International Journal of Biological Macromolecules 76 (2015) 49–57
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Effect of diameter of poly(lactic acid) fiber on the physical properties of poly(ε-caprolactone) Dandan Ju a,b , Lijing Han a,∗ , Ziqi Guo c , Junjia Bian a , Fan Li c , Shan Chen c , Lisong Dong a,∗ a b c
Key Laboratory of Polymer Ecomaterials, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China University of Chinese Academy of Sciences, Beijing 100049, China College of Life Science, Northeast Normal University, Changchun 130024, China
a r t i c l e
i n f o
Article history: Received 4 December 2014 Received in revised form 12 January 2015 Accepted 13 January 2015 Available online 20 February 2015 Keywords: Short-fiber composites Mechanical properties Enzymatic degradation
a b s t r a c t Biodegradable polymer composites based on poly(ε-caprolactone) (PCL) and poly(lactic acid) (PLA) fibers with diameters of 18, 26, 180 m were prepared by melt compounding. The PLA fiber content in the composites was constant at 20% by weight. The effects of fibers with different diameters on the physical properties and enzymatic degradation of PCL were investigated. The morphological analysis indicated good interfacial adhesion between PCL and PLA fiber, which was beneficial to improve the physical properties of PCL. With increasing PLA fiber diameter, the complex viscosity and modulus of PCL were significantly increased, especially at low frequencies, indicating that the hindered effect of the fiber on the mobility of the PCL molecular chains was more obvious when PLA fiber diameter was thicker. However, as for the mechanical properties, the reinforcement was more obvious to PCL with the smaller PLA fiber diameter. This was because increasing efficient load transfer may be appeared due to the larger surface area and better interface bonding force of the fiber with thinner diameters. The enzymatic degradation of PCL was accelerated with the addition of large PLA fiber diameter of 26 and 180 m, and hardly changed with the small PLA fiber diameter of 18 m. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The development of industrialized processes has led to increased detrimental environmental issues, in particular the generation of “white pollution”. Therefore, completely biodegradable materials are highly advantageous and are receiving increasing attention. Poly(ε-caprolactone) (PCL), a biodegradable aliphatic polyester, exhibits fair mechanical performance, good processability, brilliant biodegradability and biocompatibility, and has therefore been used in packaging and biomedical materials [1]. However, the practical application is sometimes hampered by its low stiffness, low heat resistance and slower biodegradation rate. Compounding thermoplastic polymers with natural fibers is already a well-established approach to obtain special composites with useful properties. Many studies on the PCL reinforced with plant fibers including green coconut fiber (GCF) [2,3], silk fibroin fiber (SF) [4,5], flax fiber [6], maize fiber [7], rice straw fiber (RSF) [8], chitin fiber [9–11], kenaf fiber [12], abaca [13,14], bagasse [15], alfa
∗ Corresponding authors. Tel.: +86 431 85262076; fax: +86 431 85262890. E-mail address: [email protected] (L. Han). http://dx.doi.org/10.1016/j.ijbiomac.2015.01.059 0141-8130/© 2015 Elsevier B.V. All rights reserved.
fiber mats [16], oil palm empty fruit bunch fiber (OPEFB) [17], and cellulose [18], have been investigated. The mechanical properties of these reinforced composites are reported in Table 1. It is known that the properties of the fiber-reinforced polymer composites can be influenced by many factors, and the effects of fiber types, content, length and fiber–matrix interfacial adhesion on the mechanical properties of fiber-based composites have been extensively studied in the past. Especially, for the fiber diameter, many intriguing results have been disclosed [21–25]. Obukuro et al. [21] investigated fiber-reinforced composites (FRCs) incorporating 7, 10, 13, 16, 20, 25, 30, and 45 m diameter silanized E-glass fibers, and they found an escalating flexural strength (FS) as the fibers diameter increased, except for the FRC with 45 m which had even lower FS than that of the FRC with 13 m fibers. In contrast, the elastic modulus had no significant difference with the variation of the fiber diameters. However, Rezvani et al. [22] revealed that the elastic modulus had the same trend with the FS by increasing the glass fiber diameter up to 19 m between the FRCs with two different diameters (14 and 19 m). Houshyar and Shanks [23] found that there was an optimum fiber diameter of 50 m for the stiffness and creep resistance of poly(propylene-coethylene) (PPE) reinforced by polypropylene (PP) long-fiber. The
50
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Table 1 Tensile properties of PCL/biodegradable fiber composites. Composites
Fiber content (wt%)
PCL/GCF PCL-g-AAa /GCF PCL/SF PCL/flax fiber PCL/maize fiber PCL/RSF PCL/chitin fiber PCL/kenaf fiber PCL/PLA fiber PCL/PHBVb fiber
20 20 25
a b
Fiber diameter (m)
Modulus (MPa)
20 600
17 20 15 20 20 20
60–220
25 20 134 ± 4
580 500 417 ± 10 347 ± 30
Tensile strength (MPa)
Elongation at break (%)
Reference
13.9 ± 1.6 29.1 ± 1.2 22.5 17 68.1 21 17 12.5 17.7 ± 0.4 14.6 ± 0.1
348 ± 58 610 ± 28 12.5
[2]
8.6 450 19.8 527 ± 14 498 ± 25
[4,5] [6] [7] [8] [9] [12] [19] [20]
PCL-g-AA represents acrylic acid-grafted PCL. PHBV represents poly(3-hydroxybutyrate-co-3-hydroxyvalerate).
elastic modulus increased with decreasing fiber diameter to a size of 50 m but the stiffness of the composite then decreased for fiber diameters below the 50 m size. Abraham et al. found that for the nylon fiber-reinforced polypropylene composites, the thinner fiber composites showed the higher tensile strength and modulus due to the more efficient load transfer from the matrix to the fibers ascribed to the larger surface area and the higher strength of the thinner fiber [24]. However, there has been little research completed on biodegradable thermoplastic composites reinforced with biodegradable polymer fibers. Moreover, the effect of fiber diameter on strength moduli has not been determined. In previous publications, we reported the effects of the PLA fiber [19] and PHBV fiber [20] concentration on the crystallization, mechanical properties, rheological properties, and enzymatic degradation of the PCL matrix. In this study, PLA fibers with different diameters were used as a reinforcing agent to improve the properties of PCL, and environmentally friendly PCL/PLA fiber composites were prepared by melt compounding. The influence of the PLA fibers with different diameters on the crystallization, rheological behavior, and mechanical properties and enzymatic degradation of PCL was investigated in detail.
2. Experimental 2.1. Materials A commercially available PCL (CAPA6800) with a melt flow index of 5 g/10 min and a weight-average molecular weight of 80,000 g/mol was purchased from Solvay Interox Ltd., UK. PLA fibers were kindly supplied by Nanjing GaoXin Biochemical Technology Co. Ltd. (Nanjing, China). The nominal diameters of the PLA fibers are 18, 26, and 180 m, respectively.
2.2. Preparation of the composites PCL was dried in an electric blast drying oven at 40 ◦ C for 24 h, and PLA fibers were dried in a vacuum oven at 80 ◦ C for 24 h before processing. PCL and its composites (LA18, LA26 and LA180 with PLA fiber diameter of 18, 26, and 180 m, respectively) were produced in an internal mixer (Haake Rheomix 600, Karlsruhe, Germany) with a speed of 50 rpm for a total mixing time of 5 min at 90 ◦ C. The PLA fiber content in these composites was constant at 20% by weight. After melt compounding, all the samples were cut into small pieces and then were hot-pressed at 90 ◦ C for 3 min followed by cold-press at room temperature to form the sheets with thickness of about 1 or 0.3 mm. The compression molding steps were carried out carefully in order to ensure similar processing treatment for each sample.
2.3. Characterization and measurements 2.3.1. Scanning electron microscopy (SEM) The morphology of the composites was observed using a field emission scanning electron microscopy (SEM) (XL30 ESEM FEG, FEI Co., Eindhoven, The Netherlands) at an accelerating voltage of 10 kV. The samples before and after tensile testing were immersed in liquid nitrogen for about 3 min, and then broke off. The cryofractured surfaces of the samples were coated with a thin layer of gold and then SEM micrographs recorded in order to evaluate the dispersion of the PLA fibers in the PCL matrix and the PLA fiber morphology in the PCL matrix after tensile testing. 2.3.2. Rheological measurements Rheological measurements were carried out on a rheometer (AR2000EX, TA Instruments-Waters LLC, USA) equipped with a parallel plate geometry using 25 mm diameter plates. Sheet samples of thickness of 1.0 mm were used. Frequency sweep for the samples was carried out at 80 ◦ C. To obtain reasonable signal intensities even at evaluated temperature or low frequency and to avoid nonlinear response, the small amplitude oscillatory shear (SAOS) was applied at the strain level of 1.25%. The angular frequency range used during testing was 0.01–100 rad/s. 2.3.3. Dynamic mechanical analysis (DMA) Dynamic mechanical analysis (DMA) was carried out with a dynamic mechanical analyzer SDTA861e (Mettler Toledo, Switzerland) in the tensile mode. The samples with dimensions of 9 × 4 × 1 mm3 were used. The storage modulus (E ) and the dynamic loss factor (tan ı) were determined at a frequency of 1 Hz and a heating rate of 3 ◦ C/min as a function of temperature from −80 to 40 ◦ C. 2.3.4. Tensile measurements The static mechanical properties of the samples were measured according to ISO 527-1:1993 using a tensile-testing machine (Instron-1121, USA). The samples were cut from the previously compression-molded sheets into a dumbbell shape with dimensions of 20 × 4 × 1 mm3 . The tests were conducted at a crosshead speed of 10 mm/min at room temperature (about 25 ◦ C). At least five specimens were tested for each sample to get an average value. 2.3.5. Differential scanning calorimeter (DSC) In order to investigate the degree of crystallinity of PCL in the composites before and after tensile tests, thermal analysis was performed using a TA Instruments differential scanning calorimeter DSC Q20 (USA) under N2 atmosphere. The specimens from the compression-molded sheets and the deformed specimen after tensile tests were crimp sealed in aluminum crucibles and had a
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Fig. 1. The SEM micrographs of cryo-fractured surfaces of neat PCL and PCL/PLA fiber composites with various fiber diameters: (a) PCL; (b) LA18; (c) LA26; (d) LA180.
Fig. 2. Rheological properties of PCL/PLA fiber composites: (a) storage modulus (G ); (b) loss modulus (G ); (c) complex viscosity (|*|) versus frequency at 80 ◦ C.
51
52
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
nominal weight of about 6 mg. All specimens were heated from 20 ◦ C to 75 ◦ C at a heating rate of 10 ◦ C/min, and the degree of crystallinity (c ) of PCL in the composites was calculated by the following equations [26]: c,PCL (%) = XPLA =
(Ht − Hg,PLA · XPLA ) · 100 [(1 − XPLA ) · 142]
WPLA WPCL + WPLA
(1) (2)
where WPLA and WPCL are the weights of PLA and PCL respectively, in the composites. Ht (J/g of polymer) is the enthalpy of overall transition including the glass transition enthalpy of PLA and melting enthalpy of PCL. Here we assumed the Hg,PLA to be zero as it is so small as to be undetectable, and 142 (J/g of PCL) is the melting enthalpy of PCL with 100% crystallinity reported by Crescenzi et al. [27]. 2.3.6. Enzymatic degradation The enzymatic degradation of the composites films was carried out in phosphate buffer (pH = 8) containing Pseudomonas mendocina at 30 ◦ C with shaking at 140 rpm. Sample films (approximately 10 × 10 × 0.3 mm3 ) were placed in small glass bottles containing the buffer and P. mendocina [28]. The films were periodically removed, washed with distilled water, and dried to constant weight in a vacuum, and then the weights of the films were measured. For comparison, neat PCL was treated with the same procedure. The degraded surfaces and cross-section of the samples were coated with a thin layer of gold and then they were observed with SEM at an accelerating voltage of 10 kV to obtain the morphology after enzymatic degradation. 3. Result and discussion 3.1. The morphology of PCL/PLA fiber composites The fiber–matrix interface is a critical factor that determines the properties of the composite. Fig. 1 shows the SEM micrographs of cryo-fractured surfaces of PCL/PLA fiber composites with various fiber diameters. The surface of neat PCL was smooth and featureless. With the addition of PLA fibers, the two-phase structure was visible. For all the composites, the PLA fibers were dispersed evenly in the PCL matrix, and with increasing the fiber diameter, the number of fibers visible in cross-section was reduced. Little fiber pullout was observed, with the fracture surfaces exhibiting a distinct “blocky” appearance. This was most probably caused by regions of the fiber, well bonded to the matrix, being ‘torn’ from underlying layers of the fiber [29]. The fractured surface showing evidence for fiber breakage rather than pullout indicated better interfacial adhesion between the PLA fiber and the PCL matrix which contributed to an efficient stress transfer from the matrix to the fiber causing the improvement of the storage and tensile modulus as follows reported. 3.2. Rheological behavior Rheological analysis is considered an effective method for evaluation of melt processing parameters, dispersion of fillers, filler–matrix interaction, and morphology in composites. Fig. 2 shows the complex viscosity |*|, storage modulus G , and loss modulus G versus frequency for the neat PCL and various composites at 80 ◦ C. Neat PCL showed a typical Newtonian fluid in a wide low-frequency range and pseudoplastic shear thinning behavior when the frequency increased. When the PLA fibers with different diameters were added, the |*| increased with increasing fiber diameter, and the low-frequency viscosity plateau became
Fig. 3. Han plots of G versus G (a), and Cole–Cole plots of imaginary viscosity ( ) versus real viscosity ( ) (b) for neat PCL and various composites at 80 ◦ C.
unclear and finally disappeared when the fiber diameter increased to 180 m as shown in Fig. 2a, which implied that the composite exhibited non-Newtonian fluid behavior. This could be explained by the presence of the PLA fibers acting like obstacles to resist the movement of the PCL molecular chains, resulting in an enhancement of the system |*| [30]. For the composites with larger fiber diameters, extra stress was required by the system to overcome the resistance of the PLA fiber which resulted in a higher |*| of the polymer matrix, especially at low frequencies. This also can be confirmed by the results of the tensile test (shown as follows). Like what has been observed in the case of the |*|, the values of G and G also increased with increasing the fiber diameter at all frequencies, especially at low frequencies (Fig. 2b and c). In the viscoelastic theory, G represents the elastic aspect of the material and G is the viscous aspect. A plot of G versus G (Han plots) of melts in an oscillatory flow is an empirical tool for the evaluation of the miscibility and interfacial interaction between the reinforced fiber and polymer matrix in a composite. Fig. 3a shows G versus G for the PCL/PLA fiber composites. The intersection (G = G ) determines the transition from more viscous behavior (G < G ) to more elastic behavior (G > G ). From Fig. 3a, the PCL and all composites exhibited a predominantly viscous behavior rather than an elastic behavior, especially at low frequencies. However, the elastic part was increased with an increase of the PLA fiber diameters. The slope of the curve was reduced with increasing fiber diameter, indicative of a more heterogeneous composite. It is notable that the position of the inflection point shifted to higher
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
53
increase in the modulus of virgin matrix with the incorporation of PLA fiber especially in the high temperature. The reinforcing effect was more significant for the LA18 sample. Table 2 shows E values in two different temperatures (−20 ◦ C and 20 ◦ C). The E increased about 50% at −20 ◦ C and 54% at 20 ◦ C for the LA18, indicating that the PLA fiber was an effective reinforcing agent for PCL. The reinforcing effect was probably due to higher stiffness of the PLA fiber which introduced rigid interface into the composites, which allowed a greater degree of stress transfer at the interface. As for the LA18, the larger surface area at equal fiber loading for the 18 m PLA fibers resulted more stress transfer and thus higher modulus. The increased E also indicated that the heat resistance of PCL was improved, which is useful for the choice of the proper materials for specific applications. For all the samples, with increasing temperature, the E curve exhibited a sharp decrease in the range of −60 to −20 ◦ C due to the glass transition of PCL. Fig. 4b shows the variations of tan ı as a function of temperature. Neat PCL exhibited a relaxation peak at about −46.6 ◦ C, corresponding to the glass transition temperature (Tg ) of PCL. For the composites, there was a very slight increase in the Tg compared to the neat PCL. For the LA18, the Tg value of PCL component was −44 ◦ C, about a 2 ◦ C increase. This slight increase may be indicative of an interfacial interaction between matrix and fiber, hindering the chain mobility and rearrangement [31]. The interfacial interaction was essential to improve the mechanical strength of the composites. What’s more, the height of the tan ı peak decreased with the presence of PLA fibers. It was due to a better fiber/matrix bonding that resulted not only the immobilization of polymer matrix in the presence of the PLA fiber but also an improvement in the hysteresis of the composites thus a reduction in the internal friction [32]. This phenomenon was more obvious for the LA18 due to the larger surface area at equal fiber loading for the PLA fibers of 18 m. 3.4. Static mechanical properties
Fig. 4. The dynamic mechanical traces of neat PCL and various PCL/PLA fiber composites: (a) storage modulus (E ); (b) tan ı versus temperature curves.
frequency with increasing fiber diameter. This suggested that a greater amount of energy was required for the more heterogeneous composites produced with larger fibers. The Cole–Cole plot is usually used to study viscoelastic properties of the materials with a relaxation time distribution such as heterogeneous polymeric systems. For many polymer blends, Cole–Cole plots yield two arcs, attributed to two processes of significantly different relaxation times. Fig. 3b shows Cole–Cole plots for neat PCL and PCL/PLA fiber composites at 80 ◦ C. Neat PCL showed only one circular arc. For the composites, a tail or a second circular arc appeared on the right-hand side of the arc, which indicated a second relaxation mechanism. One arc in the low-viscosity region was corresponding to the local dynamic relaxation of PCL, and the tail in the high-viscosity region was indicative of the long-term relaxation of the PCL chains restrained by the PLA fibers. An increasing effect to the high-viscosity region resulted from the composite containing increasing fiber diameters. Moreover, the long-term relaxation of those restrained PCL chains in the composites became more obvious when the fiber diameter increased to 180 m. This also indicated that the hindered effect of the PLA fiber on the movement of the PCL molecular chains was more obvious with the bigger PLA fiber diameter. 3.3. Dynamical mechanical properties Fig. 4 presents the dynamic mechanical spectra of neat PCL and its composites. It is evident from Fig. 4a that there was a notable
The effects of blending with different diameters PLA fibers on the static mechanical properties of PCL were investigated by tensile tests. The stress–strain curves of neat PCL and PCL/PLA fiber composites are shown in Fig. 5. The corresponding data of mechanical properties are summarized in Table 3. Neat PCL showed higher elongation at break and strength at break, but lower strength at yielding and tensile modulus. With the addition of PLA fibers with different diameters, the tensile yielding strength and modulus were improved at different levels. The strength at yielding and tensile modulus increased from 13.7 MPa and 270 MPa for neat PCL to 15.5 MPa and 474 MPa for the composite LA18, respectively, increases of 13% and 76%, respectively. However, with increasing the fiber diameter, the strength at yielding and tensile modulus gradually decreased, which suggested that the smaller the PLA fiber diameter was, the better the reinforcement to PCL. The reinforcement effect may be due to that the high modulus PLA fibers act as carriers of the load and stress which are transferred from the matrix to the fibers giving rise to effective and uniform stress distribution, resulting in an increase in the strength at yielding and tensile modulus of the PCL/PLA fiber composites [20]. For the composites with uniform fiber composition, the more efficient load transfer may be due to the larger surface area of the thinner fiber diameters [24], which lead to the more obvious reinforcement effect for the thin fiber compared to thick fiber. However, with increasing the fiber diameter, the elongation at break and strength at break gradually decreased. As was shown in Fig. 5a, there was a strain-hardening phenomenon after the yielding of neat PCL due to the orientation of PCL molecules during tensile tests, and the strain-hardening ability was very strong, which lead to a high strength at break. As for the PCL/PLA fiber composites, the strain-hardening capacity
54
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Table 2 Storage modulus values of the PCL/PLA fiber composites at different temperatures. Samples
Storage modulus at −20 ◦ C (MPa)
Storage modulus at 20 ◦ C (MPa)
Reinforcement imparted by the PLA fibers at 20 ◦ C (%)
PCL LA18 LA26 LA180
680 1024 745 875
424 653 477 560
− 54 12.5 32
Table 3 Tensile properties of neat PCL and PCL/PLA fiber composites with various fiber diameters. Sample
Strength at yielding (MPa)
PCL LA18 LA26 LA180
13.7 15.5 15.2 12.3
± ± ± ±
0.8 0.1 0.3 0.4
Tensile modulus (MPa) 270 474 390 319
± ± ± ±
Elongation at break (%)
7 21 8 21
of the composites decreased gradually with the increasing fiber diameters. When the diameter of the PLA fibers increased to 180 m, the strain-hardening phenomenon disappeared. It was because that the PLA fibers hindered the molecular reordering of the PCL polymer chains which was more pronounced in larger fibers than small, thereby contributing to the overall reduction in the elongation at break and the strain-hardening ability [33]. In addition, because of the discontinuous reinforcement of the short fiber and the immiscibility of the PCL and PLA fiber, the stress and strain at break dropped rapidly with increasing the fiber diameter.
1544 742 441 40
± ± ± ±
Strength at break (MPa) 49.4 ± 1.4 21.6 ± 0.3 12.7 ± 1.0 –
52 10 41 5
However, for the composites with PLA fiber diameter of 18 and 26 m, it still exhibited high ductility. Fig. 6 shows the longitudinal fracture surface morphology of PCL/PLA fiber composites after tensile tests. The fracture surfaces for neat PCL, LA18, and LA26 exhibited orderly arranged fibrillar bundle structures parallel to the stretching direction, which contributed to the significantly higher strength and toughness of the composites [29,34]. With increasing the fiber diameters, the fibrous structure became unobvious. When the PLA fiber diameter increased to 180 m, the fibrillar bundle structures disappeared. The evidence provided previously was consistent with that the mobility of the macromolecular chains and molecular orientation of the PCL during tension were impeded due to the presence of PLA fibers in the composites, thereby contributing to the overall reduction in the elongation at break and the strain-hardening ability. And the impeding effect was increased with increasing the PLA fiber diameters. That also confirmed that the |*| of the composites discussed above was increased with the increase of the PLA fiber diameters due to its hindered effect on the mobility of the PCL molecules. As is known to all, the molecular arrangement of polymer chains can be modified through stretching leading to higher crystallinity, the so-called stress-induced crystallization. Fig. 7 shows the DSC thermograms of the first heating traces before and after tensile tests for PCL/PLA fiber composites with different PLA fiber diameters, and the c of PCL was calculated as listed in Table 4. From Table 4, for all the samples, the melting temperature (Tm ) of the PCL component after tensile tests moved to a higher temperature, and the melting enthalpy (Hm ) and c increased, indicating the stress induced the orientation of the molecular chains and hence promoted the crystallization of the PCL component during drawing [34]. The c of neat PCL was increased from 40.9% before tensile tests to 58.2% after tensile tests, which increased significantly by 42.3%. While for the composites LA18, LA26 and LA180, the c of the PCL in composites was increased from 40.2%, 42.7% and 44.0% before tensile tests to 56.1%, 49.7% and 49.6% after tensile tests, respectively. The increase of 39.6%, 16.4% and 12.7% was obtained for the composites LA18,
Table 4 Thermal and crystalline properties of PCL in the composites obtained from the first heating traces at a rate of 10 ◦ C/min before and after tensile tests. Sample
Fig. 5. Tensile stress–strain curves of the neat PCL and PCL/PLA fiber composites with various fiber diameters: (b) gives details of tensile stress at low strain of (a).
PCL LA18 LA26 LA180 a
Tm (◦ C)
Hm a (J/g)
c,PCL (%)
Before
After
Before
After
Before
After
58.1 57.8 57.3 57.7
60.0 60.0 59.5 58.3
58.1 57.1 60.6 62.5
82.6 79.6 70.6 70.5
40.9 40.2 42.7 44.0
58.2 56.1 49.7 49.6
Hm are corrected for the content of PCL in the composites.
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
55
Fig. 6. The SEM micrographs of the longitudinal fracture surface of various PCL/PLA fiber composites after tensile tests at a tensile rate of 10 mm/min.
LA26 and LA180 respectively, which was smaller than that of neat PCL. It further confirmed that the PLA fibers hindered the molecular reordering of the PCL polymer chains which was more pronounced in larger fibers than small during tension therefore depressed the process of the crystallization during tensile tests. In addition, the level of the depression effect was more evident with increasing the PLA fiber diameters, which was in accordance with the result of the longitudinal fracture surface morphology of the composites after tensile tests.
Fig. 8 shows the normalized weight loss profiles of the PCL as a result of P. mendocina lipase-catalyzed degradation. As was
shown in Fig. 8, the weight loss of the films increased with time for all samples. There was little difference among the four samples for the enzymatic degradation before three days due to the incubation period for P. mendocina. However, the weight loss for the LA26 and LA180 was significantly increased after 3 days compared to the two other composites. Overall, this indicated that the degree of degradation for the polymer or composites was not differentiated. However, the rate of enzymatic degradation of PCL or PCL composite was affected by the size of composite fibers of PLA present: PCL and small-fiber LA18 were much slower to degrade than the two composites containing larger fibers. It can be further confirmed by the surface and cross-section morphology of neat PCL and the composites as shown in Fig. 9. Before enzymatic degradation, all the films presented an unchanged smooth
Fig. 7. The first heating traces of DSC thermograms for neat PCL and PCL/PLA fiber composites before and after tensile tests at a tensile rate of 10 mm/min.
Fig. 8. The normalized PCL weight loss profiles of PCL/PLA fiber composites as a function of time during the enzymatic degradation.
3.5. Enzymatic degradation
56
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Fig. 9. The surface and cross-section morphology of neat PCL and the composites before and after enzymatic degradation for different times: surface morphology of (a1 –a3 ) neat PCL; (b1 –b3 ) LA18; (c1 –c3 ) LA26; (d1 –d3 ) LA180 after degradation for 0, 3, and 6 days, respectively; (a4 –d4 ) cross-section morphology of neat PCL and the composites after enzymatic degradation for 6 days, respectively.
surface. While the PLA fibers in all the samples were exposed with time due to the enzymatic degradation of the PCL. When the composites were degraded for 6 days, the matrix around the fiber already disappeared due to the serious surface erosion, which led to the large crack near the fibers especially for the LA26 and LA180 as shown in Fig. 9c3 and d3 . This indicated that PLA fibers and their interface could act as channels to promote the water entrance and diffusion from the surface to the inside of the composites and serve as a support for the attack of the enzyme molecules, which accelerated the degradation of the PCL matrix around the fiber [20]. As for the LA18, the better interface bonding force between the PCL and the PLA fiber hindered the entrance and the diffusion of enzyme molecules, therefore having little influence on the enzymatic degradation of the PCL. This was also consistent with the cross-section morphology of the composites as shown in Fig. 9a4 –d4 . The film thickness of neat PCL decreased from 300 m (the original film thickness) to about 203 m after degraded for 6 days, and the film thickness of LA18 was similar with neat PCL under the same conditions. As for the other composites, the film thickness decreased to about 83 m for the LA26 and 58 m for the LA180, respectively. Consequently, the variation of weight loss with enzymatic degradation time, the changes of surface morphology, and the reduced film thickness, suggested that the enzymatic degradation of neat PCL and the PCL/PLA fiber composites was likely to proceed via surface erosion mechanism, and the presence of PLA fibers with bigger fiber diameters promoted the enzymatic degradation of PCL.
4. Conclusions Biodegradable polymer composites based on PCL and PLA fibers with various fiber diameters were prepared by melt compounding. The effects of PLA fiber diameter on the morphology, rheological properties, mechanical properties and enzymatic degradation of PCL composites were investigated in detail. The SEM micrographs of the cryo-fractured surfaces of PCL/PLA fiber composites showed the fiber breakage rather than pullout indicating good interfacial adhesion between the PLA fiber and the PCL matrix. With increasing PLA fiber diameters, the |*|, G and G increased at the measured temperatures, especially at low frequencies. This could be explained that the presence of the PLA fibers hindered the molecular mobility of the PCL polymer chains, and for the composites with larger diameter fiber, more stress was required for the system to overcome the resistance of the PLA fiber. This effect was rationalized by the presence of PLA fibers in the composites hindering the mobility of the macromolecular chains and molecular orientations during tension, thereby contributing to the overall reduction in the elongation at break, strain-hardening ability, and the stress-induced crystallization and hence strength at break. However, the dynamic storage modulus, tensile yielding strength and modulus of the PCL were improved with the addition of the PLA fibers, especially for the 18 m-diameter fiber. This indicated the smaller the reinforcing PLA fiber diameter was, the better the reinforcement of PCL. The enzymatic degradation of the PCL was accelerated when the PLA fiber diameters were 26 and 180 m. While for the fiber of 18 m,
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
there was little influence on the enzymatic degradation due to the better interface bonding force between the PCL and the PLA fiber. Such a PLA fiber-reinforced PCL composite could be used in the packaging field to alleviate white pollution. Additionally, the reinforced PCL by PLA fiber still can be used in a biomedical field due to the biodegradability and biocompatibility of both the PCL and PLA. Acknowledgement The authors are grateful for the support from the National Natural Science Foundation of China (51021003). References [1] Y. Ikada, H. Tsuji, Macromol. Rapid Commun. 21 (2000) 117–132. [2] C.S. Wu, J. Appl. Polym. Sci. 115 (2010) 948–956. [3] M.C.A.M. Leite, C.R.G. Furtado, L.O. Couto, F.L.B.O. Oliveira, T.R. Correia, Polimeros 20 (2010) 339–344. [4] X.Y. Qiao, W. Li, H. Watanabe, K. Sun, X.D. Chen, J. Polym. Sci. B: Polym. Phys. 47 (2009) 1957–1970. [5] W. Li, X.Y. Qiao, K. Sun, X.D. Chen, J. Appl. Polym. Sci. 110 (2008) 134–139. [6] A. Arbelaiz, B. Fernandez, A. Valea, I. Mondragon, Carbohydr. Polym. 64 (2006) 224–232. [7] M. Dauda, M. Yoshiba, K. Miura, S. Takahashi, Adv. Compos. Mater. 16 (2007) 335–347. [8] C.S. Wu, H.T. Liao, Ind. Eng. Chem. Res. 51 (2012) 3329–3337. [9] B.Q. Chen, K. Sun, T. Ren, Eur. Polym. J. 41 (2005) 453–457. [10] A.L. Yang, R.J. Wu, J. Appl. Polym. Sci. 84 (2002) 486–492.
57
[11] A. Yang, K. Sun, R. Wu, 45th International SAMPE Symposium and Exhibition on Bridging the Centuries with SAMPEs Materials and Processes Technology, vol. 45, 2000, pp. 427–432. [12] P. Pan, B. Zhu, T. Dong, S. Serizawa, M. Iji, Y. Inoue, J. Appl. Polym. Sci. 107 (2008) 3512–3519. [13] N. Teramoto, K. Urata, K. Ozawa, M. Shibata, Polym. Degrad. Stab. 86 (2004) 401–409. [14] M. Shibata, R. Yosomiya, N. Ohta, A. Sakamoto, H. Takeishi, Polym. Polym. Compos. 11 (2003) 359–367. [15] Y. Cao, K. Goda, S. Shibata, Adv. Compos. Mater. 16 (2007) 283–298. [16] Z. Marrakchi, H. Oueslati, M.N. Belgacem, F. Mhenni, E. Mauret, Compos. A 43 (2012) 742–747. [17] M.Z.A. Hamid, N.A. Ibrahim, W.M.Z.W. Yunus, K. Zaman, M. Dahlan, J. Reinf. Plast. Compos. 29 (2010) 2723–2731. [18] K.J. Kim, J.L. White, Compos. Interfaces 16 (2009) 619–637. [19] D.D. Ju, L.J. Han, F. Li, S. Chen, L.S. Dong, Polym. Compos. 34 (2013) 1745–1752. [20] D.D. Ju, L.J. Han, F. Li, S. Chen, L.S. Dong, Int. J. Biol. Macromol. 67 (2014) 343–350. [21] M. Obukuro, Y. Takahashi, H. Shimizu, Dent. Mater. J. 27 (2008) 541–548. [22] M.B. Rezvani, M. Atai, F. Hamze, Indian J. Dent. Res. 24 (2013) 237–241. [23] S. Houshyar, R.A. Shanks, Polym. Int. 53 (2004) 1752–1759. [24] T.N. Abraham, K.E. George, J. Thermoplast. Compos. Mater. 22 (2009) 5–20. [25] B. Mu, Z.J. Li, Key Eng. Mater. 145–149 (1998) 607–612. [26] H. Tsuji, Y. Ikada, J. Appl. Polym. Sci. 60 (1996) 2367–2375. [27] V. Crescenzi, G. Manzini, G. Calzolari, C. Borri, Eur. Polym. J. 8 (1972) 449–463. [28] Y. Wang, F. Li, Z.Y. Wang, D.B. Liu, H.M. Xia, L.F. Liu, S. Chen, Chem. Res. Chin. Univ. 28 (2012) 459–464. [29] M. Hughes, J. Carpenter, C. Hill, J. Mater. Sci. 42 (2007) 2499–2511. [30] R. Guo, J. Azaiez, C. Bellehumeur, Polym. Eng. Sci. 45 (2005) 385–399. [31] H. Anuar, S.H. Ahmad, R. Rasid, A. Ahmad, W.N. Wan Busu, J. Appl. Polym. Sci. 107 (2008) 4043–4052. [32] M.S. Huda, A.K. Mohanty, L.T. Drzal, E. Schut, M. Misra, J. Mater. Sci. 40 (2005) 4221–4229. [33] S. Jain, M.M. Reddy, A.K. Mohanty, M. Misra, A.K. Ghosh, Macromol. Mater. Eng. 295 (2010) 750–762. [34] F. Feng, X.W. Zhao, L. Ye, J. Macromol. Sci. Phys. 50 (2011) 1500–1507.
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Effect of diameter of poly(lactic acid) fiber on the physical properties of poly(ε-caprolactone) Dandan Ju a,b , Lijing Han a,∗ , Ziqi Guo c , Junjia Bian a , Fan Li c , Shan Chen c , Lisong Dong a,∗ a b c
Key Laboratory of Polymer Ecomaterials, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China University of Chinese Academy of Sciences, Beijing 100049, China College of Life Science, Northeast Normal University, Changchun 130024, China
a r t i c l e
i n f o
Article history: Received 4 December 2014 Received in revised form 12 January 2015 Accepted 13 January 2015 Available online 20 February 2015 Keywords: Short-fiber composites Mechanical properties Enzymatic degradation
a b s t r a c t Biodegradable polymer composites based on poly(ε-caprolactone) (PCL) and poly(lactic acid) (PLA) fibers with diameters of 18, 26, 180 m were prepared by melt compounding. The PLA fiber content in the composites was constant at 20% by weight. The effects of fibers with different diameters on the physical properties and enzymatic degradation of PCL were investigated. The morphological analysis indicated good interfacial adhesion between PCL and PLA fiber, which was beneficial to improve the physical properties of PCL. With increasing PLA fiber diameter, the complex viscosity and modulus of PCL were significantly increased, especially at low frequencies, indicating that the hindered effect of the fiber on the mobility of the PCL molecular chains was more obvious when PLA fiber diameter was thicker. However, as for the mechanical properties, the reinforcement was more obvious to PCL with the smaller PLA fiber diameter. This was because increasing efficient load transfer may be appeared due to the larger surface area and better interface bonding force of the fiber with thinner diameters. The enzymatic degradation of PCL was accelerated with the addition of large PLA fiber diameter of 26 and 180 m, and hardly changed with the small PLA fiber diameter of 18 m. © 2015 Elsevier B.V. All rights reserved.
1. Introduction The development of industrialized processes has led to increased detrimental environmental issues, in particular the generation of “white pollution”. Therefore, completely biodegradable materials are highly advantageous and are receiving increasing attention. Poly(ε-caprolactone) (PCL), a biodegradable aliphatic polyester, exhibits fair mechanical performance, good processability, brilliant biodegradability and biocompatibility, and has therefore been used in packaging and biomedical materials [1]. However, the practical application is sometimes hampered by its low stiffness, low heat resistance and slower biodegradation rate. Compounding thermoplastic polymers with natural fibers is already a well-established approach to obtain special composites with useful properties. Many studies on the PCL reinforced with plant fibers including green coconut fiber (GCF) [2,3], silk fibroin fiber (SF) [4,5], flax fiber [6], maize fiber [7], rice straw fiber (RSF) [8], chitin fiber [9–11], kenaf fiber [12], abaca [13,14], bagasse [15], alfa
∗ Corresponding authors. Tel.: +86 431 85262076; fax: +86 431 85262890. E-mail address: [email protected] (L. Han). http://dx.doi.org/10.1016/j.ijbiomac.2015.01.059 0141-8130/© 2015 Elsevier B.V. All rights reserved.
fiber mats [16], oil palm empty fruit bunch fiber (OPEFB) [17], and cellulose [18], have been investigated. The mechanical properties of these reinforced composites are reported in Table 1. It is known that the properties of the fiber-reinforced polymer composites can be influenced by many factors, and the effects of fiber types, content, length and fiber–matrix interfacial adhesion on the mechanical properties of fiber-based composites have been extensively studied in the past. Especially, for the fiber diameter, many intriguing results have been disclosed [21–25]. Obukuro et al. [21] investigated fiber-reinforced composites (FRCs) incorporating 7, 10, 13, 16, 20, 25, 30, and 45 m diameter silanized E-glass fibers, and they found an escalating flexural strength (FS) as the fibers diameter increased, except for the FRC with 45 m which had even lower FS than that of the FRC with 13 m fibers. In contrast, the elastic modulus had no significant difference with the variation of the fiber diameters. However, Rezvani et al. [22] revealed that the elastic modulus had the same trend with the FS by increasing the glass fiber diameter up to 19 m between the FRCs with two different diameters (14 and 19 m). Houshyar and Shanks [23] found that there was an optimum fiber diameter of 50 m for the stiffness and creep resistance of poly(propylene-coethylene) (PPE) reinforced by polypropylene (PP) long-fiber. The
50
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Table 1 Tensile properties of PCL/biodegradable fiber composites. Composites
Fiber content (wt%)
PCL/GCF PCL-g-AAa /GCF PCL/SF PCL/flax fiber PCL/maize fiber PCL/RSF PCL/chitin fiber PCL/kenaf fiber PCL/PLA fiber PCL/PHBVb fiber
20 20 25
a b
Fiber diameter (m)
Modulus (MPa)
20 600
17 20 15 20 20 20
60–220
25 20 134 ± 4
580 500 417 ± 10 347 ± 30
Tensile strength (MPa)
Elongation at break (%)
Reference
13.9 ± 1.6 29.1 ± 1.2 22.5 17 68.1 21 17 12.5 17.7 ± 0.4 14.6 ± 0.1
348 ± 58 610 ± 28 12.5
[2]
8.6 450 19.8 527 ± 14 498 ± 25
[4,5] [6] [7] [8] [9] [12] [19] [20]
PCL-g-AA represents acrylic acid-grafted PCL. PHBV represents poly(3-hydroxybutyrate-co-3-hydroxyvalerate).
elastic modulus increased with decreasing fiber diameter to a size of 50 m but the stiffness of the composite then decreased for fiber diameters below the 50 m size. Abraham et al. found that for the nylon fiber-reinforced polypropylene composites, the thinner fiber composites showed the higher tensile strength and modulus due to the more efficient load transfer from the matrix to the fibers ascribed to the larger surface area and the higher strength of the thinner fiber [24]. However, there has been little research completed on biodegradable thermoplastic composites reinforced with biodegradable polymer fibers. Moreover, the effect of fiber diameter on strength moduli has not been determined. In previous publications, we reported the effects of the PLA fiber [19] and PHBV fiber [20] concentration on the crystallization, mechanical properties, rheological properties, and enzymatic degradation of the PCL matrix. In this study, PLA fibers with different diameters were used as a reinforcing agent to improve the properties of PCL, and environmentally friendly PCL/PLA fiber composites were prepared by melt compounding. The influence of the PLA fibers with different diameters on the crystallization, rheological behavior, and mechanical properties and enzymatic degradation of PCL was investigated in detail.
2. Experimental 2.1. Materials A commercially available PCL (CAPA6800) with a melt flow index of 5 g/10 min and a weight-average molecular weight of 80,000 g/mol was purchased from Solvay Interox Ltd., UK. PLA fibers were kindly supplied by Nanjing GaoXin Biochemical Technology Co. Ltd. (Nanjing, China). The nominal diameters of the PLA fibers are 18, 26, and 180 m, respectively.
2.2. Preparation of the composites PCL was dried in an electric blast drying oven at 40 ◦ C for 24 h, and PLA fibers were dried in a vacuum oven at 80 ◦ C for 24 h before processing. PCL and its composites (LA18, LA26 and LA180 with PLA fiber diameter of 18, 26, and 180 m, respectively) were produced in an internal mixer (Haake Rheomix 600, Karlsruhe, Germany) with a speed of 50 rpm for a total mixing time of 5 min at 90 ◦ C. The PLA fiber content in these composites was constant at 20% by weight. After melt compounding, all the samples were cut into small pieces and then were hot-pressed at 90 ◦ C for 3 min followed by cold-press at room temperature to form the sheets with thickness of about 1 or 0.3 mm. The compression molding steps were carried out carefully in order to ensure similar processing treatment for each sample.
2.3. Characterization and measurements 2.3.1. Scanning electron microscopy (SEM) The morphology of the composites was observed using a field emission scanning electron microscopy (SEM) (XL30 ESEM FEG, FEI Co., Eindhoven, The Netherlands) at an accelerating voltage of 10 kV. The samples before and after tensile testing were immersed in liquid nitrogen for about 3 min, and then broke off. The cryofractured surfaces of the samples were coated with a thin layer of gold and then SEM micrographs recorded in order to evaluate the dispersion of the PLA fibers in the PCL matrix and the PLA fiber morphology in the PCL matrix after tensile testing. 2.3.2. Rheological measurements Rheological measurements were carried out on a rheometer (AR2000EX, TA Instruments-Waters LLC, USA) equipped with a parallel plate geometry using 25 mm diameter plates. Sheet samples of thickness of 1.0 mm were used. Frequency sweep for the samples was carried out at 80 ◦ C. To obtain reasonable signal intensities even at evaluated temperature or low frequency and to avoid nonlinear response, the small amplitude oscillatory shear (SAOS) was applied at the strain level of 1.25%. The angular frequency range used during testing was 0.01–100 rad/s. 2.3.3. Dynamic mechanical analysis (DMA) Dynamic mechanical analysis (DMA) was carried out with a dynamic mechanical analyzer SDTA861e (Mettler Toledo, Switzerland) in the tensile mode. The samples with dimensions of 9 × 4 × 1 mm3 were used. The storage modulus (E ) and the dynamic loss factor (tan ı) were determined at a frequency of 1 Hz and a heating rate of 3 ◦ C/min as a function of temperature from −80 to 40 ◦ C. 2.3.4. Tensile measurements The static mechanical properties of the samples were measured according to ISO 527-1:1993 using a tensile-testing machine (Instron-1121, USA). The samples were cut from the previously compression-molded sheets into a dumbbell shape with dimensions of 20 × 4 × 1 mm3 . The tests were conducted at a crosshead speed of 10 mm/min at room temperature (about 25 ◦ C). At least five specimens were tested for each sample to get an average value. 2.3.5. Differential scanning calorimeter (DSC) In order to investigate the degree of crystallinity of PCL in the composites before and after tensile tests, thermal analysis was performed using a TA Instruments differential scanning calorimeter DSC Q20 (USA) under N2 atmosphere. The specimens from the compression-molded sheets and the deformed specimen after tensile tests were crimp sealed in aluminum crucibles and had a
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Fig. 1. The SEM micrographs of cryo-fractured surfaces of neat PCL and PCL/PLA fiber composites with various fiber diameters: (a) PCL; (b) LA18; (c) LA26; (d) LA180.
Fig. 2. Rheological properties of PCL/PLA fiber composites: (a) storage modulus (G ); (b) loss modulus (G ); (c) complex viscosity (|*|) versus frequency at 80 ◦ C.
51
52
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
nominal weight of about 6 mg. All specimens were heated from 20 ◦ C to 75 ◦ C at a heating rate of 10 ◦ C/min, and the degree of crystallinity (c ) of PCL in the composites was calculated by the following equations [26]: c,PCL (%) = XPLA =
(Ht − Hg,PLA · XPLA ) · 100 [(1 − XPLA ) · 142]
WPLA WPCL + WPLA
(1) (2)
where WPLA and WPCL are the weights of PLA and PCL respectively, in the composites. Ht (J/g of polymer) is the enthalpy of overall transition including the glass transition enthalpy of PLA and melting enthalpy of PCL. Here we assumed the Hg,PLA to be zero as it is so small as to be undetectable, and 142 (J/g of PCL) is the melting enthalpy of PCL with 100% crystallinity reported by Crescenzi et al. [27]. 2.3.6. Enzymatic degradation The enzymatic degradation of the composites films was carried out in phosphate buffer (pH = 8) containing Pseudomonas mendocina at 30 ◦ C with shaking at 140 rpm. Sample films (approximately 10 × 10 × 0.3 mm3 ) were placed in small glass bottles containing the buffer and P. mendocina [28]. The films were periodically removed, washed with distilled water, and dried to constant weight in a vacuum, and then the weights of the films were measured. For comparison, neat PCL was treated with the same procedure. The degraded surfaces and cross-section of the samples were coated with a thin layer of gold and then they were observed with SEM at an accelerating voltage of 10 kV to obtain the morphology after enzymatic degradation. 3. Result and discussion 3.1. The morphology of PCL/PLA fiber composites The fiber–matrix interface is a critical factor that determines the properties of the composite. Fig. 1 shows the SEM micrographs of cryo-fractured surfaces of PCL/PLA fiber composites with various fiber diameters. The surface of neat PCL was smooth and featureless. With the addition of PLA fibers, the two-phase structure was visible. For all the composites, the PLA fibers were dispersed evenly in the PCL matrix, and with increasing the fiber diameter, the number of fibers visible in cross-section was reduced. Little fiber pullout was observed, with the fracture surfaces exhibiting a distinct “blocky” appearance. This was most probably caused by regions of the fiber, well bonded to the matrix, being ‘torn’ from underlying layers of the fiber [29]. The fractured surface showing evidence for fiber breakage rather than pullout indicated better interfacial adhesion between the PLA fiber and the PCL matrix which contributed to an efficient stress transfer from the matrix to the fiber causing the improvement of the storage and tensile modulus as follows reported. 3.2. Rheological behavior Rheological analysis is considered an effective method for evaluation of melt processing parameters, dispersion of fillers, filler–matrix interaction, and morphology in composites. Fig. 2 shows the complex viscosity |*|, storage modulus G , and loss modulus G versus frequency for the neat PCL and various composites at 80 ◦ C. Neat PCL showed a typical Newtonian fluid in a wide low-frequency range and pseudoplastic shear thinning behavior when the frequency increased. When the PLA fibers with different diameters were added, the |*| increased with increasing fiber diameter, and the low-frequency viscosity plateau became
Fig. 3. Han plots of G versus G (a), and Cole–Cole plots of imaginary viscosity ( ) versus real viscosity ( ) (b) for neat PCL and various composites at 80 ◦ C.
unclear and finally disappeared when the fiber diameter increased to 180 m as shown in Fig. 2a, which implied that the composite exhibited non-Newtonian fluid behavior. This could be explained by the presence of the PLA fibers acting like obstacles to resist the movement of the PCL molecular chains, resulting in an enhancement of the system |*| [30]. For the composites with larger fiber diameters, extra stress was required by the system to overcome the resistance of the PLA fiber which resulted in a higher |*| of the polymer matrix, especially at low frequencies. This also can be confirmed by the results of the tensile test (shown as follows). Like what has been observed in the case of the |*|, the values of G and G also increased with increasing the fiber diameter at all frequencies, especially at low frequencies (Fig. 2b and c). In the viscoelastic theory, G represents the elastic aspect of the material and G is the viscous aspect. A plot of G versus G (Han plots) of melts in an oscillatory flow is an empirical tool for the evaluation of the miscibility and interfacial interaction between the reinforced fiber and polymer matrix in a composite. Fig. 3a shows G versus G for the PCL/PLA fiber composites. The intersection (G = G ) determines the transition from more viscous behavior (G < G ) to more elastic behavior (G > G ). From Fig. 3a, the PCL and all composites exhibited a predominantly viscous behavior rather than an elastic behavior, especially at low frequencies. However, the elastic part was increased with an increase of the PLA fiber diameters. The slope of the curve was reduced with increasing fiber diameter, indicative of a more heterogeneous composite. It is notable that the position of the inflection point shifted to higher
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
53
increase in the modulus of virgin matrix with the incorporation of PLA fiber especially in the high temperature. The reinforcing effect was more significant for the LA18 sample. Table 2 shows E values in two different temperatures (−20 ◦ C and 20 ◦ C). The E increased about 50% at −20 ◦ C and 54% at 20 ◦ C for the LA18, indicating that the PLA fiber was an effective reinforcing agent for PCL. The reinforcing effect was probably due to higher stiffness of the PLA fiber which introduced rigid interface into the composites, which allowed a greater degree of stress transfer at the interface. As for the LA18, the larger surface area at equal fiber loading for the 18 m PLA fibers resulted more stress transfer and thus higher modulus. The increased E also indicated that the heat resistance of PCL was improved, which is useful for the choice of the proper materials for specific applications. For all the samples, with increasing temperature, the E curve exhibited a sharp decrease in the range of −60 to −20 ◦ C due to the glass transition of PCL. Fig. 4b shows the variations of tan ı as a function of temperature. Neat PCL exhibited a relaxation peak at about −46.6 ◦ C, corresponding to the glass transition temperature (Tg ) of PCL. For the composites, there was a very slight increase in the Tg compared to the neat PCL. For the LA18, the Tg value of PCL component was −44 ◦ C, about a 2 ◦ C increase. This slight increase may be indicative of an interfacial interaction between matrix and fiber, hindering the chain mobility and rearrangement [31]. The interfacial interaction was essential to improve the mechanical strength of the composites. What’s more, the height of the tan ı peak decreased with the presence of PLA fibers. It was due to a better fiber/matrix bonding that resulted not only the immobilization of polymer matrix in the presence of the PLA fiber but also an improvement in the hysteresis of the composites thus a reduction in the internal friction [32]. This phenomenon was more obvious for the LA18 due to the larger surface area at equal fiber loading for the PLA fibers of 18 m. 3.4. Static mechanical properties
Fig. 4. The dynamic mechanical traces of neat PCL and various PCL/PLA fiber composites: (a) storage modulus (E ); (b) tan ı versus temperature curves.
frequency with increasing fiber diameter. This suggested that a greater amount of energy was required for the more heterogeneous composites produced with larger fibers. The Cole–Cole plot is usually used to study viscoelastic properties of the materials with a relaxation time distribution such as heterogeneous polymeric systems. For many polymer blends, Cole–Cole plots yield two arcs, attributed to two processes of significantly different relaxation times. Fig. 3b shows Cole–Cole plots for neat PCL and PCL/PLA fiber composites at 80 ◦ C. Neat PCL showed only one circular arc. For the composites, a tail or a second circular arc appeared on the right-hand side of the arc, which indicated a second relaxation mechanism. One arc in the low-viscosity region was corresponding to the local dynamic relaxation of PCL, and the tail in the high-viscosity region was indicative of the long-term relaxation of the PCL chains restrained by the PLA fibers. An increasing effect to the high-viscosity region resulted from the composite containing increasing fiber diameters. Moreover, the long-term relaxation of those restrained PCL chains in the composites became more obvious when the fiber diameter increased to 180 m. This also indicated that the hindered effect of the PLA fiber on the movement of the PCL molecular chains was more obvious with the bigger PLA fiber diameter. 3.3. Dynamical mechanical properties Fig. 4 presents the dynamic mechanical spectra of neat PCL and its composites. It is evident from Fig. 4a that there was a notable
The effects of blending with different diameters PLA fibers on the static mechanical properties of PCL were investigated by tensile tests. The stress–strain curves of neat PCL and PCL/PLA fiber composites are shown in Fig. 5. The corresponding data of mechanical properties are summarized in Table 3. Neat PCL showed higher elongation at break and strength at break, but lower strength at yielding and tensile modulus. With the addition of PLA fibers with different diameters, the tensile yielding strength and modulus were improved at different levels. The strength at yielding and tensile modulus increased from 13.7 MPa and 270 MPa for neat PCL to 15.5 MPa and 474 MPa for the composite LA18, respectively, increases of 13% and 76%, respectively. However, with increasing the fiber diameter, the strength at yielding and tensile modulus gradually decreased, which suggested that the smaller the PLA fiber diameter was, the better the reinforcement to PCL. The reinforcement effect may be due to that the high modulus PLA fibers act as carriers of the load and stress which are transferred from the matrix to the fibers giving rise to effective and uniform stress distribution, resulting in an increase in the strength at yielding and tensile modulus of the PCL/PLA fiber composites [20]. For the composites with uniform fiber composition, the more efficient load transfer may be due to the larger surface area of the thinner fiber diameters [24], which lead to the more obvious reinforcement effect for the thin fiber compared to thick fiber. However, with increasing the fiber diameter, the elongation at break and strength at break gradually decreased. As was shown in Fig. 5a, there was a strain-hardening phenomenon after the yielding of neat PCL due to the orientation of PCL molecules during tensile tests, and the strain-hardening ability was very strong, which lead to a high strength at break. As for the PCL/PLA fiber composites, the strain-hardening capacity
54
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Table 2 Storage modulus values of the PCL/PLA fiber composites at different temperatures. Samples
Storage modulus at −20 ◦ C (MPa)
Storage modulus at 20 ◦ C (MPa)
Reinforcement imparted by the PLA fibers at 20 ◦ C (%)
PCL LA18 LA26 LA180
680 1024 745 875
424 653 477 560
− 54 12.5 32
Table 3 Tensile properties of neat PCL and PCL/PLA fiber composites with various fiber diameters. Sample
Strength at yielding (MPa)
PCL LA18 LA26 LA180
13.7 15.5 15.2 12.3
± ± ± ±
0.8 0.1 0.3 0.4
Tensile modulus (MPa) 270 474 390 319
± ± ± ±
Elongation at break (%)
7 21 8 21
of the composites decreased gradually with the increasing fiber diameters. When the diameter of the PLA fibers increased to 180 m, the strain-hardening phenomenon disappeared. It was because that the PLA fibers hindered the molecular reordering of the PCL polymer chains which was more pronounced in larger fibers than small, thereby contributing to the overall reduction in the elongation at break and the strain-hardening ability [33]. In addition, because of the discontinuous reinforcement of the short fiber and the immiscibility of the PCL and PLA fiber, the stress and strain at break dropped rapidly with increasing the fiber diameter.
1544 742 441 40
± ± ± ±
Strength at break (MPa) 49.4 ± 1.4 21.6 ± 0.3 12.7 ± 1.0 –
52 10 41 5
However, for the composites with PLA fiber diameter of 18 and 26 m, it still exhibited high ductility. Fig. 6 shows the longitudinal fracture surface morphology of PCL/PLA fiber composites after tensile tests. The fracture surfaces for neat PCL, LA18, and LA26 exhibited orderly arranged fibrillar bundle structures parallel to the stretching direction, which contributed to the significantly higher strength and toughness of the composites [29,34]. With increasing the fiber diameters, the fibrous structure became unobvious. When the PLA fiber diameter increased to 180 m, the fibrillar bundle structures disappeared. The evidence provided previously was consistent with that the mobility of the macromolecular chains and molecular orientation of the PCL during tension were impeded due to the presence of PLA fibers in the composites, thereby contributing to the overall reduction in the elongation at break and the strain-hardening ability. And the impeding effect was increased with increasing the PLA fiber diameters. That also confirmed that the |*| of the composites discussed above was increased with the increase of the PLA fiber diameters due to its hindered effect on the mobility of the PCL molecules. As is known to all, the molecular arrangement of polymer chains can be modified through stretching leading to higher crystallinity, the so-called stress-induced crystallization. Fig. 7 shows the DSC thermograms of the first heating traces before and after tensile tests for PCL/PLA fiber composites with different PLA fiber diameters, and the c of PCL was calculated as listed in Table 4. From Table 4, for all the samples, the melting temperature (Tm ) of the PCL component after tensile tests moved to a higher temperature, and the melting enthalpy (Hm ) and c increased, indicating the stress induced the orientation of the molecular chains and hence promoted the crystallization of the PCL component during drawing [34]. The c of neat PCL was increased from 40.9% before tensile tests to 58.2% after tensile tests, which increased significantly by 42.3%. While for the composites LA18, LA26 and LA180, the c of the PCL in composites was increased from 40.2%, 42.7% and 44.0% before tensile tests to 56.1%, 49.7% and 49.6% after tensile tests, respectively. The increase of 39.6%, 16.4% and 12.7% was obtained for the composites LA18,
Table 4 Thermal and crystalline properties of PCL in the composites obtained from the first heating traces at a rate of 10 ◦ C/min before and after tensile tests. Sample
Fig. 5. Tensile stress–strain curves of the neat PCL and PCL/PLA fiber composites with various fiber diameters: (b) gives details of tensile stress at low strain of (a).
PCL LA18 LA26 LA180 a
Tm (◦ C)
Hm a (J/g)
c,PCL (%)
Before
After
Before
After
Before
After
58.1 57.8 57.3 57.7
60.0 60.0 59.5 58.3
58.1 57.1 60.6 62.5
82.6 79.6 70.6 70.5
40.9 40.2 42.7 44.0
58.2 56.1 49.7 49.6
Hm are corrected for the content of PCL in the composites.
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
55
Fig. 6. The SEM micrographs of the longitudinal fracture surface of various PCL/PLA fiber composites after tensile tests at a tensile rate of 10 mm/min.
LA26 and LA180 respectively, which was smaller than that of neat PCL. It further confirmed that the PLA fibers hindered the molecular reordering of the PCL polymer chains which was more pronounced in larger fibers than small during tension therefore depressed the process of the crystallization during tensile tests. In addition, the level of the depression effect was more evident with increasing the PLA fiber diameters, which was in accordance with the result of the longitudinal fracture surface morphology of the composites after tensile tests.
Fig. 8 shows the normalized weight loss profiles of the PCL as a result of P. mendocina lipase-catalyzed degradation. As was
shown in Fig. 8, the weight loss of the films increased with time for all samples. There was little difference among the four samples for the enzymatic degradation before three days due to the incubation period for P. mendocina. However, the weight loss for the LA26 and LA180 was significantly increased after 3 days compared to the two other composites. Overall, this indicated that the degree of degradation for the polymer or composites was not differentiated. However, the rate of enzymatic degradation of PCL or PCL composite was affected by the size of composite fibers of PLA present: PCL and small-fiber LA18 were much slower to degrade than the two composites containing larger fibers. It can be further confirmed by the surface and cross-section morphology of neat PCL and the composites as shown in Fig. 9. Before enzymatic degradation, all the films presented an unchanged smooth
Fig. 7. The first heating traces of DSC thermograms for neat PCL and PCL/PLA fiber composites before and after tensile tests at a tensile rate of 10 mm/min.
Fig. 8. The normalized PCL weight loss profiles of PCL/PLA fiber composites as a function of time during the enzymatic degradation.
3.5. Enzymatic degradation
56
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
Fig. 9. The surface and cross-section morphology of neat PCL and the composites before and after enzymatic degradation for different times: surface morphology of (a1 –a3 ) neat PCL; (b1 –b3 ) LA18; (c1 –c3 ) LA26; (d1 –d3 ) LA180 after degradation for 0, 3, and 6 days, respectively; (a4 –d4 ) cross-section morphology of neat PCL and the composites after enzymatic degradation for 6 days, respectively.
surface. While the PLA fibers in all the samples were exposed with time due to the enzymatic degradation of the PCL. When the composites were degraded for 6 days, the matrix around the fiber already disappeared due to the serious surface erosion, which led to the large crack near the fibers especially for the LA26 and LA180 as shown in Fig. 9c3 and d3 . This indicated that PLA fibers and their interface could act as channels to promote the water entrance and diffusion from the surface to the inside of the composites and serve as a support for the attack of the enzyme molecules, which accelerated the degradation of the PCL matrix around the fiber [20]. As for the LA18, the better interface bonding force between the PCL and the PLA fiber hindered the entrance and the diffusion of enzyme molecules, therefore having little influence on the enzymatic degradation of the PCL. This was also consistent with the cross-section morphology of the composites as shown in Fig. 9a4 –d4 . The film thickness of neat PCL decreased from 300 m (the original film thickness) to about 203 m after degraded for 6 days, and the film thickness of LA18 was similar with neat PCL under the same conditions. As for the other composites, the film thickness decreased to about 83 m for the LA26 and 58 m for the LA180, respectively. Consequently, the variation of weight loss with enzymatic degradation time, the changes of surface morphology, and the reduced film thickness, suggested that the enzymatic degradation of neat PCL and the PCL/PLA fiber composites was likely to proceed via surface erosion mechanism, and the presence of PLA fibers with bigger fiber diameters promoted the enzymatic degradation of PCL.
4. Conclusions Biodegradable polymer composites based on PCL and PLA fibers with various fiber diameters were prepared by melt compounding. The effects of PLA fiber diameter on the morphology, rheological properties, mechanical properties and enzymatic degradation of PCL composites were investigated in detail. The SEM micrographs of the cryo-fractured surfaces of PCL/PLA fiber composites showed the fiber breakage rather than pullout indicating good interfacial adhesion between the PLA fiber and the PCL matrix. With increasing PLA fiber diameters, the |*|, G and G increased at the measured temperatures, especially at low frequencies. This could be explained that the presence of the PLA fibers hindered the molecular mobility of the PCL polymer chains, and for the composites with larger diameter fiber, more stress was required for the system to overcome the resistance of the PLA fiber. This effect was rationalized by the presence of PLA fibers in the composites hindering the mobility of the macromolecular chains and molecular orientations during tension, thereby contributing to the overall reduction in the elongation at break, strain-hardening ability, and the stress-induced crystallization and hence strength at break. However, the dynamic storage modulus, tensile yielding strength and modulus of the PCL were improved with the addition of the PLA fibers, especially for the 18 m-diameter fiber. This indicated the smaller the reinforcing PLA fiber diameter was, the better the reinforcement of PCL. The enzymatic degradation of the PCL was accelerated when the PLA fiber diameters were 26 and 180 m. While for the fiber of 18 m,
D. Ju et al. / International Journal of Biological Macromolecules 76 (2015) 49–57
there was little influence on the enzymatic degradation due to the better interface bonding force between the PCL and the PLA fiber. Such a PLA fiber-reinforced PCL composite could be used in the packaging field to alleviate white pollution. Additionally, the reinforced PCL by PLA fiber still can be used in a biomedical field due to the biodegradability and biocompatibility of both the PCL and PLA. Acknowledgement The authors are grateful for the support from the National Natural Science Foundation of China (51021003). References [1] Y. Ikada, H. Tsuji, Macromol. Rapid Commun. 21 (2000) 117–132. [2] C.S. Wu, J. Appl. Polym. Sci. 115 (2010) 948–956. [3] M.C.A.M. Leite, C.R.G. Furtado, L.O. Couto, F.L.B.O. Oliveira, T.R. Correia, Polimeros 20 (2010) 339–344. [4] X.Y. Qiao, W. Li, H. Watanabe, K. Sun, X.D. Chen, J. Polym. Sci. B: Polym. Phys. 47 (2009) 1957–1970. [5] W. Li, X.Y. Qiao, K. Sun, X.D. Chen, J. Appl. Polym. Sci. 110 (2008) 134–139. [6] A. Arbelaiz, B. Fernandez, A. Valea, I. Mondragon, Carbohydr. Polym. 64 (2006) 224–232. [7] M. Dauda, M. Yoshiba, K. Miura, S. Takahashi, Adv. Compos. Mater. 16 (2007) 335–347. [8] C.S. Wu, H.T. Liao, Ind. Eng. Chem. Res. 51 (2012) 3329–3337. [9] B.Q. Chen, K. Sun, T. Ren, Eur. Polym. J. 41 (2005) 453–457. [10] A.L. Yang, R.J. Wu, J. Appl. Polym. Sci. 84 (2002) 486–492.
57
[11] A. Yang, K. Sun, R. Wu, 45th International SAMPE Symposium and Exhibition on Bridging the Centuries with SAMPEs Materials and Processes Technology, vol. 45, 2000, pp. 427–432. [12] P. Pan, B. Zhu, T. Dong, S. Serizawa, M. Iji, Y. Inoue, J. Appl. Polym. Sci. 107 (2008) 3512–3519. [13] N. Teramoto, K. Urata, K. Ozawa, M. Shibata, Polym. Degrad. Stab. 86 (2004) 401–409. [14] M. Shibata, R. Yosomiya, N. Ohta, A. Sakamoto, H. Takeishi, Polym. Polym. Compos. 11 (2003) 359–367. [15] Y. Cao, K. Goda, S. Shibata, Adv. Compos. Mater. 16 (2007) 283–298. [16] Z. Marrakchi, H. Oueslati, M.N. Belgacem, F. Mhenni, E. Mauret, Compos. A 43 (2012) 742–747. [17] M.Z.A. Hamid, N.A. Ibrahim, W.M.Z.W. Yunus, K. Zaman, M. Dahlan, J. Reinf. Plast. Compos. 29 (2010) 2723–2731. [18] K.J. Kim, J.L. White, Compos. Interfaces 16 (2009) 619–637. [19] D.D. Ju, L.J. Han, F. Li, S. Chen, L.S. Dong, Polym. Compos. 34 (2013) 1745–1752. [20] D.D. Ju, L.J. Han, F. Li, S. Chen, L.S. Dong, Int. J. Biol. Macromol. 67 (2014) 343–350. [21] M. Obukuro, Y. Takahashi, H. Shimizu, Dent. Mater. J. 27 (2008) 541–548. [22] M.B. Rezvani, M. Atai, F. Hamze, Indian J. Dent. Res. 24 (2013) 237–241. [23] S. Houshyar, R.A. Shanks, Polym. Int. 53 (2004) 1752–1759. [24] T.N. Abraham, K.E. George, J. Thermoplast. Compos. Mater. 22 (2009) 5–20. [25] B. Mu, Z.J. Li, Key Eng. Mater. 145–149 (1998) 607–612. [26] H. Tsuji, Y. Ikada, J. Appl. Polym. Sci. 60 (1996) 2367–2375. [27] V. Crescenzi, G. Manzini, G. Calzolari, C. Borri, Eur. Polym. J. 8 (1972) 449–463. [28] Y. Wang, F. Li, Z.Y. Wang, D.B. Liu, H.M. Xia, L.F. Liu, S. Chen, Chem. Res. Chin. Univ. 28 (2012) 459–464. [29] M. Hughes, J. Carpenter, C. Hill, J. Mater. Sci. 42 (2007) 2499–2511. [30] R. Guo, J. Azaiez, C. Bellehumeur, Polym. Eng. Sci. 45 (2005) 385–399. [31] H. Anuar, S.H. Ahmad, R. Rasid, A. Ahmad, W.N. Wan Busu, J. Appl. Polym. Sci. 107 (2008) 4043–4052. [32] M.S. Huda, A.K. Mohanty, L.T. Drzal, E. Schut, M. Misra, J. Mater. Sci. 40 (2005) 4221–4229. [33] S. Jain, M.M. Reddy, A.K. Mohanty, M. Misra, A.K. Ghosh, Macromol. Mater. Eng. 295 (2010) 750–762. [34] F. Feng, X.W. Zhao, L. Ye, J. Macromol. Sci. Phys. 50 (2011) 1500–1507.
