile strength of the crosslinked samples firstly increases and then decreases with the gel fraction and crosslink density increasing. The highest tensile strength of crosslinked PLA is obtained for PLA-2 with 59.14% Gel fraction and 18.38 DS. At higher crosslinking level, the crosslinked PLA is difficult to melt during hot-pressing, and a homogeneous morphology could not be formed in PLA-3 and PLA-4 samples, so that a lot of defects exist in these samples, as

Table 2 TGA results for PLA and chemical crosslinked PLA samples Samples

T0.1 ( C)

T0.5 ( C)

DT0.5 ( C)

PLA PLA-2 PLA-3 PLA-4

292 298 304 324

325 336 338 352

– 11 13 27

T0.1 ¼ temperature at 10% mass loss; T0.5 ¼ temperature at 50% mass loss; DT0.5 ¼ temperature difference at 50% mass loss between the crosslinked sample and the neat PLA.

suggested in Fig. 7. This directly deteriorates the mechanical properties of these crosslinked samples. Fig. 8 shows the storage modulus (E0 ) of PLA and chemical crosslinked PLA samples as a function of temperature. The curve of pure PLA demonstrates that PLA exhibits glassy, glass-transition, cold crystallization and liquid-flow behaviors. The curves of partially crosslinked samples almost have the same shape as pure PLA below the melting temperature (about 170  C). The increase of gel fraction and crosslinking density results in increased storage modulus in the glassy state and decreased cold crystallinity. Meanwhile, the onset temperature of cold crystallization tends to shift to higher temperature. These results are consistent with the DSC results. It can also be observed that E0 of pure PLA drops dramatically (almost to 0 Pa) after the melting temperature, while the E0 of PLA-2 and PLA-3 firstly drops and then decreases slightly with a gentle curve up to 200  C, showing further evidence for better thermal stability than the non-crosslinked PLA sample. This also results from the crosslinking networks which prevent the mobility of the chain segments to some extent after melting. The crosslinked PLA samples showed pronounced heat shrinkage behavior after the glass-transition temperature under heating, which was also observed in the radiation crosslinked PLA films [22]. Moreover, the degree of shrinking increased with increase of the gel fraction and crosslink density. For PLA-4, the high degree of shrinkage and many defects directly resulted in fracture near its melting point (about150  C).

80

PLA-2 PLA-4

70

PLA-1

PLA-3

PLA

60

Stress (MPa)

Weight Percent (%)

80

961

50 40 30 20 10 0

0

1

2

3

4

5

6

7

8

9

10

Strain (%) Fig. 6. Stress–strain curves of PLA and chemical crosslinked PLA samples.

962

S.-lin Yang et al. / Polymer Testing 27 (2008) 957–963

Fig. 7. Scanning Electron Micrograph (SEM) photos of the fracture surfaces of the PLA and chemical crosslinked samples.

4. Conclusions 10

10

10

PLA PLA-2 PLA-3 PLA-4

9

E (Pa)

108 107 106 105 104 -20

0

20

40

60

80 100 120 140 160 180 200

Temperature (°C) Fig. 8. Dynamic mechanical analysis of PLA and chemical crosslinked PLA samples.

Crosslinking structures can be effectively introduced into PLA by the initiation of DCP in the presence of a small amount of crosslinking agent (TAIC). The thermal and mechanical properties of chemical crosslinked PLA have been mainly determined by the gel fraction and crosslink density. With the introduction of crosslinking structure, the tensile strength was improved. The Tm shifted to lower temperature and the thermal stability was improved, which was indicated by TGA and DMA results. However, the increase of brittleness with the introduction of a highly crosslinked structure is a problem needing to be overcome and will be considered in our future work.

Acknowledgements The authors gratefully acknowledge the financial support of this work by National Natural Science

S.-lin Yang et al. / Polymer Testing 27 (2008) 957–963

Foundation of China (No. 20734005). The authors would also like to express their sincere thanks to the Analytical and Testing Center of Sichuan University for the assistance of SEM testing. References [1] A.J. Nijenhuis, D.W. Gripma, A.J. Pennings, Macromolecules 25 (1992) 6419. [2] H. Tsuji, H. Daimon, K. Fujie, Biomacromolecules 4 (2003) 835. [3] E.T.H. Vink, K.R. Rabago, D.A. Glassner, P.R. Gruber, Polym. Degrad. Stab. 80 (2003) 403. [4] G. Kale, R. Auras, S.P. Singh, R. Narayan, Polymer Test. 26 (2007) 1049. [5] D. Garlotta, J. Polym. Environ. 9 (2002) 63. [6] J.R. Dorgan, H. Lehermeir, M. Mang, J. Polym. Environ. 8 (2000) 1. [7] H. Xu, C.Q. Teng, M.H. Yu, Polymer 47 (2006) 3922. [8] M. Shibata, N. Teramoto, Y. Inoue, Polymer 48 (2007) 2768. [9] P. Nugroho, H. Mitomo, F. Yoshii, T. Kume, Polym. Degrad. Stab. 72 (2001) 337. [10] H. Urayama, T. Kanamori, K. Fukushima, Y. Kimura, Polymer 44 (2003) 5635. [11] H. Tsuji, Y. Ikada, Polymer 36 (1995) 2709. [12] N. Ogata, G. Jimenez, H. Kawai, T. Ogihara, J. Polym. Sci., Part B: Polym. Phys. 35 (1997) 389. [13] H. Urayama, C.H. Ma, Y. Kimura, Macromol. Mater. Eng. 288 (2003) 562. [14] T. Trimaille, C. Pichot, A. Elaissari, H. Fessi, S. Briancon, T. Delair, Colloid Polym. Sci. 281 (2003) 1184.

963

[15] X. Hu, H.S. Xu, Z.M. Li, Macromol. Mater. Eng. 292 (2007) 646. [16] Y.Z. Wan, Y.L. Wang, X.H. Xu, Q.Y. Li, J. Appl. Polym. Sci. 82 (2001) 150. [17] B.H. Li, M.C. Yang, Polym. Adv. Technol. 17 (2006) 439. [18] Y.W. Di, S. Iannace, E. Di Maio, L. Nicolais, Macromol. Mater. Eng. 290 (2005) 1083. [19] F.Z. Jin, S.H. Hyon, H. Iwata, S. Tsutsumi, Macromol. Rapid Commun 23 (2002) 909. [20] T.M. Quynh, H. Mitomo, N. Nagasawa, Y. Wada, F. Yoshii, M. Tamada, Eur. Polym. J. 43 (2007) 1779. [21] T.M. Quynh, H. Mitomo, L. Zhao, S. Asai, Carbohydr. Polym. 72 (2008) 673. [22] H. Mitomo, A. Kaneda, T.M. Quynh, N. Nagasawa, F. Yoshii, Polymer 46 (2005) 4695. [23] N. Nagasawa, A. Kaneda, S. Kanazawa, T. Yagi, H. Mitomo, F. Yoshii, M. Tamada, Nucl. Instrum. Methods Phys. Res., Sect. AdBeam Interact. Mater. Atoms 236 (2005) 611. [24] A.J. Nijienhuis, D.W. Grijpma, A.J. Pennings, Polymer 37 (1996) 2783. [25] M.G. Rimoli, L. Avallone, P. Caprariis, A. Galeone, F. Forni, M.A. Vandelli, J. Controlled Release 58 (1999) 61. [26] J.M. Zhang, H. Sato, H. Tsuji, I. Noda, Y. Ozaki, J. Mol. Struct. 735–736 (2005) 249. [27] J.Y. Liu, W. Yu, C.X. Zhao, C.X. Zhou, Polymer 48 (2007) 2882. [28] T. Semba, K. Kitagawa, U.S. Ishiaku, H. Hamada, J. Appl. Polym. Sci. 101 (2006) 1816. [29] J. Pan, Y.L. Wang, S.H. Qin, B.B. Zhang, Y.F. Luo, J. Biomed. Mater. Res. B: Appl. Biomater. 74B (2005) 476. [30] D. Carlson, N. Li, R. Narayan, P. Dubios, J. Appl. Polym. Sci. 72 (1999) 477. [31] T. Miyata, T. Masuko, Polymer 39 (1998) 5515.