Properties of Linear Poly(Lactic Acid)/Polyethylene Glycol Blends

K. Sungsanit, N. Kao, S.N. Bhattacharya School of Civil, Environmental and Chemical Engineering, Rheology and Materials Processing Centre, RMIT University, Melbourne, Victoria 3001, Australia

Poly(lactic acid) (PLA) has great potentials to be processed into lms for packaging applications. However, lm production is difcult to carry out due to the brittleness and low melt strength of PLA. In this investigation, linear PLA (L-PLA) was plasticized with poly(ethylene glycol) (PEG) having MW of 1000 g mol21 in various PEG concentrations (0, 5, 10, 15, and 20 wt%). In relation to plasticizer content, the impact resistance and crystallinity of L-PLA was increased, whereas a decrease in glass transition temperature and lower stiffness was observed. Nevertheless, the phase separation has been found in samples which contained PEG greater than 10 wt%. The dynamic and shear rheological studies showed that the plasticized PLA possessed lower viscosity and more pronounced elastic properties than that of pure PLA. Both storage and loss moduli decreased with PEG loading at all frequencies while storage modulus exhibited weak frequency dependence with increasing PEG content. POLYM. ENG. SCI.,
52:108116, 2012. 2011 Society of Plastics Engineers

INTRODUCTION Poly(lactic acid) (PLA) is produced from renewable resources that has become a useful material, especially in packaging applications due to its good clarity, high strength, and moderate barrier properties. PLA resins have mostly been used for biomedical applications such as drug delivery systems and packaging applications such as lms and food containers [1]. Some of the environmental benets of PLA and opportunities for the future are presented by Henton et al. [2]. These include PLA requiring less energy to produce as well as reduced green house gas production. Generally, commercial grades PLA are copolymers of poly(L-lactic acid)(PLLA) and poly(D,L-lactic acid) (PDLLA), which are produced from L-lactides and D,L-lactides, respectively. PLLA is a semicrystalline polymer whereas PDLLA is an amorphous polymer.
Correspondence to: K. Sungsanit; e-mail: Kullawadee_s@hotmail.com K. Sungsanit is currently at Rajamangala University of Technology Thanyaburi, Pathumthani, Thailand. DOI 10.1002/pen.22052 Published online in Wiley Online Library (wileyonlinelibrary.com). C 2011 Society of Plastics Engineers V

PLAs use in lm packaging applications is highly desirable due to its environmentally friendly nature. However, this requires lm extrusion of PLA to be performed, which is difcult to process due to the brittleness and low melt strength of PLA. The rheological and mechanical properties of PLA may be enhanced by blending it with a plasticizer or with a second polymer. For instance, it could be blended with other polymers such as linear low density polyethylene [3], poly(vinyl acetate) [4] and polyethylene glycol (PEG) [5, 6]. Possible plasticizers for PLA include oligomeric lactic acid, lactide and low molecular weight esters such as citrates [7]. These blends have been shown to improve the exibility of PLA. However, some of the above-mentioned polymers are not biodegradable; hence, they could not be considered as a possible blend with PLA in this investigation. In consumer goods packaging applications, only nontoxic substances approved for food contact can be considered as plasticizing agents. The selection of a plasticizer to be used in a specic PLA composition requires the consideration of many criteria: compatibility, low volatility, resistance to migration, extraction during service life, lack of toxicity, etc. [8]. Low molecular weight PEG is the most suitable material to be classed as an impact modier for PLA due to its miscibility, biodegradability, and food contactable applications [6, 911]. Jacobsen et al. [9] investigated the plasticized PLA, with 2.510 wt% of PEG (MW 1.5 3 103) by melt blending. They found that the addition of PEG to PLA led to a decrease of both tensile strength and elasticity modulus but an increase of percentage elongation at break. Adding 10 wt% PEG resulted in an enhanced impact resistance of more than ve times that of a pure PLA. Kulinski et al. [11] investigated the blending of semicrystalline and amorphous PLA with 5 and 10 wt% of PEG. They reported that at 10 wt%, an amorphous plasticized PLA could be deformed to about 550% while a semicrystalline PLA exhibited nonuniform plasticization of the amorphous phase and showed less ability to the plastic deformation. Sheth et al. [5] have found that PLA/PEG blends varied from completely miscible to partially miscible, depending on the PEG concentration. Below 50 wt% PEG content, the blends achieved the higher elongations and lower modulus values. Above 50 wt% PEG content, the blend morphology was driven by
POLYMER ENGINEERING AND SCIENCE -2012

TABLE 1. Properties of PLA and plasticizer used in this study. Mw (g/mol) 155,000a Tg Tm (K)c (K)c 442 334 Chemical formula

Materials Linear poly (lactic acid) (L-PLA)

Poly(ethylene glycol) (PEG)

1,000b

313

203

a Determined by Gel Permeation Chromatography (GPC) versus polystyrene standards. b As stated by the manufacturer. c Determined by Differential Scanning Calorimetry (DSC).

the increasing crystallinity of PEG, resulting in an increase in modulus and a corresponding decrease in elongation at break. Recent studies [6, 10] of PLA plasticized with PEG in various contents have shown a limit of miscibility of polymer blends. The PLA blended to PEGs became very brittle as a function of plasticizer content and molecular weight. The plasticizing efciency increased with decreasing molecular weight of PEG. In contrast, at the same molecular weight of PEG, material became brittle at higher content because of a lack of cohesion between the separate phases. This article investigates the effect of plasticization on the properties of L-PLA/PEG blends at various PEG contents. Differential scanning calorimetry (DSC) was carried out in a wide range of temperatures in order to evaluate crystallization and melting behavior of the blends. Furthermore, impact and tensile properties were investigated to evaluate the outcome of PLA plasticization with PEG for lm applications. Surprisingly, there is very little literature examining the plasticizing effects of PEG on the rheological properties in terms of steady shear and dynamic measurements of PLA/PEG blends which play important roles in polymer processing. Thus, the melt rheology of pure L-PLA and plasticized L-PLA was conducted in this study and the results are presented below. EXPERIMENTAL Materials The poly(lactic acid) (4032D-grade) from NatureWorks produced by Cargill Dow LLC, used in this study, comprised of 2% D-LA content. Polyethylene glycol (molecular weight 1000 g/mol), a food-contact approved grade from Sigma-Aldrich, was chosen as a plasticizer for the L-PLA. The materials used and their properties are listed in Table 1. Blending Preparation L-PLA pellets were dried in an oven under vacuum at 508C overnight (1215 h) prior to blending. Drying was
DOI 10.1002/pen

necessary to minimize the hydrolytic degradation of the polymers during melt processing in the extruder. L-PLA and PEG were melt-blended using a Brabender twinscrew extruder in the ratio of 100/0, 95/5, 90/10, 85/15, and 80/20 where the rst and second number represent LPLA and PEG by weight percentage, respectively. For better comparison, the pure L-PLA sample was also processed in the same twin screw extruder and processing conditions to ensure identical thermal history with those of L-PLA/PEG blends. Dry-mixing of each polymer was rst carried out in a zip-lock bag before blending. The twin screw extruder had a screw diameter of 17.8 mm and an L/D ratio of 40. The extruder had three controlled temperature zones which were set from 1808C (next to the feeding segment), 1908C (compression zone) and 2008C (metering zone). The screw speed was maintained at 30 rpm for all runs. Subsequently, plasticized L-PLA pellets were dried again under vacuum at 508C overnight prior to sample preparation by injection and compression molding. Differential Scanning Calorimetry Differential scanning calorimetry (DSC) was carried out with a DSC TA Instrument 2920. The samples were preliminarily heated to 473 K to discard any anterior thermal history and held at that temperature for 5 min. It was then cooled to 253 K at a rate of 10 K min21 and kept at 253 K for 5 min before a second heating scan from 253 to 473 K at 10 K min21 scan rate was carried out. During the second heating scan the glass transition, cold crystallization and melting temperature of the material could be determined; whilst the crystallization temperature was determined from the cooling scan. The degree of crystallinity of all samples were calculated by Crystallinity % DHm DHcc =UPLA 100: 93:6 (1)

Most commonly, an enthalpy of fusion of 93.6 J g21 is used for a 100% crystalline PLLA or PDLA homopolymers [12], where DHm is the measured heat of fusion and DHcc is the heat of cold crystallization. This value is used throughout the PLA literature. FPLA is the PLA content in the component. Mechanical Characterization Tensile measurements were carried out to determine the tensile strength, tensile modulus and % strain at break using the Instron 4467 Universal testing machine at room temperature. The testing was carried out at a rate of 5 mm min21 according to ASTM D638 on standard Type I dog-bone shaped samples with sample thickness of 4.1 mm. The impact resistance was determined for each material with 10 samples, which were tested under the temperPOLYMER ENGINEERING AND SCIENCE -2012 109

out to determine the dynamic moduli and complex viscosity over a frequency range of 0.1100 rad s21 at 10% strain. All tests were started when the temperature had stabilized after loading the sample into the rheometer. Steady shear measurements were performed over a shear rate range of 0.0110 s21.

Fractured Surface Characterization The fractured surfaces of all materials were observed by Environmental Scanning Electron Microscopy (ESEM), using a Philips XL-20 SEM at an accelerating voltage of 30 kV. All of the samples were fractured specimens after impact tests. The fractured surface was then coated with a thin layer of gold prior to observation. RESULTS AND DISCUSSION Thermal Properties The results obtained from differential scanning calorimetry of the pure L-PLA (unprocessed and 0 wt% PEG) and plasticized L-PLA are shown in Fig. 1 and important numerical values are summarized in Table 2. Table 2 summarizes the glass transition temperature (Tg) of plasticized L-PLA as well as crystallization and melting temperatures (Tc and Tm). Pure PLA showed a sharp Tg and its magnitude decreased gradually with increasing PEG concentration. A minor decrease in the melting temperature (Tm) of 3 to 4 K was also observed, indicating that the melting temperature of L-PLA was not greatly affected by the addition of PEG. According to DSC thermograms (Fig. 1a), the crystallization temperature (Tc) seemed to be increased with increasing PEG content. This was due to the presence of the plasticizer, which facilitated the crystallization process of PLA. As noted previously, the increased molecular mobility increased the rate of crystallization, which allowed LPLA to crystallize to a higher degree during cooling [1] and allowing the crystallization of L-PLA to occur at a higher temperature (i.e., with less sub-cooling). This is in accordance with the crystallization behavior of PLA as reported by Miyata et al. [13]. However, there was an irregular characteristic of Tc which showed sudden decrease from 369 K to 363 K in the presence of a small amount of plasticizer at 5 wt% PEG. It is presumed that during the non-isothermal crystallization temperature from 473 K to 253 K, L-PLA initially crystallized before the formation of PEG crystals at temperature range between 373 K and 363 K. On the other hand, PEG crystallized following the formation of L-PLA crystals between 303 K and 293 K (determined by DSC). Consequently, during the formation of L-PLA crystals some molecules of PEG probably could be trapped in the intra-spherulitic region of L-PLA and led to hindering the crystallization of LPLA. Nevertheless, addition more wt% of PEG enhances
DOI 10.1002/pen

FIG. 1. Effect of PEG concentration on crystallization and melting of L-PLA/PEG blends: (a) cooling thermograms obtained with a cooling rate of 10 K min21; (b) subsequent heating thermograms obtained with a heating rate of 10 K min21.

ature like the tensile test and following the ASTM standard specication ASTM D256, the so-called izod method, with a Davenport impact tester. The samples were notched and a pendulum hammer of 1.36 J was used. Both tensile and impact samples were injection molded and were dried in the vacuum oven at 508C over night prior to testing. Rheological Characterization The rheological properties of L-PLA with various PEG contents were measured using an Advanced Rheometric Expansion System (ARES) tted with a 25 mm parallel plate geometry. Tests were performed at 1808C under a nitrogen atmosphere to avoid any degradation. Sample disks for the rheometer were compression molded at 2008C into 25 mm diameter discs $3.2 mm thick and were again dried in the vacuum oven overnight at 508C prior to testing. Dynamic strain sweep tests were carried out to conrm the linearity of the viscoelastic region up to 100% strain at 10 rad s21 frequency. Frequency sweeps were carried
110 POLYMER ENGINEERING AND SCIENCE -2012

TABLE 2. Results from DSC for the L-PLA/PEG Blends. Cooling PEG content (wt%) Un-processed PLA 0 5 10 15 20 Tc (K) 369 363 368 371 378 DH (J g ) 12 13 21 30 30
21

Subsequent heating Tg (K) 334 335 319 311 306 306 Tcc (K) 373 365 363 DH (J g21) 12 7 1 Tm (K) 442 444 443 443 441 440 DH (J g21) 12 38 39 34 30 33 Xc (%) 13 27 36 40 39 44

All data are the average data.

degree of the crystallization rate of L-PLA as seen in Table 2. In contrast, as seen in Fig. 1b for 0 wt% PEG, the Tg was observed to be approximately 335 K. At 5 wt% PEG the Tg decreased to approximately 319 K and down to 306 K when up to 15 wt% PEG was added. However, for blend compositions having higher concentration of PEG, there was no change in the Tg of L-PLA/PEG blends. Moreover, the plasticized L-PLA showed crystallinity between 36% and 44% for all the four different polyethylene glycol contents examined. Fig. 1b shows that the cold crystallization temperature (Tcc) of L-PLA decreased in the presence of the plasticizer. For the un-processed LPLA, the thermograms did not show the cold crystallization peak since the material did not go through the thermal process. While the processed L-PLA without PEG (0 wt % of PEG) showed the cold crystallization temperature of 373 K and suddenly decreased to 263 K for the LPLA/PEG sample containing 10 wt% of PEG. For samples containing higher PEG content, cold crystallization was no longer visible. In the present blends, compared with that of pure L-PLA, the depression of Tg and the signicant decrease in Tcc indicated that polyethylene glycol was compatible and effective with L-PLA. It clearly appeared that the decreasing of Tcc and Tg of L-PLA due to enhanced chain mobility as the plasticizer level increased. Pillin et al. [6] has also reported this behavior. It was also interesting to see that the small melting and crystallization peaks of PEG were seen for L-PLA/PEG blends containing 15 and 20 wt% plasticizer concentration (in the circle of Fig. 1a and b) instead of the observation of the glass transition of blends. It is possible that there was the phase-separation of pure PEG in this blend (compared with melting temperature of PEG 1000 in Table 1). It is obvious that to facilitate the plastic deformation of semicrystalline polymer, with glassy amorphous phase, the plasticization of the latter and sufcient decrease of Tg are required. However, crystallization while increasing the average plasticizer content in the amorphous phase may also induce a rejection of plasticizer from growing spherulites and its accumulation at interspherulitic boundaries [14]. In the L-PLA/PEG blends investigated in this work, segregation of a pure PEG phase always occurred at plasticizer content higher than 10 wt%. From the DSC
DOI 10.1002/pen

analysis, it could be concluded that the PEG plasticizer has ability to increase ductility of L-PLA by increasing the mobility of L-PLA molecules. Tensile Properties The objective of plasticization is to improve the ductility while maintaining the blends strength and stiffness. Results of tensile experiments are shown in Figs. 24. When L-PLA sample was stretched without PEG, the tensile strength, tensile modulus and % tensile strain at break were 49 MPa, 4.3 GPa, and 1.3%, respectively. For plasticized L-PLA at 5 wt% and 10 wt% PEG, the tensile strength was 56 and 51, MPa, respectively. The higher tensile strength values were due to the higher crystallinity content than that of the L-PLA with 0 wt% of PEG (whose tensile strength was only 49 MPa). Tensile strength of the L-PLA/PEG blends gradually decreased to 29 and 22 MPa, when higher concentrations (15 and 20 wt %, respectively) of PEG were used (see Fig. 2). From thermal characterization, it was clear that phase separation of L-PLA/PEG blends in this investigation could be observed at PEG concentration of more than 10 wt%. Moreover, from Fig. 3, a gradual decrease in the tensile

FIG. 2. Effect of PEG concentration on tensile strength of L-PLA/PEG blends.

POLYMER ENGINEERING AND SCIENCE -2012 111

FIG. 3. Effect of PEG concentration on tensile modulus of L-PLA/PEG blends.

FIG. 5. Effect of PEG concentration on the impact strength of L-PLA/ PEG blends.

modulus from 4.3 to 1.9 GPa was observed as the concentration of PEG increased from 0 to 20 wt% in the L-PLA/ PEG blends. In the meantime, the tensile strain at break (see Fig. 4) increased from 1.5% to 15% with increasing PEG content of 0 wt% and 20 wt% PEG, respectively.

Impact Properties The impact resistance properties of L-PLA/PEG blends are shown in Fig. 5. Clearly, it could be seen that the presence of PEG as a plasticizer in L-PLA marginally increases the toughness of L-PLA. For L-PLA/PEG bland (0 wt% PEG), the average value was 13.3 kJ/m2 but the data increased dramatically to 15.4 and 14.9 kJ m22 for L-PLA/PEG samples containing 5 and 10 wt% PEG, respectively. As expected, the increase in impact strength

FIG. 4. Effect of PEG concentration on % strain at break of L-PLA/ PEG blends.

of L-PLA was observed when plasticized with PEG because the plasticizer interacted with the polymer chains distributing itself uniformly within the polymer, hence creating additional free volume. However, the decrease in impact strength was also observed when PEG concentration reached 15 and 20 wt% due to separation of PLA and PEG phases as seen from thermal characterization. This indicated that the blends were ductile at less than 10 wt% plasticizer content and were brittle at greater than 15 wt% plasticizer content. It should be noted that it is important to have information about the impact resistance behavior at other deformation rates as well. Because of some differences between tensile and impact deformation rates, it has been found in this investigation that blends with high elongation at break were characterized by relatively weak impact values. Therefore, it is important to note that for plasticized blends with high content of plasticizer (PEG) a decrease in impact strength has been reported. Jacobsen and Fritz [9] attributed this to a disturbance created by the plasticizer composition into PLA matrix. Surprisingly, the abrupt change of all mechanical properties occurred regularly in L-PLA/PEG blends whose PEG concentrations were greater than 10 wt%. For higher PEG concentrations, the material became brittle due to an occurrence of phase separation at the saturated points of PEG content. This could result only from a phase separation in the amorphous phase leading to the formation of a plasticizer rich phase and depleting the PLA of plasticizer. SEM studies of fractured surfaces of impact bar as shown in Fig. 6 illustrated some emptied voids and round white particles in L-PLA/PEG blends (15 and 20 wt% of PEG) where PEG accumulated during phase separation. The fractured surfaces represented the dispersion of PEG phases in L-PLA matrix uniformly and the size of the PEG domains was of sub-micrometer order. Additionally, the rich PEG phase and smooth plane could be seen more
DOI 10.1002/pen

112 POLYMER ENGINEERING AND SCIENCE -2012

on fractured surfaces of PLA/PEG blends (15 and 20 wt% PEG). It has been reported [15] that this kind of bril formation appeared to be related to the increase of the temperature in the crack-tip region above the glass transition temperature due to heat generation at high strain-rate. The Fig. 6b and c. shows the surface of the blend at 510 wt% PEG. There were more brils and roughness on the surface compared with sample without PEG. This proved that the PEG was equally dispersed into L-PLA matrix, which made the sample slightly more ductile. When PEG content was above 10 wt%, the small cavities and white round shapes of PEG were observed in Fig. 6d and e. The white round shapes showed broad distribution at 15 wt% PEG content. The distribution of the white round shapes increased (as shown in Fig. 6e) with increasing PEG content (20 wt%). In addition, the sample containing 20 wt% PEG content also showed a smooth fractured surface indicating brittle fractured areas with considerable voids of submicron size are clearly observed, which was probably caused by the accumulation of PEG during phase separation.

Rheological Properties To fully understand the processing properties of plasticized L-PLA blends, a detailed investigation of the rheological behavior of these L-PLA/PEG blends with varying PEG concentration was necessary. Figure 7 presents the dynamic viscosity properties on frequency at various PEG concentrations at 1808C. The unprocessed L-PLA data were obtained from dynamic frequency sweep measurement carried out on the nonextruded pellets. It exhibited a clear Newtonian Plateau at low oscillation frequency with a zero-shear rate viscosity around 1000 Pa s and seemed to be shear thinning behavior at high oscillation frequency. On the other hand, the processed L-PLA also exhibited a non-Newtonian behavior, but with a much lower zero-shear rate viscosity value (around 840 Pa s). All L-PLA/PEG blends of varying

FIG. 6. SEM micrographs (15003) of the fractured surfaces of the LPLA/PEG blends with (a) 0, (b) 5, (c) 10, (d) 15, and (e) 20 wt% PEG.

clearly at the fractured surface of 20 wt% PEG, indicating the phase separation and brittle failure of L-PLA/PEG blends. Furthermore, the phase separation in the amorphous phase was already reected in additional low temperature glass transitions of the blends and in melting peaks of PEG crystals in L-PLA/PEG blends evidenced by the DSC measurements as shown in Fig. 1b. Upon phase separation the PEG plasticizer accumulated in distinct pools. Fractured Surface Analysis by SEM The SEM micrographs revealed rather brittle fracture of L-PLA/PEG blend (0 wt% PEG) with little amount of plastic deformation with signicant lateral contraction of the test bar and many striated ridges as shown in Fig. 6 (15003 magnication). The lateral contraction and ridges are the morphological manifestation of the fact that shearyielding had occurred during the impact test, resulting in a ductile break. Figure 6 SEM micrographs of the impactfractured surfaces show more evidences of ductile fractures as more and longer brils can be observed from the surfaces with the increase in PEG content. Ductile bril formations were observed on the impact surfaces as shown in Fig. 6ac. In contrast, these brils were observed barely in sample without PEG at $5 lm length between the striated ridges and there was no longer bril
DOI 10.1002/pen

FIG. 7. Dependence of dynamic viscosity on frequency at various PEG concentrations at 1808C. POLYMER ENGINEERING AND SCIENCE -2012 113

FIG. 8. Dependence of storage modulus (G0 ) on frequency at various PEG concentration in L-PLA/PEG blends at 1808C.

PEG concentrations exhibited a more pronounced Newtonian response with an extended Newtonian plateau compared with unprocessed PLA and also showed the decreasing zero-shear viscosity values as the PEG concentration increased. This was the effect of more disentanglement of PLA chain due to increased segmental mobility of PLA chains. Additionally, it could be noticed from Fig. 7 that by blending in the twin screw extruder the polymer was exposed to excessively high shear strains resulting in a greater degradation of the PLA. Moreover, the degradation in PLA during processing in the presence of plasticizers with ester groups could also be due to potential transesterication reactions leading to a decrease of PLA molecular weights [16] which resulted in a decrease in PLA viscosity. The corresponding storage and loss moduli for these blends are shown in Figs. 8 and 9, respectively. As expected, the moduli of L-PLA decreased with increasing plasticizer loading at all frequencies. Unprocessed L-PLA and plasticized L-PLA (0-20 wt% of PEG) exhibited the rheological behavior of a typical polymer melt as characterized by a storage modulus (G0 , Fig. 8) smaller than the loss modulus (G00 , Fig. 9). Both the G0 and G00 decreased with increasing PEG concentration. However, at low-frequency G0 of all blends presented lower frequency dependency. Only at high frequencies, all samples approximately showed a common storage modulus. Such a nonterminal behavior, sometime occurs at a medium frequency region, as has been observed on many polymer blends and is accepted to be attributed to the change of the shape of the discrete phase in the polymer matrix during the oscillatory shear deformation, namely shape relaxation [17, 18]. In this investigation, at medium to low frequency region the storage modulus exhibited weak frequency dependency with increasing PEG content, such that there were gradual changes of behavior from liquid-like (G0 (x) ! x2) to solid-like with increasing PEG content. At frequency less than 1 rad s21, the frequency dependent
114 POLYMER ENGINEERING AND SCIENCE -2012

transition of the blend with PEG concentration less than 10 wt% could be observed. On the other hand, the frequency dependent transition of L-PLA/PEG blends at PEG concentration higher than 10 wt% showed a medium frequency dependent region between 1 and 10 rad s21. It could be concluded that at higher PEG concentration the G0 curves exhibited a plateau distinctly at the low frequencies as the blends seemed to be a solid like behavior. However it illustrated the discrete phase as well in the matrix if the plasticizer saturation point was reached. As seen in Fig. 8, the slope of log G0 vs. log x for the unprocessed L-PLA was close to 2, similar to the thermo-rheologically simple polymer in the terminal regime. In contrast, the slopes of the storage moduli, in the terminal region of low frequencies (0.11 rad s21), for L-PLA/ PEG blends were much smaller than 1 (in fact close to 0.5), especially for the blends containing 15 wt% and 20 wt% of PEG. Zheng et al. and Du et al., [19, 20] reported that the experimental values of the slope for G0 obtained from other phase separated or degraded polymer blends varied between 0.5 and 1. Therefore, the small values of these exponents suggested that the high concentration of PEG may have contributed to the phase separation in these blends as veried in the thermal and mechanical characterization. In addition, the use of a plasticizer reduces the intermolecular force and increases the mobility of the polymeric chains, thereby improving the exibility and the extensibility of the lasticized polymer [21]. The dependences of the dynamic loss moduli of L-PLA/ PEG blends on frequency (see Fig. 9) indicated that the blends with higher PEG content had lower G00 values than that of unprocessed L-PLA over the frequency range covered. This was due to the fact that G00 represented the viscous behavior (i.e., the amount of energy dissipated), and the addition of PEG to the L-PLA produced a material with the lowest energy dissipation. From the view of miscibility of blends, the interaction between blends decreased to a certain extent at higher PEG content. Hence decreasing in

FIG. 9. Dependence of loss modulus (G00 ) on frequency at various PEG concentration in L- PLA/PEG blends at 1808C. DOI 10.1002/pen

amount of energy dissipated of blends melting with all load resulted in the decreased loss modulus [22]. Figure 10 shows the steady shear viscosity of L-PLA/ PEG blends at various PEG content. As expected both unplasticized and plasticized L-PLA melts behaved as typical non-Newtonian uids. At all shear rates, the shear viscosities of L-PLA/PEG blends were lower than those of pure L-PLA melt and decreased considerably with increasing PEG content. The slight shear-thinning behavior could be observed in unplasticized and plasticized LPLA in whole the shear region. The calculated parameters from the measured rheological properties are summarized in Table 3. All parameters were obtained by tting the modied Cross model as follows [23]: Z Z0 ; 1 t0g m (2)

TABLE 3. Rheology characterization of L-PLA and plasticized PLA melts. 0 wt% PEG 868.8 0.12 0.11 0.968 5 wt% PEG 394.7 0.10 0.15 0.945 10 wt% PEG 307.2 0.09 0.11 0.983 15 wt% PEG 186.1 0.03 0.07 0.984 20wt% PEG 139.1 0.01 0.09 0.965

Sample g0 s0 m r2

CONCLUSIONS This article demonstrated that plasticizing L-PLA with PEG could produce a more exible material with different mechanical and rheological properties. It was found that PLA/PEG blends lowered the glass transition temperature and modied the crystallization properties. It clearly appeared that the blends obtain a miscibility of the components at 5 wt% and 10 wt% PEG content. Mechanical characteristics of these materials showed a decrease in modulus and stress at break, but an increase in % elongation at break and impact strength. Nevertheless, L-PLA blended with PEG became very brittle at higher plasticizer content due to phase separation of PEG phase as evidence from SEM micrographs. Rheological study concluded that both unplasticized and plasticized L-PLA exhibited a slightly shear thinning behavior at all frequency regions. Viscosities of blends decreased as PEG content was increased due to increasing in chain mobility in the system. Moreover, during processing in the presence of plasticizers with ester groups could also be potential trans-esterication reactions leading to a decrease of PLA molecular weights. The moduli of PLA/PEG blends decreased with increasing plasticizer loading at all frequencies. At low frequency region, the storage modulus exhibited weak frequency dependence with increasing PEG content. On the other hand, loss modulus decreased monotonically with increasing plasticizer loading at all frequencies. In this article, plasticizing PEG with linear PLA just have ability to improve the impact strength. Therefore with this study we indicate a point of attention for plasticizing branched PLA. A study probing the effect of PEG on the properties of branched PLA is already in progress. REFERENCES
1. O. Martin and L. Averous, Polymer, 42, 6209 (2001). 2. D.E. Henton, P. Gruber, J. Lunt, and J. Randall, Natural Fibers, Biopolymers and Biocomposites, CRC Press, Boca Raton, FL (2005). 3. K.S. Anderson, S.H. Lim, and M.A. Hillmyer, J. Appl. Polym. Sci., 89, 3757 (2003). 4. A.M. Gajria, V. Dave, R.A. Gross, and S.P. McCarthy, Polymer, 37, 437 (1996).
POLYMER ENGINEERING AND SCIENCE -2012 115

where Z0 represents the zero shear rate viscosity (Pa s), t0 represents the characteristic relaxation time (s), and m characterizes the slope of the line over the pseudoplastic region in the logarithmic plot. The Modied Cross model ts the data well. All the equations have correlation coefcient (r2) close to 0.95. As shown in Table 3, the incorporation of PEG led to the decreasing value of m and decreasing in s0. This referred that adding more PEG content into L-PLA could lead to decreasing in relaxation time and slight shear thinning tendency. This was attributed to the addition of PEG was easier to cover with PLA chains and the chain of PLA was disentangled under higher shear rate. Moreover, PLA blending with PEG also improved the elastomeric behavior of matrix, which would resist the ow and make the value of m tend to decrease. As shown in Table 3, g0 decreased signicantly with increasing PEG content and the saturated points of PEG content seem to be reached resulting in phase separation [24]. This effect was observed in the PEG concentration at 15 wt% and 20 wt%. According to theory [25], this reduction in viscosity can be interpreted as an enhanced mobility of polymer chain in the system.

FIG. 10. Dependence of shear viscosity on shear rate at various PEG concentration of L-PLA/PEG blends at 1808C. DOI 10.1002/pen

5. M. Sheth, R.A. Kumar, V. Dave, R.A. Gross, and S.P. McCarthy, J. Appl. Polym. Sci., 66, 1495 (1997). 6. I. Pillin, N. Montrelay, and Y. Grohens, Polymer, 47, 4676 (2006). 7. N. Ljungberg and B. Wesslen, J. Appl. Polym. Sci., 86, 1227 (2002). 8. M.C. Hansen, Prog. Org. Coat., 51, 109 (2004). 9. S. Jacobsen and H.G. Fritz, Polym. Eng. Sci., 39, 1303 (1999). 10. M. Baiardo, G. Frisoni, M. Scandola, M. Rimelen, D. Lips, K. Rufeux, and E. Wintermantal, J. Appl. Polym. Sci., 90, 1731 (2003). 11. Z. Kulinski and E. Piorkowska, Polymer, 46, 10290 (2005). 12. E.W. Fisher, H.J. Sterzel, and G. Wegner, Colloid. Polym. Sci., 251, 980 (1973). 13. T. Miyata and T. Masuko, Polymer, 39, 5515 (1998). 14. E. Piorkowska, Z. Kulinski, A. Galeski, and R. Masirek, Polymer 47, 7178 (2006). 15. S.D. Park, M. Todo, K. Arakawa, and M. Koganemaru, Polymer, 47, 1357 (2005).

16. M. Murariu, A. Silva Ferreira, M. Alexandre, and P. Dubois, Polym. Adv. Technol., 19, 636 (2008). 17. J. Ferry, Viscoelastic Properties of Polymer, Wiley, New York (1980). 18. M. Bousmina, P. Bataille, S. Sapieha, and H. Schreiber, J. Rheol, 39, 499 (1995). 19. Q. Zheng, M. Du, B.B. Yang, and W.G. Polymer, 42, 5473 (2001). 20. M. Du, J. Gong, and Q. Zheng, Polymer, 45, 6725 (2004). n, Handbook of Plasticizers, 21. A. Marcilla and M. Beltra ChemTec Publishing Toronto (2004). 22. N. Song, L. Zhu, X. Yan, Y. Xu, and X. Xu, J. Mater. Sci., 43, 3218 (2008). 23. J.M. Dealy and K.F. Wissbrun, Melt Rheology and Its Role in Plastics Processing, Van Nostrand Reinhold, New York (1990). 24. H. Li and M.A. Huneault, Polymer, 48, 6855 (2007). 25. L.A. Utracki, Polymer Alloys and Blends-Thermodynamics and Rheology, Hanser Publishers, Munich (1990).

116 POLYMER ENGINEERING AND SCIENCE -2012

DOI 10.1002/pen