- Email: [email protected]
PHB-HHx composites
Materials Science and Engineering C 33 (2013) 2131–2137
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Effect of composition ratio on the thermal and physical properties of semicrystalline PLA/PHB-HHx composites Jung Seop Lim a,⁎, Ku-il Park a, Gun Soo Chung a, Jong Hoon Kim b a b
R&D Center, LG Hausys, Ltd, 533, Hogae-1-dong, Dongan-gu, Anyang-city, Gyoungki-do, Republic of Korea Korea High Tech Textile Research Institute, 666-2, Sangsu-ri, Nam-myun, Yangju-si, Gyoungki-do, Republic of Korea
a r t i c l e
i n f o
Article history: Received 8 September 2012 Received in revised form 26 December 2012 Accepted 15 January 2013 Available online 20 January 2013 Keywords: Bioplastic composites PLA PHB-HHx Miscibility Morphology
a b s t r a c t In this study, composites of semicrystalline, biodegradable polylactide (PLA) and poly(3-hydroxybutyrate-co3-hydroxyhexanoate) (PHB-HHx) were prepared by direct melt compounding. The physical and thermal properties of the composites were investigated as a function of the composition ratio. Differential scanning calorimetry analysis indicated that PLA and PHB-HHx formed immiscible composites over the observed range of composition. The crystallization of PLA was gradually suppressed by increasing proportions of PHB-HHx. Dynamic mechanical analysis results confirmed that the innate ductility of PHB-HHX and its inhibiting effect on PLA crystallization improved the stiffness of the composite compared to those of neat PLA. The infrared spectra of the immiscible PLA/PHB-HHx composites at two crystallization temperatures (30 °C, 130 °C) were obtained and presented. At 30 °C, PHB-HHx existed as crystalline domains in the PLA matrix, while, amorphous phase of molten PHB-HHx was diffused within the crystalline phase of PLA at 130 °C. The interaction between PHB-HHX and PLA could not be elucidated from the temperature data. Mechanical tests showed that the addition of PHB-HHx improves ductility of PLA/PHB-HHx composite. Morphological analysis revealed that small proportions of PHB-HHx exhibited less tendency to aggregate, which resulted in greater plastic deformation and improved toughness. From this study, PLA blended with small portions of PHB-HHx may further expand the use of bio-friendly resources in a variety of applications such as flexible films, food packaging and something like that. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Concerns about the potentially damaging effects of nondegradable plastics on the environment and the increasing cost of petrochemical feedstocks have raised new challenges in polymer research. Biodegradable polymers such as polylactide (PLA) or polyhydroxyalkanoates (PHA) are traditional alternatives to petroleum-based thermoplastics and represent an interesting and growing market. Poly(3-hydroxybutyrate) (PHB), discovered by Lemoigne in 1925, is the most representative member of the PHA family. However, it is rigid and stiff due to high crystallinity and relatively large spherulites [1,2]. Furthermore, PHB is thermally unstable during conventional melt processing and only suitable for processing in a narrow temperature window [3]. To overcome these drawbacks, several PHB copolymers have been biosynthesized by a variety of microorganisms. Recently, poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHB-HHx) was synthesized from oils and fats using Aeromonas sp. OL338 and Aeromonas sp. FA440 [4]. This new co-polyester provides an opportunity to circumvent the shortcomings of conventional PHAs such as PHB and PHBV. Randomly distributed second monomer units (3HHx) are excluded from the 3HB crystalline lattice, forming a short chain branch. ⁎ Corresponding author. Tel.: +82 31 450 4452; fax: +82 31 450 4444. E-mail address: [email protected] (J.S. Lim). 0928-4931/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2013.01.030
This branch acts as a molecular defect, which disrupts the excessive regularity of the polymer chain and consequently lowers the melt temperature (Tm) and crystallinity. This unique characteristic may induce substantially more flexibility and ductility, which would make PHB-HHx more suitable for making films and other soft articles [5,6]. Polylactide (PLA) is synthesized from lactic acid, which is obtained from renewable resources such as starch from corn or maize. PLA is now produced on a large scale and used in various applications including packaging, agriculture, and textiles. However, its high brittleness, slow crystallization rate, and low heat distortion temperature restrict its widespread use [7,8]. To overcome these limitations, many researchers have examined ways of modifying PLA, such as copolymerization or composite formation [9–12]. Creating composites of PLA with other polymers may be the most credible methodology, since this process allows the incorporation of new materials with enhanced properties as well as is less expensive than chemical modifications or syntheses of tailor-made macromolecules. In a composite system, the selection of individual components and their relative proportions must be carefully considered, as these factors can have a strong influence on the physical and thermal properties of the final materials [13,14]. In addition, the relationship between morphology and composition must be clearly investigated, since morphology depends on the miscibility of the components and can
2132
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
affect the overall properties of the composite [15,16]. Compared to PLA and other PHA family composites [17,18], very little has been reported on PLA/PHB-HHx composites. Moreover, no systematic research has been published concerning the impact of composition ratio on the physical, thermal, and morphological properties of these composites. In this work, PLA/PHB-HHx composites were prepared via direct melt compounding. The morphological, thermal, and physical properties of the composites were evaluated as a function of composition ratio. Commercial polymers without any compatibilizer were used to confirm innate polymer properties, and the neat polymers were examined first to determine processing conditions. We hypothesized that a composite of PLA, which is naturally hard and brittle, with the more ductile PLA would effectively balance the shortcomings of the individual polymers. 2. Experimental section 2.1. Materials and sample preparation A commercial linear and semicrystalline PLA (Ingeo™, 2003D) was provided in pellet form by NatureWorks LLC (Blair, NE, USA). According to the supplier, the D-isomer content of the PLA was 4.3%, with a melt flow rate of 5–7 g/10 min and a density of 1.24 g/cm 3. Additionally, this semi-crystalline PLA exhibited a weight-average molecular weight (Mw) of 210,000 g/mol and polydispersity of 1.74 as determined by gel permeation chromatography. The bacterial polyester, poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHB-HHx, Aonilex®), was supplied by the Kaneka Corporation (Osaka, Japan). It contained a 3-hydroxy hexanoate content of about 11 mol% and a weight average molecular weight (Mw) of 1,388,000 with a polydispersity of about 2.0. Before manufacturing the composite samples, the two biodegradable polymers were dried at 60 °C overnight. Fig. 1 shows the chemical structures of PLA and PHB-HHx. PLA/PHB-HHx composites were prepared by direct melt compounding using a twin roll counterrotating mixer (Thermo Haake, Paramus, NJ, USA). The barrel temperature was 180 °C and the screw speed was 100 rpm. The total mixing time was 3 min with the screw speed of 100 rpm at a barrel temperature of 180 °C. The mixtures produced were then compression molded in a hydraulic compression press for 5 min at 180 °C with pressure of 6000 psi. The weight ratios of PLA/PHB-HHx were 10/0, 9/1, 8/2, 6/4, and 0/10, respectively. 2.2. Measurements Differential scanning calorimetry (DSC) data from samples were obtained in a DSC7; Perkin-Elmer, Waltham, MA, USA. Samples of 5 mg taken from the central sections of test specimens were sealed in aluminum pans and subjected to heating from − 40 °C to 200 °C at 5 °C/min followed by holding at 200 °C for 3 min before rapidly cooling to − 40 °C. Samples were reheated to 200 °C with the same
Fig. 1. Chemical structures of PLA and PHB-HHx.
heating rate. The dynamic mechanical properties were measured on a dynamic mechanical analyzer (DMA 2980; TA Instruments, New Castle, DE, USA) set in tension mode at a frequency of 1 Hz in a nitrogen atmosphere. Samples (15 × 5 × 0.5 mm) were cooled to −40 °C and then heated gradually to 200 °C at a rate of 2 °C min−1. For infrared (IR) analyses, PLA/PHB-HHx composites were crystallized using the following procedure. First, a Mettler-Toledo FP82HT hot stage (Mettler-Toledo, Greifensee, Switzerland) was equilibrated at the desired crystallization temperature. Each sample was then melted on the hot stage at 200 °C for 5 min followed by quickly transferring to a second hot stage. Crystallization occurred over 30 min. Fourier transform (FT)-IR absorption spectra were obtained on the crystallized samples at room temperature using an IFS 88-IR spectrometer (Bruker AXS GmbH, Karlsruhe, Germany). Spectra were acquired from 4000 to 400 cm−1 at a resolution of 2 cm−1, and 16 scans were averaged for each sample. Mechanical tests were conducted at room temperature with a tensile testing machine (Instron 4465; Instron, Norwood, MA, USA) based on ASTM D638. Crosshead speed was set to 10 mm/min. The samples were prepared on a dumbbell-shaped film with the following dimensions: 3.5 mm, 14 mm and 0.3 mm in width, gage length and thickness, respectively. Each value from 10 successful measurements was averaged and expressed with a standard deviation (SD). The morphologies of PLA/PHB-HHx were investigated by scanning electron microscopy, SEM, (JSM-6700F; JEOL, Tokyo, Japan). Crosssections of gold-coated specimens were examined after fracturing in liquid nitrogen. Surface morphology was examined in air using atomic force microscopy (AFM; 2010 Discoverer; Topometrix, Santa Clara, CA, USA) in contact mode with silicon nitride tips. Images were recorded at scan speeds of 10 mm/s using the minimal applied force.
3. Results and discussion Fig. 2 shows second heating differential scanning calorimetry (DSC) thermograms of neat PLA, PHB-HHx and their composites, respectively. Neat PLA yielded a broad crystallization peak at 113 °C and a melting peak at about 150 °C. For neat PHB-HHx, a sharp exothermal peak and double melting peak were observed at about 50 °C and 100–130 °C, respectively, indicating that PHB-HHx crystallizes more rapidly than PLA at a lower temperature. The glass transition temperatures, Tg, of the two biopolymers were −2 °C and 60 °C, indicating the segmental mobility of PHB-HHx increases at a low temperature, resulting in a faster crystallization rate compared to that of PLA. In addition to, at 40% PHB-HHx, the Tg of PLA cannot be discerned due to the crystallization peak of PHB-HHx. However, at other composite proportions, the Tg of each biopolymer can be observed at 60 °C and −2 °C. No additional evidence of miscibility of polymers, such as a remarkable melting point depression [19,20]. For these reasons, we suggest that PLA and PHB-HHx are immiscible. Detailed investigation of miscibility between PLA and PHB-HHx was performed by dynamic mechanical analysis (DMA). With increasing proportions of PHB-HHx, two crystallization and melting peaks, corresponding to PLA and PHB-HHx, became apparent. This result suggests that this system can be classified as a semicrystalline/ semicrystalline composite. Moreover, 10 wt.% and 20 wt.% additions of PHB-HHx resulted in a shift of the PLA crystallization peak to higher temperature, while the heat of fusion of PLA was decreased. This may presumably due to restriction of the chain mobility of PLA by PHB-HHx, resulting in slower crystallization kinetics in the composites. The long chain length of PHB-HHx may affect entanglement of PLA, thereby reducing the crystallization rate and decreasing regularity of the PLA crystal. In the case of 40% PHB-HHx sample, the crystallization peak of PHB-HHx shifted to a higher temperature compared to the corresponding peak for neat PHB-HHx, which suggests that the PLA component may also restrict the molecular mobility of PHB-HHx in the composite materials. We will plan to investigate the crystallization kinetic of PLA/PHB-HHx
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
2133
Fig. 2. Second heating DSC thermograms of neat PLA, PHB-HHx and their composites.
composites, in detail, by analyzing the isothermal and non-isothermal crystallization experiments. Fig. 3(a) shows the storage modulus of PLA and PLA/PHB-HHx composites as a function of temperature. The high modulus of PLA at room temperature indicates that PLA is brittle, since the storage modulus of a semicrystalline polymer is indicative of material stiffness under shear deformation [21]. This peak drops abruptly near 60 °C and rises around 120 °C due to a glass transition and crystallization of PLA, respectively. Below the Tg region, the storage modulus decreases as the weight percent of PHB-HHx increases in composite, resulting in a reduction of stiffness In the glassy state, the mobility of the molecular chain is restricted and stiffness is related to stored elastic energy [22]. Therefore, these results imply that PHB-HHx may restrict the chain mobility of PLA, resulting in low stored elastic energy and a glassy modulus. From DSC analysis, it can be pointed out that PHB-HHx can hinder the
crystallization of PLA, resulting in lowering the storage modulus of PLA. PHB-HHx, which is composed of a 3-hydroxyhexanonate comonomer with a 3-hydroxybutyrate backbone, is naturally more flexible and ductile than PLA [5,6]. From these results, it is evident that innate properties of PHB-HHx dominate mechanical properties of composites, showing a decreased storage modulus as the weight percent of PHB-HHx is increased. Fig. 3(b) shows loss modulus curves of PLA and PLA/PHB-HHx composites. Generally, the glass transition, which is reflected in the loss modulus peaks, can be detected at about 60 °C in neat PLA. With increasing the weight percent of PHB-HHx, a broad peak related to the Tg of PHB-HHx gradually emerges near 20 °C; this is presumably due to poor miscibility of PLA and PHB-HHx as well as phase separation. In addition, the Tg of PLA did not shift to a lower temperature, regardless of the proportion of PHB-HHx. This indicates that PHB-HHx does not act
2134
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
(a) Storage Modulus 3000
Storage Modulus (MPa)
2500
PLA PLA/PHB-HHx=9:1 PLA/PHB-HHx=8:2
2000
PLA/PHB-HHx=6:4
1500
1000
500
0 -50
0
50
100
150
Temperature (oC) Fig. 4. The carbonyl-stretching region of the IR absorption spectra of PLA/PHB-HHx composites crystallized at 30 °C after melting.
(b) Loss Modulus 600 PLA
Loss Modulus (MPa)
PLA/PHB-HHx=9:1
500
PLA/PHB-HHx=8:2 PLA/PHB-HHx=6:4
400 300 200 100 0 -50
0
50
100
150
Temperature (oC) Fig. 3. DMA result of PLA/PHB-HHx composites.
as a plasticizer in the PLA/PHB-HHx composites. Generally, the addition of a plasticizer will decrease the Tg due to the diffusion within the molecular chains, which enhances flexibility and ductility. Moreover, the extent of this effect enlarges with increasing proportions of plasticizer [23]. Therefore, PHB-HHx does not act as a plasticizer. Instead, its innate ductility and its inhibition of crystallization affect the viscoelastic properties of the composites. Neat PLA exhibited a very sharp and intense tanδ peak (data not shown) due to no restriction in chain motion exists. The height of the tanδ peak decreased with increasing proportions of PHB-HHx, indicating that PHB-HHx may inhibit the chain mobility of PLA, which is consistent with the DSC results. Fig. 4 shows the carbonyl-stretching region of IR absorption spectra of PLA/PHB-HHx composites crystallized at 30 °C after melting. In this region, it is obvious that both PLA and PHB-HHx are in a crystalline state. The spectrum of neat PLA contains a singlet peak near 1750 cm−1, corresponding to the stretching vibration of a carbonyl group [24]. The spectral features of neat PHB-HHx can be divided into two major peaks at 1740 cm−1 and 1723 cm−1, corresponding respectively to amorphous and crystalline carbonyl groups [20]. As the proportion of PHB-HHx was increased, the crystalline peaks, which are related to PHB-HHx, increased gradually. However, no change in the main carbonyl peak, which corresponds to PLA, was observed. This result indicates that PHB-HHx may be more easily crystallized as its proportions in the composite increase, forming phase-separated domains in the PLA matrix.
The existence of hydrogen bonding between PLA and PHB-HHx was investigated by analyzing the C\O\C absorption peak in the IR spectra of the composites. Fig. 5 shows the C\O\C peaks of PLA/PHB-HHx composites crystallized at 30 °C after melting. The spectrum of neat PLA contained the following absorption peaks [24]: 1384 and 1362 cm−1 assigned to \CH\ deformation including symmetric and asymmetric bends, whereas 1267, 1182, 1129, 1090, and 1045 cm−1 attributable to \C\O\C\ stretches. The spectrum of neat PHB-HHx contains characteristic bands at 1288, 1275, 1261, and 1227 cm − 1, corresponding to the C\O\C stretching modes of the crystalline components. These four peaks may be due to helical structures formed by the PHB-HHx backbone. The bands near 1180 cm−1 probably consist of two bands, ascribed to crystalline and amorphous states, respectively [5,6]. With the addition of PHB-HHx, no unique changes were observed except the appearance of peaks related to crystallized PHB-HHx, near 1300–1250 cm−1. Therefore, no evidence of hydrogen bonding between PLA and PHB-HHx was observed. Fig. 6 shows the carbonyl-stretching region of IR spectra of the PLA/PHB-HHx composites crystallized at 130 °C after melting. The composite systems are in a crystalline/amorphous state since the temperature is near the Tm of PHB-HHx. Neat PLA exhibits the same
Fig. 5. C\O\C absorption peaks in the IR spectra of PLA/PHB-HHx composites crystallized at 30 °C after melting.
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
2135
Fig. 8. The stress–strain curve of PLA/PHB-HHx composites. Fig. 6. The carbonyl-stretching region of the IR absorption spectra of PLA/PHB-HHx composites crystallized at 130 °C after melting.
carbonyl peak regardless of temperature, while the carbonyl peak of neat PHB-HHx is observed at 1740 cm −1 due to the broad amorphous band of molten PHB-HHx. As the PHB-HHx content increases, the carbonyl peak shifts to lower energy state. The vibrational frequency of a functional group is usually determined by the force constant and mass of the bonded atoms. Any interaction with the carbonyl group may result in a decrease in bond energy, which may have lowered the force constant followed by shifting the peak to lower frequency [19,25]. Recently, several research groups have reported the appearance of a C\H\O hydrogen bond between the CH3 group of one helical structure and C_O groups in the neighboring helical structure within PLA and PHA [26,27]. Zhang et al. showed that a PLLA/PDLA stereocomplex can form intermolecular hydrogen bonds between CH3/O_C groups as evidenced by low frequency shifts of the C_O stretching band [26]. Therefore, the above results suggest that a hydrogen bond may also exist between crystalline PLA and amorphous PHB-HHx, although the formation of a hydrogen bond could not be fully ascertained based only on the carbonyl peak. A hydrogen bonding between PLA and PHB-HHx could also result in some change near the C\H peak, such as shifting or creation of shoulder peak. Fig. 7 shows the C\H stretch region of PLA/PHB-HHx composites crystallized at 130 °C after melting. No
changes were observed near the C\H stretch peak, regardless of PHB-HHx content. Therefore, we suggest that little significant hydrogen bonding presents between crystalline PLA and amorphous PHB-HHx; also, the hydrogen bonding effects are much weaker than those observed in the PLLA/PDLA stereocomplex. For the next time, our research groups have a plan to investigate IR analysis by using 2-D or 3-D IR instrument. The stress–strain behavior of the PLA/PHB-HHx composites is shown in Fig. 8. Neat PLA fails as soon as its yield stress exceeded. Neat PLA has a strong strain softening that is not stabilized by strain hardening; in tensile tests, strain softening may stimulate strain localization, which causes the buildup of local, multiaxial stresses. If the local strain is not delocalized, it induces void nucleation and cracks in the matrix, which lead to a brittle failure behavior [28,29]. In the contrast, composites containing PHB-HHx greater than 20 wt.% show stable neck growth and strain hardening, indicating that PHB-HHx can be used to improve the ductility of PLA. The mechanical properties of the PLA/PHB-HHx composites are summarized in Table 1. Effect of PHB-HHx was much more pronounced for mechanical properties of composites; yield stress and Young's modulus were decreased to 54 MPa and 1400–1450 MPa, respectively, by adding 10 wt.% of PHB-HHx. Toughness and elongation at break were 4.0 MPa and 7.6%, respectively, demonstrating that only a small amount of PHB-HHx is effective in reducing the stiffness of PLA. At 20 wt.% PHB-HHx, the ductility of the composite was significantly enhanced; however, 40 wt.% of PHB-HHx/PLA composite had a lower ductility. Several criteria must be met to improve the toughness of PLA or any other polymers. Polymers those that contribute to enhance toughness must be uniformly distributed throughout the PLA matrix with small size of domains; large size of domains are less effective on absorbing and dissipating energy under mechanical stress [30,31]. Domain size and phase separation were investigated by AFM and SEM (shown in Figs. 9, 10). The composites exhibited larger domains and rougher surfaces than those of neat PLA and PHB-HHx. Surface roughness, including Ra (average roughness) and Rq (RMS roughness) is given in Table 2. Among the tested samples, neat PLA exhibited a higher roughness than neat PHB-HHx, with values of 111.9 nm and Table 1 Mechanical properties of PLA/PHB-HHx composites. Tensile Young's Toughness strength (MPa) modulus (MPa) (MPa)
Fig. 7. The C\H stretch region of the IR spectra of PLA/PHB-HHx composites crystallized at 130 °C after melting.
PLA 62.2 (±3.5) PLA/PHB-HHx=9/1 54.1 (±2.7) PLA/PHB-HHx=8/2 45.3 (±3.3) PLA/PHB-HHx=6/4 40.1 (±2.4) PHB-HHx 21.6 (±1.8)
1603.0 (±7.0) 1416.0 (±6.0) 1265.0 (±8.0) 1093.0 (±7.0) 309.0 (±4.0)
Elongation at break (%)
3.2 (±0.5) 3.6 (±0.3) 4.0 (±0.3) 7.6 (±0.6) 68.7 (±2.7) 113.1 (±2.5) 20.4 (±1.4) 37.6 (±1.3) 160.6 (±3.4) 524.8 (±3.7)
2136
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
(a) PLA
(b) PLA/PHB-HHx=8/2
(c) PLA/PHB-HHx/6/4
(d) PHB-HHx
Fig. 9. AFM images of the PLA/PHB-HHx composites.
(a) PLA
(b) PHB-HHx
(c) PLA/PHB-HHx=8/2
Fig. 10. SEM micrographs of the PLA/PHB-HHx composites.
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137 Table 2 Surface roughness of PLA/PHB-HHx composites. Parameter Sample PLA Ra(nm) Rq(nm)
PLA/PHB-HHx=8/2 PLA/PHB-HHx=6/4 PHB-HHx
111.9 (±0.5) 98.2 (±0.2) 132.9 (±0.3) 127.2 (±0.6)
120.7 (±0.5) 144.3 (±0.5)
49.6 (±0.6) 62.1 (±0.5)
49.6 nm, respectively. This result is supported by the enhanced ductility and flexibility of the composite materials as described above. Although neat PHB-HHx has a lower surface roughness than PLA, the roughness of PLA/PHB-HHx composites containing high weight percent of PHB-HHx is increased. PLA/PHB-HHx with the ratio of 6:4 had the greatest surface roughness (Ra = 120.7 nm). This implies that phase separation between PLA and PHB-HHx becomes more pronounced with increasing proportions of PHB-HHx. Noda et al. reported that a very ductile polymer alloy can be produced when a small amount of PHB-HHx is mixed with hard and brittle PLA [32]. Their findings showed that fine dispersions of small PHA particles are created in such composites as long as the proportion of PHA does not exceed about 20 wt.%. As shown in Fig. 10, SEM micrographs showed that neat PHB-HHx exhibited a much smoother surface than neat PLA which is consistent with the result is consistent with the AFM images. In the 8/2 PLA/ PHB-HHx composites, small domains of PHB-HHx were homogeneously dispersed in the PLA matrix. The interface between PLA and PHB-HHx domains is vague, confirming that phase separation between the two components is not absolute. We also analyzed the SEM of PLA/PHB-HHx (6:4). Compared with PLA/PHB-HHx (8:2), it was observed that PHB-HHx domain size was large and phase separation between PLA and PHB-HHx (data not shown). Moreover, this result may be a good agreement with AFM data. These findings indicate that small amounts of PHB-HHx are less likely to aggregate, resulting in a relatively large overall surface area in contact with the PLA matrix. Consequently, the tensile energy was efficiently dissipated by the elastic PHB-HHx phase, leading to greater plastic deformation and improved toughness of the composite over that of neat PLA. 4. Conclusions Binary bioplastic composites of PLA and PHB-HHx were prepared by melt mixing. Mechanical and thermal properties and surface morphology were investigated as a function of the composition ratio. DSC analyses indicated that PLA and PHB-HHx formed immiscible composites over the observed range of compositions. The crystallization of PLA was suppressed with increasing proportions of PHB-HHx. DMA results confirmed that PHB-HHX could be used to improve the stiffness of PLA due to their innate ductility and inhibiting effect on crystallization. At 30 °C, the crystallinity of PHB-HHx domains in the composites increased with greater proportions of PHB-HHx. At 130 °C, an amorphous
2137
phase of molten PHB-HHx diffused within the crystalline phase of PLA. However, there is no evidence of a direct chemical interaction between PLA and PHB-HHx within the composites. Mechanical tests showed that the material changed from brittle (PLA) to ductile failure with the addition of PHB-HHx, although no optimal proportion of PHB-HHx was determined that maximized toughness and elongation at break without severe loss in tensile strength and modulus. Morphological analyses revealed that small amounts of PHB-HHx were less likely to aggregate, resulting in greater plastic deformation and enhanced toughness in the final composite. Thus, for industrial applications, the toughness of PLA can be improved with the addition of small amounts of PHB-HHx.
References [1] A.J. Anderson, E.A. Dawes, Microbiol. Rev. 54 (1990) 450. [2] M. Kunioka, Y. Doi, Macromolecules 23 (1990) 1933. [3] S.F. Williams, D.P. Martin, D.M. Horowitz, O.P. Peoples, Int. J. Biol. Macromol. 25 (1999) 111. [4] Y. Doi, S. Kitamura, H. Abe, Macromolecules 28 (1995) 4822. [5] I. Noda, P.R. Green, M.M. Satkowski, L.A. Schechtman, Biomacromolecules 6 (2005) 580. [6] I. Noda, E.B. Bond, P.R. Green, D.H. Melik, K. Narasimhan, L.A. Schechtman, Polym. Biocatal. Biomater. 900 (2005) 280. [7] K.A.M. Thakur, R.T. Kean, J.M. Zupfer, N.U. Buehler, M.A. Doscotch, E.J. Munson, Macromolecules 29 (1996) 8844. [8] H. Urayama, T. Kanamori, K. Fukushima, Y. Kimura, Polymer 44 (2003) 5635. [9] Y.L. Zhao, Q. Cai, X.T. Shuai, J.Z. Bei, C.F. Chen, F. Xi, Polymer 43 (2002) 5819. [10] E.S. Kim, B.C. Kim, S.H. Kim, J. Polym. Sci., Part B: Polym. Phys. 42 (2004) 939. [11] H. Liu, F. Chen, B. Liu, G. Estep, J. Zhang, Macromolecules 43 (2010) 6058. [12] L. Jiang, M.P. Wolcott, J. Zhang, Biomacromolecules 7 (2006) 199. [13] G.I. Williams, R.P. Wool, Appl. Compos. Mater. 7 (2000) 421. [14] K. Oksman, J.F. Selin, in: F.T. Wallenberger, N.E. Weston (Eds.), Natural Fibre: Plastics and Composites, Kluwer Academic Publishers, Dordrecht Boston London, 2004, p. 149. [15] P. Martin, C. Maquet, R. Legras, C. Bailly, L. Leemans, M. van Gurp, Polymer 45 (2004) 5111. [16] P. Martin, C. Maquet, R. Legras, C. Bailly, L. Leemans, M. van Gurp, Polymer 45 (2004) 3277. [17] V. Christian, G.H. Günter, W.S. Heinz, Vib. Spectrosc. 49 (2009) 284. [18] T. Gerard, T. Budtova, Eur. Polym. J. 48 (2012) 1110. [19] J.S. Lim, I. Noda, S.S. Im, J. Polym. Sci., Part B: Polym. Phys. 44 (2006) 2852. [20] J.S. Lim, I. Noda, S.S. Im, Eur. Polym. J. 44 (2008) 1428. [21] J.S. Lim, J.H. Kim, Mater. Sci. Eng. A 527 (2010) 4641. [22] H. Ishida, D.J. Allen, Polymer 37 (1996) 4487. [23] Z. Kulinski, E. Piorkowska, Polymer 46 (2005) 10290. [24] Y. Gao, L. Kong, L. Zhang, Y. Gong, G. Chen, N. Zhao, X. Zhang, Eur. Polym. J. 42 (2006) 764. [25] J.S. Lim, I. Noda, S.S. Im, Polymer 48 (2007) 2745. [26] J. Zhanga, H. Sato, H. Tsujib, I. Noda, Y. Ozaki, J. Mol. Struct. 735 (2005) 249. [27] H. Sato, R. Murakami, A. Padermshoke, F. Hirose, K. Senda, I. Noda, Macromolecules 37 (2004) 7203. [28] P. Ma, D.G. Hristova-Bogaerds, J.G.P. Goossens, A.B. Spoelstra, Y. Zhang, P.J. Lemstra, Eur. Polym. J. 48 (2012) 146. [29] B. Meng, J. Deng, Q. Liu, Z. Wu, W. Yang, Eur. Polym. J. 48 (2012) 127. [30] K. Cho, J.H. Yang, C.E. Park, Polymer 39 (1998) 3073. [31] P. Nawadon, C. Sirijutaratana, E. Nukul, Mater. Sci. Eng. A 532 (2012) 64. [32] I. Noda, M.M. Satkowski, A.E. Dowrey, C. Marcott, Macromol. Biosci. 4 (2004) 269.
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Effect of composition ratio on the thermal and physical properties of semicrystalline PLA/PHB-HHx composites Jung Seop Lim a,⁎, Ku-il Park a, Gun Soo Chung a, Jong Hoon Kim b a b
R&D Center, LG Hausys, Ltd, 533, Hogae-1-dong, Dongan-gu, Anyang-city, Gyoungki-do, Republic of Korea Korea High Tech Textile Research Institute, 666-2, Sangsu-ri, Nam-myun, Yangju-si, Gyoungki-do, Republic of Korea
a r t i c l e
i n f o
Article history: Received 8 September 2012 Received in revised form 26 December 2012 Accepted 15 January 2013 Available online 20 January 2013 Keywords: Bioplastic composites PLA PHB-HHx Miscibility Morphology
a b s t r a c t In this study, composites of semicrystalline, biodegradable polylactide (PLA) and poly(3-hydroxybutyrate-co3-hydroxyhexanoate) (PHB-HHx) were prepared by direct melt compounding. The physical and thermal properties of the composites were investigated as a function of the composition ratio. Differential scanning calorimetry analysis indicated that PLA and PHB-HHx formed immiscible composites over the observed range of composition. The crystallization of PLA was gradually suppressed by increasing proportions of PHB-HHx. Dynamic mechanical analysis results confirmed that the innate ductility of PHB-HHX and its inhibiting effect on PLA crystallization improved the stiffness of the composite compared to those of neat PLA. The infrared spectra of the immiscible PLA/PHB-HHx composites at two crystallization temperatures (30 °C, 130 °C) were obtained and presented. At 30 °C, PHB-HHx existed as crystalline domains in the PLA matrix, while, amorphous phase of molten PHB-HHx was diffused within the crystalline phase of PLA at 130 °C. The interaction between PHB-HHX and PLA could not be elucidated from the temperature data. Mechanical tests showed that the addition of PHB-HHx improves ductility of PLA/PHB-HHx composite. Morphological analysis revealed that small proportions of PHB-HHx exhibited less tendency to aggregate, which resulted in greater plastic deformation and improved toughness. From this study, PLA blended with small portions of PHB-HHx may further expand the use of bio-friendly resources in a variety of applications such as flexible films, food packaging and something like that. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Concerns about the potentially damaging effects of nondegradable plastics on the environment and the increasing cost of petrochemical feedstocks have raised new challenges in polymer research. Biodegradable polymers such as polylactide (PLA) or polyhydroxyalkanoates (PHA) are traditional alternatives to petroleum-based thermoplastics and represent an interesting and growing market. Poly(3-hydroxybutyrate) (PHB), discovered by Lemoigne in 1925, is the most representative member of the PHA family. However, it is rigid and stiff due to high crystallinity and relatively large spherulites [1,2]. Furthermore, PHB is thermally unstable during conventional melt processing and only suitable for processing in a narrow temperature window [3]. To overcome these drawbacks, several PHB copolymers have been biosynthesized by a variety of microorganisms. Recently, poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHB-HHx) was synthesized from oils and fats using Aeromonas sp. OL338 and Aeromonas sp. FA440 [4]. This new co-polyester provides an opportunity to circumvent the shortcomings of conventional PHAs such as PHB and PHBV. Randomly distributed second monomer units (3HHx) are excluded from the 3HB crystalline lattice, forming a short chain branch. ⁎ Corresponding author. Tel.: +82 31 450 4452; fax: +82 31 450 4444. E-mail address: [email protected] (J.S. Lim). 0928-4931/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2013.01.030
This branch acts as a molecular defect, which disrupts the excessive regularity of the polymer chain and consequently lowers the melt temperature (Tm) and crystallinity. This unique characteristic may induce substantially more flexibility and ductility, which would make PHB-HHx more suitable for making films and other soft articles [5,6]. Polylactide (PLA) is synthesized from lactic acid, which is obtained from renewable resources such as starch from corn or maize. PLA is now produced on a large scale and used in various applications including packaging, agriculture, and textiles. However, its high brittleness, slow crystallization rate, and low heat distortion temperature restrict its widespread use [7,8]. To overcome these limitations, many researchers have examined ways of modifying PLA, such as copolymerization or composite formation [9–12]. Creating composites of PLA with other polymers may be the most credible methodology, since this process allows the incorporation of new materials with enhanced properties as well as is less expensive than chemical modifications or syntheses of tailor-made macromolecules. In a composite system, the selection of individual components and their relative proportions must be carefully considered, as these factors can have a strong influence on the physical and thermal properties of the final materials [13,14]. In addition, the relationship between morphology and composition must be clearly investigated, since morphology depends on the miscibility of the components and can
2132
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
affect the overall properties of the composite [15,16]. Compared to PLA and other PHA family composites [17,18], very little has been reported on PLA/PHB-HHx composites. Moreover, no systematic research has been published concerning the impact of composition ratio on the physical, thermal, and morphological properties of these composites. In this work, PLA/PHB-HHx composites were prepared via direct melt compounding. The morphological, thermal, and physical properties of the composites were evaluated as a function of composition ratio. Commercial polymers without any compatibilizer were used to confirm innate polymer properties, and the neat polymers were examined first to determine processing conditions. We hypothesized that a composite of PLA, which is naturally hard and brittle, with the more ductile PLA would effectively balance the shortcomings of the individual polymers. 2. Experimental section 2.1. Materials and sample preparation A commercial linear and semicrystalline PLA (Ingeo™, 2003D) was provided in pellet form by NatureWorks LLC (Blair, NE, USA). According to the supplier, the D-isomer content of the PLA was 4.3%, with a melt flow rate of 5–7 g/10 min and a density of 1.24 g/cm 3. Additionally, this semi-crystalline PLA exhibited a weight-average molecular weight (Mw) of 210,000 g/mol and polydispersity of 1.74 as determined by gel permeation chromatography. The bacterial polyester, poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHB-HHx, Aonilex®), was supplied by the Kaneka Corporation (Osaka, Japan). It contained a 3-hydroxy hexanoate content of about 11 mol% and a weight average molecular weight (Mw) of 1,388,000 with a polydispersity of about 2.0. Before manufacturing the composite samples, the two biodegradable polymers were dried at 60 °C overnight. Fig. 1 shows the chemical structures of PLA and PHB-HHx. PLA/PHB-HHx composites were prepared by direct melt compounding using a twin roll counterrotating mixer (Thermo Haake, Paramus, NJ, USA). The barrel temperature was 180 °C and the screw speed was 100 rpm. The total mixing time was 3 min with the screw speed of 100 rpm at a barrel temperature of 180 °C. The mixtures produced were then compression molded in a hydraulic compression press for 5 min at 180 °C with pressure of 6000 psi. The weight ratios of PLA/PHB-HHx were 10/0, 9/1, 8/2, 6/4, and 0/10, respectively. 2.2. Measurements Differential scanning calorimetry (DSC) data from samples were obtained in a DSC7; Perkin-Elmer, Waltham, MA, USA. Samples of 5 mg taken from the central sections of test specimens were sealed in aluminum pans and subjected to heating from − 40 °C to 200 °C at 5 °C/min followed by holding at 200 °C for 3 min before rapidly cooling to − 40 °C. Samples were reheated to 200 °C with the same
Fig. 1. Chemical structures of PLA and PHB-HHx.
heating rate. The dynamic mechanical properties were measured on a dynamic mechanical analyzer (DMA 2980; TA Instruments, New Castle, DE, USA) set in tension mode at a frequency of 1 Hz in a nitrogen atmosphere. Samples (15 × 5 × 0.5 mm) were cooled to −40 °C and then heated gradually to 200 °C at a rate of 2 °C min−1. For infrared (IR) analyses, PLA/PHB-HHx composites were crystallized using the following procedure. First, a Mettler-Toledo FP82HT hot stage (Mettler-Toledo, Greifensee, Switzerland) was equilibrated at the desired crystallization temperature. Each sample was then melted on the hot stage at 200 °C for 5 min followed by quickly transferring to a second hot stage. Crystallization occurred over 30 min. Fourier transform (FT)-IR absorption spectra were obtained on the crystallized samples at room temperature using an IFS 88-IR spectrometer (Bruker AXS GmbH, Karlsruhe, Germany). Spectra were acquired from 4000 to 400 cm−1 at a resolution of 2 cm−1, and 16 scans were averaged for each sample. Mechanical tests were conducted at room temperature with a tensile testing machine (Instron 4465; Instron, Norwood, MA, USA) based on ASTM D638. Crosshead speed was set to 10 mm/min. The samples were prepared on a dumbbell-shaped film with the following dimensions: 3.5 mm, 14 mm and 0.3 mm in width, gage length and thickness, respectively. Each value from 10 successful measurements was averaged and expressed with a standard deviation (SD). The morphologies of PLA/PHB-HHx were investigated by scanning electron microscopy, SEM, (JSM-6700F; JEOL, Tokyo, Japan). Crosssections of gold-coated specimens were examined after fracturing in liquid nitrogen. Surface morphology was examined in air using atomic force microscopy (AFM; 2010 Discoverer; Topometrix, Santa Clara, CA, USA) in contact mode with silicon nitride tips. Images were recorded at scan speeds of 10 mm/s using the minimal applied force.
3. Results and discussion Fig. 2 shows second heating differential scanning calorimetry (DSC) thermograms of neat PLA, PHB-HHx and their composites, respectively. Neat PLA yielded a broad crystallization peak at 113 °C and a melting peak at about 150 °C. For neat PHB-HHx, a sharp exothermal peak and double melting peak were observed at about 50 °C and 100–130 °C, respectively, indicating that PHB-HHx crystallizes more rapidly than PLA at a lower temperature. The glass transition temperatures, Tg, of the two biopolymers were −2 °C and 60 °C, indicating the segmental mobility of PHB-HHx increases at a low temperature, resulting in a faster crystallization rate compared to that of PLA. In addition to, at 40% PHB-HHx, the Tg of PLA cannot be discerned due to the crystallization peak of PHB-HHx. However, at other composite proportions, the Tg of each biopolymer can be observed at 60 °C and −2 °C. No additional evidence of miscibility of polymers, such as a remarkable melting point depression [19,20]. For these reasons, we suggest that PLA and PHB-HHx are immiscible. Detailed investigation of miscibility between PLA and PHB-HHx was performed by dynamic mechanical analysis (DMA). With increasing proportions of PHB-HHx, two crystallization and melting peaks, corresponding to PLA and PHB-HHx, became apparent. This result suggests that this system can be classified as a semicrystalline/ semicrystalline composite. Moreover, 10 wt.% and 20 wt.% additions of PHB-HHx resulted in a shift of the PLA crystallization peak to higher temperature, while the heat of fusion of PLA was decreased. This may presumably due to restriction of the chain mobility of PLA by PHB-HHx, resulting in slower crystallization kinetics in the composites. The long chain length of PHB-HHx may affect entanglement of PLA, thereby reducing the crystallization rate and decreasing regularity of the PLA crystal. In the case of 40% PHB-HHx sample, the crystallization peak of PHB-HHx shifted to a higher temperature compared to the corresponding peak for neat PHB-HHx, which suggests that the PLA component may also restrict the molecular mobility of PHB-HHx in the composite materials. We will plan to investigate the crystallization kinetic of PLA/PHB-HHx
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
2133
Fig. 2. Second heating DSC thermograms of neat PLA, PHB-HHx and their composites.
composites, in detail, by analyzing the isothermal and non-isothermal crystallization experiments. Fig. 3(a) shows the storage modulus of PLA and PLA/PHB-HHx composites as a function of temperature. The high modulus of PLA at room temperature indicates that PLA is brittle, since the storage modulus of a semicrystalline polymer is indicative of material stiffness under shear deformation [21]. This peak drops abruptly near 60 °C and rises around 120 °C due to a glass transition and crystallization of PLA, respectively. Below the Tg region, the storage modulus decreases as the weight percent of PHB-HHx increases in composite, resulting in a reduction of stiffness In the glassy state, the mobility of the molecular chain is restricted and stiffness is related to stored elastic energy [22]. Therefore, these results imply that PHB-HHx may restrict the chain mobility of PLA, resulting in low stored elastic energy and a glassy modulus. From DSC analysis, it can be pointed out that PHB-HHx can hinder the
crystallization of PLA, resulting in lowering the storage modulus of PLA. PHB-HHx, which is composed of a 3-hydroxyhexanonate comonomer with a 3-hydroxybutyrate backbone, is naturally more flexible and ductile than PLA [5,6]. From these results, it is evident that innate properties of PHB-HHx dominate mechanical properties of composites, showing a decreased storage modulus as the weight percent of PHB-HHx is increased. Fig. 3(b) shows loss modulus curves of PLA and PLA/PHB-HHx composites. Generally, the glass transition, which is reflected in the loss modulus peaks, can be detected at about 60 °C in neat PLA. With increasing the weight percent of PHB-HHx, a broad peak related to the Tg of PHB-HHx gradually emerges near 20 °C; this is presumably due to poor miscibility of PLA and PHB-HHx as well as phase separation. In addition, the Tg of PLA did not shift to a lower temperature, regardless of the proportion of PHB-HHx. This indicates that PHB-HHx does not act
2134
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
(a) Storage Modulus 3000
Storage Modulus (MPa)
2500
PLA PLA/PHB-HHx=9:1 PLA/PHB-HHx=8:2
2000
PLA/PHB-HHx=6:4
1500
1000
500
0 -50
0
50
100
150
Temperature (oC) Fig. 4. The carbonyl-stretching region of the IR absorption spectra of PLA/PHB-HHx composites crystallized at 30 °C after melting.
(b) Loss Modulus 600 PLA
Loss Modulus (MPa)
PLA/PHB-HHx=9:1
500
PLA/PHB-HHx=8:2 PLA/PHB-HHx=6:4
400 300 200 100 0 -50
0
50
100
150
Temperature (oC) Fig. 3. DMA result of PLA/PHB-HHx composites.
as a plasticizer in the PLA/PHB-HHx composites. Generally, the addition of a plasticizer will decrease the Tg due to the diffusion within the molecular chains, which enhances flexibility and ductility. Moreover, the extent of this effect enlarges with increasing proportions of plasticizer [23]. Therefore, PHB-HHx does not act as a plasticizer. Instead, its innate ductility and its inhibition of crystallization affect the viscoelastic properties of the composites. Neat PLA exhibited a very sharp and intense tanδ peak (data not shown) due to no restriction in chain motion exists. The height of the tanδ peak decreased with increasing proportions of PHB-HHx, indicating that PHB-HHx may inhibit the chain mobility of PLA, which is consistent with the DSC results. Fig. 4 shows the carbonyl-stretching region of IR absorption spectra of PLA/PHB-HHx composites crystallized at 30 °C after melting. In this region, it is obvious that both PLA and PHB-HHx are in a crystalline state. The spectrum of neat PLA contains a singlet peak near 1750 cm−1, corresponding to the stretching vibration of a carbonyl group [24]. The spectral features of neat PHB-HHx can be divided into two major peaks at 1740 cm−1 and 1723 cm−1, corresponding respectively to amorphous and crystalline carbonyl groups [20]. As the proportion of PHB-HHx was increased, the crystalline peaks, which are related to PHB-HHx, increased gradually. However, no change in the main carbonyl peak, which corresponds to PLA, was observed. This result indicates that PHB-HHx may be more easily crystallized as its proportions in the composite increase, forming phase-separated domains in the PLA matrix.
The existence of hydrogen bonding between PLA and PHB-HHx was investigated by analyzing the C\O\C absorption peak in the IR spectra of the composites. Fig. 5 shows the C\O\C peaks of PLA/PHB-HHx composites crystallized at 30 °C after melting. The spectrum of neat PLA contained the following absorption peaks [24]: 1384 and 1362 cm−1 assigned to \CH\ deformation including symmetric and asymmetric bends, whereas 1267, 1182, 1129, 1090, and 1045 cm−1 attributable to \C\O\C\ stretches. The spectrum of neat PHB-HHx contains characteristic bands at 1288, 1275, 1261, and 1227 cm − 1, corresponding to the C\O\C stretching modes of the crystalline components. These four peaks may be due to helical structures formed by the PHB-HHx backbone. The bands near 1180 cm−1 probably consist of two bands, ascribed to crystalline and amorphous states, respectively [5,6]. With the addition of PHB-HHx, no unique changes were observed except the appearance of peaks related to crystallized PHB-HHx, near 1300–1250 cm−1. Therefore, no evidence of hydrogen bonding between PLA and PHB-HHx was observed. Fig. 6 shows the carbonyl-stretching region of IR spectra of the PLA/PHB-HHx composites crystallized at 130 °C after melting. The composite systems are in a crystalline/amorphous state since the temperature is near the Tm of PHB-HHx. Neat PLA exhibits the same
Fig. 5. C\O\C absorption peaks in the IR spectra of PLA/PHB-HHx composites crystallized at 30 °C after melting.
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
2135
Fig. 8. The stress–strain curve of PLA/PHB-HHx composites. Fig. 6. The carbonyl-stretching region of the IR absorption spectra of PLA/PHB-HHx composites crystallized at 130 °C after melting.
carbonyl peak regardless of temperature, while the carbonyl peak of neat PHB-HHx is observed at 1740 cm −1 due to the broad amorphous band of molten PHB-HHx. As the PHB-HHx content increases, the carbonyl peak shifts to lower energy state. The vibrational frequency of a functional group is usually determined by the force constant and mass of the bonded atoms. Any interaction with the carbonyl group may result in a decrease in bond energy, which may have lowered the force constant followed by shifting the peak to lower frequency [19,25]. Recently, several research groups have reported the appearance of a C\H\O hydrogen bond between the CH3 group of one helical structure and C_O groups in the neighboring helical structure within PLA and PHA [26,27]. Zhang et al. showed that a PLLA/PDLA stereocomplex can form intermolecular hydrogen bonds between CH3/O_C groups as evidenced by low frequency shifts of the C_O stretching band [26]. Therefore, the above results suggest that a hydrogen bond may also exist between crystalline PLA and amorphous PHB-HHx, although the formation of a hydrogen bond could not be fully ascertained based only on the carbonyl peak. A hydrogen bonding between PLA and PHB-HHx could also result in some change near the C\H peak, such as shifting or creation of shoulder peak. Fig. 7 shows the C\H stretch region of PLA/PHB-HHx composites crystallized at 130 °C after melting. No
changes were observed near the C\H stretch peak, regardless of PHB-HHx content. Therefore, we suggest that little significant hydrogen bonding presents between crystalline PLA and amorphous PHB-HHx; also, the hydrogen bonding effects are much weaker than those observed in the PLLA/PDLA stereocomplex. For the next time, our research groups have a plan to investigate IR analysis by using 2-D or 3-D IR instrument. The stress–strain behavior of the PLA/PHB-HHx composites is shown in Fig. 8. Neat PLA fails as soon as its yield stress exceeded. Neat PLA has a strong strain softening that is not stabilized by strain hardening; in tensile tests, strain softening may stimulate strain localization, which causes the buildup of local, multiaxial stresses. If the local strain is not delocalized, it induces void nucleation and cracks in the matrix, which lead to a brittle failure behavior [28,29]. In the contrast, composites containing PHB-HHx greater than 20 wt.% show stable neck growth and strain hardening, indicating that PHB-HHx can be used to improve the ductility of PLA. The mechanical properties of the PLA/PHB-HHx composites are summarized in Table 1. Effect of PHB-HHx was much more pronounced for mechanical properties of composites; yield stress and Young's modulus were decreased to 54 MPa and 1400–1450 MPa, respectively, by adding 10 wt.% of PHB-HHx. Toughness and elongation at break were 4.0 MPa and 7.6%, respectively, demonstrating that only a small amount of PHB-HHx is effective in reducing the stiffness of PLA. At 20 wt.% PHB-HHx, the ductility of the composite was significantly enhanced; however, 40 wt.% of PHB-HHx/PLA composite had a lower ductility. Several criteria must be met to improve the toughness of PLA or any other polymers. Polymers those that contribute to enhance toughness must be uniformly distributed throughout the PLA matrix with small size of domains; large size of domains are less effective on absorbing and dissipating energy under mechanical stress [30,31]. Domain size and phase separation were investigated by AFM and SEM (shown in Figs. 9, 10). The composites exhibited larger domains and rougher surfaces than those of neat PLA and PHB-HHx. Surface roughness, including Ra (average roughness) and Rq (RMS roughness) is given in Table 2. Among the tested samples, neat PLA exhibited a higher roughness than neat PHB-HHx, with values of 111.9 nm and Table 1 Mechanical properties of PLA/PHB-HHx composites. Tensile Young's Toughness strength (MPa) modulus (MPa) (MPa)
Fig. 7. The C\H stretch region of the IR spectra of PLA/PHB-HHx composites crystallized at 130 °C after melting.
PLA 62.2 (±3.5) PLA/PHB-HHx=9/1 54.1 (±2.7) PLA/PHB-HHx=8/2 45.3 (±3.3) PLA/PHB-HHx=6/4 40.1 (±2.4) PHB-HHx 21.6 (±1.8)
1603.0 (±7.0) 1416.0 (±6.0) 1265.0 (±8.0) 1093.0 (±7.0) 309.0 (±4.0)
Elongation at break (%)
3.2 (±0.5) 3.6 (±0.3) 4.0 (±0.3) 7.6 (±0.6) 68.7 (±2.7) 113.1 (±2.5) 20.4 (±1.4) 37.6 (±1.3) 160.6 (±3.4) 524.8 (±3.7)
2136
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137
(a) PLA
(b) PLA/PHB-HHx=8/2
(c) PLA/PHB-HHx/6/4
(d) PHB-HHx
Fig. 9. AFM images of the PLA/PHB-HHx composites.
(a) PLA
(b) PHB-HHx
(c) PLA/PHB-HHx=8/2
Fig. 10. SEM micrographs of the PLA/PHB-HHx composites.
J.S. Lim et al. / Materials Science and Engineering C 33 (2013) 2131–2137 Table 2 Surface roughness of PLA/PHB-HHx composites. Parameter Sample PLA Ra(nm) Rq(nm)
PLA/PHB-HHx=8/2 PLA/PHB-HHx=6/4 PHB-HHx
111.9 (±0.5) 98.2 (±0.2) 132.9 (±0.3) 127.2 (±0.6)
120.7 (±0.5) 144.3 (±0.5)
49.6 (±0.6) 62.1 (±0.5)
49.6 nm, respectively. This result is supported by the enhanced ductility and flexibility of the composite materials as described above. Although neat PHB-HHx has a lower surface roughness than PLA, the roughness of PLA/PHB-HHx composites containing high weight percent of PHB-HHx is increased. PLA/PHB-HHx with the ratio of 6:4 had the greatest surface roughness (Ra = 120.7 nm). This implies that phase separation between PLA and PHB-HHx becomes more pronounced with increasing proportions of PHB-HHx. Noda et al. reported that a very ductile polymer alloy can be produced when a small amount of PHB-HHx is mixed with hard and brittle PLA [32]. Their findings showed that fine dispersions of small PHA particles are created in such composites as long as the proportion of PHA does not exceed about 20 wt.%. As shown in Fig. 10, SEM micrographs showed that neat PHB-HHx exhibited a much smoother surface than neat PLA which is consistent with the result is consistent with the AFM images. In the 8/2 PLA/ PHB-HHx composites, small domains of PHB-HHx were homogeneously dispersed in the PLA matrix. The interface between PLA and PHB-HHx domains is vague, confirming that phase separation between the two components is not absolute. We also analyzed the SEM of PLA/PHB-HHx (6:4). Compared with PLA/PHB-HHx (8:2), it was observed that PHB-HHx domain size was large and phase separation between PLA and PHB-HHx (data not shown). Moreover, this result may be a good agreement with AFM data. These findings indicate that small amounts of PHB-HHx are less likely to aggregate, resulting in a relatively large overall surface area in contact with the PLA matrix. Consequently, the tensile energy was efficiently dissipated by the elastic PHB-HHx phase, leading to greater plastic deformation and improved toughness of the composite over that of neat PLA. 4. Conclusions Binary bioplastic composites of PLA and PHB-HHx were prepared by melt mixing. Mechanical and thermal properties and surface morphology were investigated as a function of the composition ratio. DSC analyses indicated that PLA and PHB-HHx formed immiscible composites over the observed range of compositions. The crystallization of PLA was suppressed with increasing proportions of PHB-HHx. DMA results confirmed that PHB-HHX could be used to improve the stiffness of PLA due to their innate ductility and inhibiting effect on crystallization. At 30 °C, the crystallinity of PHB-HHx domains in the composites increased with greater proportions of PHB-HHx. At 130 °C, an amorphous
2137
phase of molten PHB-HHx diffused within the crystalline phase of PLA. However, there is no evidence of a direct chemical interaction between PLA and PHB-HHx within the composites. Mechanical tests showed that the material changed from brittle (PLA) to ductile failure with the addition of PHB-HHx, although no optimal proportion of PHB-HHx was determined that maximized toughness and elongation at break without severe loss in tensile strength and modulus. Morphological analyses revealed that small amounts of PHB-HHx were less likely to aggregate, resulting in greater plastic deformation and enhanced toughness in the final composite. Thus, for industrial applications, the toughness of PLA can be improved with the addition of small amounts of PHB-HHx.
References [1] A.J. Anderson, E.A. Dawes, Microbiol. Rev. 54 (1990) 450. [2] M. Kunioka, Y. Doi, Macromolecules 23 (1990) 1933. [3] S.F. Williams, D.P. Martin, D.M. Horowitz, O.P. Peoples, Int. J. Biol. Macromol. 25 (1999) 111. [4] Y. Doi, S. Kitamura, H. Abe, Macromolecules 28 (1995) 4822. [5] I. Noda, P.R. Green, M.M. Satkowski, L.A. Schechtman, Biomacromolecules 6 (2005) 580. [6] I. Noda, E.B. Bond, P.R. Green, D.H. Melik, K. Narasimhan, L.A. Schechtman, Polym. Biocatal. Biomater. 900 (2005) 280. [7] K.A.M. Thakur, R.T. Kean, J.M. Zupfer, N.U. Buehler, M.A. Doscotch, E.J. Munson, Macromolecules 29 (1996) 8844. [8] H. Urayama, T. Kanamori, K. Fukushima, Y. Kimura, Polymer 44 (2003) 5635. [9] Y.L. Zhao, Q. Cai, X.T. Shuai, J.Z. Bei, C.F. Chen, F. Xi, Polymer 43 (2002) 5819. [10] E.S. Kim, B.C. Kim, S.H. Kim, J. Polym. Sci., Part B: Polym. Phys. 42 (2004) 939. [11] H. Liu, F. Chen, B. Liu, G. Estep, J. Zhang, Macromolecules 43 (2010) 6058. [12] L. Jiang, M.P. Wolcott, J. Zhang, Biomacromolecules 7 (2006) 199. [13] G.I. Williams, R.P. Wool, Appl. Compos. Mater. 7 (2000) 421. [14] K. Oksman, J.F. Selin, in: F.T. Wallenberger, N.E. Weston (Eds.), Natural Fibre: Plastics and Composites, Kluwer Academic Publishers, Dordrecht Boston London, 2004, p. 149. [15] P. Martin, C. Maquet, R. Legras, C. Bailly, L. Leemans, M. van Gurp, Polymer 45 (2004) 5111. [16] P. Martin, C. Maquet, R. Legras, C. Bailly, L. Leemans, M. van Gurp, Polymer 45 (2004) 3277. [17] V. Christian, G.H. Günter, W.S. Heinz, Vib. Spectrosc. 49 (2009) 284. [18] T. Gerard, T. Budtova, Eur. Polym. J. 48 (2012) 1110. [19] J.S. Lim, I. Noda, S.S. Im, J. Polym. Sci., Part B: Polym. Phys. 44 (2006) 2852. [20] J.S. Lim, I. Noda, S.S. Im, Eur. Polym. J. 44 (2008) 1428. [21] J.S. Lim, J.H. Kim, Mater. Sci. Eng. A 527 (2010) 4641. [22] H. Ishida, D.J. Allen, Polymer 37 (1996) 4487. [23] Z. Kulinski, E. Piorkowska, Polymer 46 (2005) 10290. [24] Y. Gao, L. Kong, L. Zhang, Y. Gong, G. Chen, N. Zhao, X. Zhang, Eur. Polym. J. 42 (2006) 764. [25] J.S. Lim, I. Noda, S.S. Im, Polymer 48 (2007) 2745. [26] J. Zhanga, H. Sato, H. Tsujib, I. Noda, Y. Ozaki, J. Mol. Struct. 735 (2005) 249. [27] H. Sato, R. Murakami, A. Padermshoke, F. Hirose, K. Senda, I. Noda, Macromolecules 37 (2004) 7203. [28] P. Ma, D.G. Hristova-Bogaerds, J.G.P. Goossens, A.B. Spoelstra, Y. Zhang, P.J. Lemstra, Eur. Polym. J. 48 (2012) 146. [29] B. Meng, J. Deng, Q. Liu, Z. Wu, W. Yang, Eur. Polym. J. 48 (2012) 127. [30] K. Cho, J.H. Yang, C.E. Park, Polymer 39 (1998) 3073. [31] P. Nawadon, C. Sirijutaratana, E. Nukul, Mater. Sci. Eng. A 532 (2012) 64. [32] I. Noda, M.M. Satkowski, A.E. Dowrey, C. Marcott, Macromol. Biosci. 4 (2004) 269.