The Effect of Mold Conditions on Heat Resistance of Injection- Molded Stereocomplex Polylactide-b-polyethylene Glycol-b- Polylactide Bioplastic

The effect of mold conditions was investigated in terms of mold temperature (30C and 90C) and cooling time (30 s and 60 s) on the heat resistance of injection-molded bars for stereocomplex polylactide-b-polyethylene glycol-b-polylactide (scPLA-PEG-PLA). Comparative study was performed for poly(L-lactide) (PLLA) and PLLA-b-PEG-b-PLLA (PLLA-PEG-PLLA). scPLA-PEG-PLA was 90/10 (w/w) PLLA-PEG-PLLA/poly(D-lactide) blend. scPLA-PEG-PLA exhibited the easiest crystallization upon cooling scan as shown by differential scanning calorimetry (DSC). Higher mold-temperature and longer cooling-time induced higher degree of crystallinity as assessed by X-ray diffractometry (XRD) except for PLLA bars. The heat resistance of both PLLA-PEG-PLLA and scPLA-PEG-PLA bars was improved with increased mold-temperature and cooling-time as shown by dynamic mechanical analysis (DMA), vicat softening temperature (VST) and heat distortion-resistance tests except for PLLA bars. In conclusion, the heat resistance of injection-molded bars prepared at 90 ̊C mold temperature was in the order scPLA-PEG-PLA > PLLA-PEG-PLLA > PLLA. The results suggested that flexible PLLA-PEGPLLA and scPLA-PEG-PLA with high degrees of crystallinity were successfully obtained by injection molding for use as good heat-resistant bioplastic products.


Introduction
Poly(L-lactic acid) or poly(L-lactide) (PLLA) is an important bio-based plastic because it features non-toxicity, biodegradability, bio-renewability and good processability [1−3]. PLLA and modified PLLA have been successfully utilized in several industrial applications such as biomedical, pharmaceutical, automotive interiors and packaging [2,4−6]. However, low flexibility, poor crystallizability and low heat-resistance were the problems in practical use of PLLA [7−9].
Poly(L-lactide)-b-polyethylene glycol-b-poly(L-lactide) triblock copolymers (PLLA-PEG-PLLA) showed greater flexibility and faster crystallization-rate than PLLA due to flexibility of PEG middleblocks causing an enhanced plasticizing effect [10,11]. However, the heat resistance of PLLA-PEG-PLLA was still poor [12]. The heat resistance of PLLA and PLLA-PEG-PLLA is directly related to degree of crystallinity [13−15]. This indicates that crystallizability of PLLA-PEG-PLLA is not enough to improve its heat resistance. Nucleating agents and heat treatment have been used to increase the degree of crystallinity of PLLA and so improve its heat-resistance [13−15].
Poly(D-lactide) (PDLA) has been blended with PLLA to form a stereocomplex polylactides (scPLA) which consisted of homo-and stereocomplex (sc) crystallites [16−19]. The homo-and sc-crystallites of PLA had melting temperatures at lower and higher than 200 o C, respectively [20]. The scPLA showed better thermo-mechanical properties and heat resistance than PLLA due to faster crystallization of sccrystallites upon cooling [19,21,22]. However, scPLA with high PDLA contents or high content of sccrystallites have to be processed at high temperature (240 − 250 o C) which risks thermal decomposition [12, 23−25]. Meanwhile PLLA has been blended with small amounts of PDLA for enhancing PLLA crystallization [19, 26−28]. The result was that sc-crystallites can act as nucleating sites to enhance crystallizability of PLLA phases. This process can be performed at 200 o C or below [19,27].
Higher mold-temperature was reported to enhance crystallizability and heat resistance of injectionmolded bars for both PLLA [14] and scPLA [29]. However, no reports were present to study optimum mold-conditions for injection-molded bars of flexible PLLA-PEG-PLLA and scPLA-PEG-PLA for these purposes.
This paper studied the effects of mold temperature and cooling time on crystallizability and heat resistance of injection-molded bars of PLLA-PEG-PLLA and scPLA-PEG-PLA. PDLA of 10 wt% was blended with PLLA-PEG-PLLA to form scPLA-PEG-PLA. Annealing effect by post heat-treatment of injection-molded bars was also investigated for comparison. The thermal properties, crystalline structures, mechanical properties and heat resistance of injection-molded bars obtained were determined and discussed. Injection-molded PLLA bars were also prepared and studied for comparison.

Materials
PLLA grade 3052D was supplied by NatureWork LLC. It had melt flow index (MFI) of 14 g/10 min (210 o C, 2.16 kg). In situ chain-extended PLLA-PEG-PLLA with MFI of 26 g/10 min (190 o C, 2.16 kg) was synthesized by ring-opening polymerization of L-lactide monomer in the presence of 2.0 phr Joncryl  ADR4368 (chain extender) according to our previous work [30]. PDLA with number-average molecular weight and dispersity index of 90,000 g/mol and 2.8 from GPC respectively, was synthesized as described in our previous work [25].

Preparation of injection-molded bars
scPLA-PEG-PLA was prepared by melt blending of 90/10 (w/w) PLLA-PEG-PLLA/PDLA mixture. The mixture was dried overnight at 50 o C in a vacuum oven before melt blending using an internal batch mixer (Hakke Polylab OS system) at 190 o C with a rotor speed of 100 rpm for 4 min. The injectionmolded bars with dumbbell (ASTM D 638) and rectangular (ASTM D 790) shapes of PLLA3052D, PLLA-PEG-PLLA and scPLA-PEG-PLA were obtained by injection molding the samples in an injection-molding machine (Nissei Plastic Industrial Co., Ltd., Japan, Model PS40E5ASE) at 170 o C. Mold temperatures and cooling times of 30 o C for 60 s, 90 o C for 30 s and 90 o C for 60 s were used. The obtained bars prepared with mold temperature at 30 o C for 60 s were annealed at 90 o C for 24 h in an air oven in order to study the annealing effect.

Characterization
Thermal transition properties of PLLA3052D, PLLA-PEG-PLLA and scPLA-PEG-PLA pellets were determined using a Differential Scanning Calorimeter (DSC, Perkin-Elmer, Model Pyris Diamond) under nitrogen gas flow. For DSC heating scan, the sample (3-5 mg) in a sealed aluminum pan was firstly melted at 200 o C for 3 min. to remove thermal history, after which it was quenched to 0 o C before scanning from 0 o C to 200 o C with a heating rate of 10 o C/min. The degree of crystallinity from DSC of the PLA homo-crystallites (DSC-Xc,hc) was calculated from the enthalpies of melting (∆Hm) and cold crystallization (∆Hcc) with the following equation. where the values 93.7 J/g is the Hm for 100%Xc PLLA [31]. WPLLA is the PLLA weight-fraction of the samples that is calculated from the PLLA fraction (PLLA =1.00 and PLLA-PEG-PLLA = 0.83 obtained from 1 H-NMR) [11]. For DSC cooling scans, the sample was firstly melted at 200 o C for 3 min to erase the thermal history before scanning from 200 o C to 0 o C at a cooling rate of 10 o C/min under nitrogen gas flow.
The crystalline structures of injection-molded bars were investigated using a X-Ray Diffractometer (XRD, Bruker, Model D8 Advance) with CuK radiation at 40 kV and 40 mA. The samples were scanned in the range of 2 from 5 o to 30 o at a rate of 3 o /min. The degree of crystallinity from XRD for PLA hc-crystallites (XRD-Xc,hc) and degree of crystallinity from XRD for PLA sc-crystallites (XRD-Xc,sc) were determined from equations (2) and (3), respectively [32]. Total degree of crystallinity from XRD (XRD-Xc) and percentage of sterecomplex (%SC) were calculated from equations (4) and (5), respectively.
Thermo-mechanical properties of injection-molded bars, 3 × 13 × 60 mm in size, were tested using a Dynamic Mechanical Analyzer (DMA, TA Instruments Model Q800) in a multi-frequency strain mode. The samples were heated at a rate of 2 o C/min from 30 o C to 140 o C at a scan amplitude of 20 m and a scan frequency of 1 Hz.
The heat deflection temperature (HDT) of injection-molded bar samples was measured using a HDT/Vicat machine (Yasuda, Model HD-PC). The bar samples were 3  13  125 mm in size and were heated in silicone oil at a rate of 2 o C/min under 0.455 MPa load according to ASTM D 648-01. The HDT value was obtained for a 0.25 mm deflection of the bar sample. Average value was obtained from triplicate measurements.
The Vicat softening temperature (VST) of injection-molded bar samples was determined using a HDT/Vicat machine (Yasuda, Model HD-PC). Bar samples (3  20  20 mm) were heated in silicone oil at a rate of 2 o C/min under 50 N load of a flat-ended needle according to ISO 306. A surface area of a flat-ended needle was 1 mm 2 . The VST value was obtained when the needle penetrated the sample by 1 mm. Average value was obtained from triplicate measurements.
Heat distortion resistance of injection-molded bars was directly observed by half hanging in an oven at 85 o C for 5 min by following the method [9] with some modifications. Photographs of bar samples after the test were recorded to compare heat deformation of bar samples.

Thermal transition properties
DSC method was used to determine the thermal transition properties of samples including glass transition temperature (Tg), cold-crystallization temperature (Tcc), enthalpy of cold crystallization (Hcc), melting temperature (Tm) and enthalpy of melting (Hm) from DSC heating curves ( Figure  1(above)) as well as crystallization temperature (Tc) and enthalpy of crystallization (Hc) from DSC cooling curves (Figure 1(below)). The DSC results are summarized in Table 1.  From DSC heating curves, Tg of PLLA3052D was 59 o C while PLLA-PEG-PLLA exhibited a lower Tg at 31 o C. This seemed to due to plasticizing effect of flexible PEG middle-blocks [10,11]. Tg of scPLA-PEG-PLA could not be detected. From DSC cooling curves, PLLA3052D had no Tc peak. PLLA-PEG-PLLA and scPLA-PEG-PLA had a Tc at 103 o C and 120 o C, respectively indicating that crystallization upon cooling of scPLA-PEG-PLA was easier than that of PLLA-PEG-PLLA. sc-crystallites in scPLA-PEG-PLA enhanced crystallization of PLLA end-blocks by acting as nucleating sites [26][27][28]. Tm of sccrystallites in PLLA-PEG-PLLA/PDLA blends were higher than 200 o C. Thus, sc-crystallites did not melt during removal of their thermal history at 200 o C and acted as nucleating sites during the cooling scan [26].

Crystalline structures
The crystalline structures of injection-molded bars were estimated from XRD patterns as shown in Figure 2. PLLA3052D bars prepared at 30 o C mold temperature for 60 s had no XRD reflexes indicating that they were completely amorphous. The small XRD reflexes of PLLA homo-crystallites (hc) at 2 = 16.8 o [33,34] were found for both PLLA3052D bars prepared at 90 o C mold temperatures for 30 s and 60s. Therefore, increasing the mold temperature from The degrees of crystallinity of injection-molded bars from XRD are summarized in Table 2. It can be seen that at 90 o C mold temperature for both the cooling times could not significantly improve crystallization of injection-molded PLLA3052D bars. The XRD-Xc,hc of PLLA3502D bars was increased up to 51.7% when they were annealed at 90 o C for 24 h. The XRD-Xc,hc of PLLA-PEG-PLLA bars prepared at 90 o C mold temperature were higher than PLLA3052D. The flexible PEG middle-blocks enhanced chain mobility to improve PLLA end-block crystallization of PLLA-PEG-PLLA by a plasticizing effect [10]. The XRD-Xc,hc of PLLA-PEG-PLLA bars prepared at 90 o C mold temperature increased from 25.6% to 30.8% as the cooling time was increased. The annealed PLLA-PEG-PLLA bars exhibited the highest XRD-Xc,hc (46.2%).
The scPLA-PEG-PLA bars prepared at 30 o C mold temperature had the XRD-Xc,hc and XRD-Xc,sc of 1.4% and 11.2%, respectively. Both the XRD-Xc,hc and XRD-Xc,sc of scPLA-PEG-PLA bars prepared at 90 o C mold temperature steadily increased as the cooling time increased from 30 to 60 s. The XRD-Xc of scPLA-PEG-PLA bars increased and %SC decreased with the increase of mold temperature and cooling time. The annealed scPLA-PEG-PLA bars had the highest XRD-Xc (53.1%) and the lowest %SC (26.0%). It should be noted that the XRD-Xc of scPLA-PEG-PLA bars were larger than those of PLLA-PEG-PLLA bars at the same mold temperature and cooling time. This suggests higher crystallizability of the scPLA-PEG-PLA bars. The crystallinity of injection-molded PLLA bars cannot be largely improved at mold temperature less than 100 o C [15]. But the injection-molded bars of PLLA-PEG-PLLA and scPLA-PEG-PLA exhibited largely increased in their crystallinity contents at 90 o C mold temperature.

Mechanical properties
The mechanical properties of injection-molded bars were determined by tensile tests as summarized in Table 3. The ultimate tensile stress of PLLA3052D bars prepared with different mold-conditions and annealing were in range 57.1−63.2 MPa suggesting that the XRD-Xc changes of PLLA3052D bars did not significantly affect the ultimate tensile stress. However, the strain at break slightly decreased and Young's modulus increased as the mold temperature and cooling time increased. This may be explained by an increasing XRD-Xc of PLLA3052D bars inhibiting chain mobility for extension of bar samples and improving their stiffness [14].
With increasing mold temperature and cooling time, the ultimate tensile stress and Young's modulus of PLLA-PEG-PLLA and scPLA-PEG-PLA bars increased as well as strain at break decreased. It should be noted that their strains at break dramatically dropped as the mold temperature was increased. This is due to the XRD-Xc largely increasing as the mold temperature increased ( Table 2). All the annealed bars with the high XRD-Xc exhibited the highest tensile stress and Young's modulus whereas they had the lowest strain at break. However, both the PLLA-PEG-PLLA and scPLA-PEG-PLA bars exhibited strain at break higher than the PLLA3052D indicated they were more flexible than the PLLA3052D. Finally, the tensile properties of the scPLA-PEG-PLA bars were better than the PLLA-PEG-PLLA bars. This can be explained by the sc-crystallites enhanced mechanical properties by acting as physical crosslinking sites in amorphous regions [36].

DMA
The DMA method has been widely used to investigate heat resistance of PLLA and modified PLLA using storage-modulus change as a function of temperature [16,19,22,37,38]. Usually, the storage modulus of PLLA is high at temperature below its Tg (around 60 o C) because it is in the glassy state. The storage modulus of PLLA drastically dropped when the temperature passed the Tg region before increasing again caused by cold crystallization of PLLA [16]. This demonstrates the PLLA has poor heat-resistance due to its low degree of crystallinity after melt processing. Figure 3 shows the storage-modulus changes on temperature of injection-molded bars. From Figure  3a, all the PLLA3052D bars prepared with mold temperatures at 30 o C and 90 o C exhibited large decreases of the storage moduli in the Tg region (60−80 o C) suggesting that they had poor heat-resistance. This could be explained by the PLLA3052D bars containing low XRD-Xc. Whereas the annealed PLLA3052D bars were slightly decreased in storage modulus indicated that they were good heatresistance. This is because the PLLA3052D bars had the highest XRD-Xc. The results confirmed that the PLLA with high Xc exhibited good heat-resistance [14,16,19].
For PLLA-PEG-PLLA and scPLA-PEG-PLA bars, the lowest storage-moduli increased with the mold temperature and cooling time as shown in Figures 3b and 3c, respectively. The results suggested that the heat resistance of both PLLA-PEG-PLLA and scPLA-PEG-PLA bars were improved by increasing the mold temperature and cooling time. The annealed PLLA-PEG-PLLA and scPLA-PEG-PLA bars showed the best heat-resistance. This is due to higher mold-temperature and longer coolingtime of injection-molded bars induced greater XRD-Xc. The storage moduli at 85 o C of injection-molded bars were compared in Figure 4. A temperature at 85 o C was selected for comparison because the storage moduli increased again after 85 o C due to coldcrystallization effects. All the injection-molded bars prepared at 30 o C mold temperature had low storage moduli (Figure 4a). The storage moduli at 85 o C increased significantly as the mold temperature increased from 30 to 90 o C and cooling time increased from 30 to 60 s except PLLA3052D bars ( Figure  4b and 4c). All the annealed bars showed the highest storage moduli at 85 o C (Figure 4d). From DMA results, it can be concluded that the heat resistance of injection-molded bars are in order scPLA-PEG-PLA > PLLA-PEG-PLLA > PLLA3052D for the same mold-temperature and cooling-time.

HDT and VST
HDT and VST have been used to determine the heat resistance of injection-molded PLLA bars [9,17]. The resulting HDT and VST are also summarized in Table 3. The HDT and VST of PLLA3052D bars did not significantly change with the mold temperature and cooling time that were in ranges 45.9-46.9 o C and 56.5-57.0 o C, respectively. However, the HDT and VST were increased to 63.2 o C and 83.4 o C, respectively for the annealed PLLA3052D bars.
The HDT of PLLA-PEG-PLLA and scPLA-PEG-PLA bars increased steadily with the mold temperature and cooling time. It can be said that the HDT and VST of these injection-molded bars also depended on their XRD-Xc. The results supported the conclusion that increasing mold-temperatures and cooling-times improved heat resistance of injection-molded bars. It should be noted that the HDT of both PLLA-PEG-PLLA and scPLA-PEG-PLA bars were lower than PLLA3052D bars. This may be due to the fact tha PLLA-PEG-PLLA and scPLA-PEG-PLA are more flexible than PLLA3052D as observed from the results of strain at break from tensile test.

Heat Distortion Resistance
A half-hanging test was performed in an air oven at 85 o C for 5 min to investigate the heat-distortion resistance of injection-molded bars. The photographs of injection-molded bars after test are shown in  It can be seen that all the PLLA3052D bars prepared with mold temperatures at 30 o C and 90 o C were largely deformed indicating that they had poor heat-resistance ( Figure 5(left)). Whereas the annealed PLLA3052D bars retained original shape after the test corresponding to the results of DMA, HDT and VST as described above. The annealed PLLA3052D bars had the highest storage-modulus at 85 o C (from DMA), HDT and VST.
PLLA-PEG-PLLA and scPLA-PEG-PLA bars were deformed after the test when the mold temperature was at 30 o C as shown in Figure 5(middle, a) and 5(right, a), respectively. This is due to their low storage-moduli at 85 o C (from DMA), HDT and VST. When the mold temperature was at 90 o C, the PLLA-PEG-PLLA and scPLA-PEG-PLA bars exhibited good heat-distortion resistance for both the cooling times because their storage moduli at 85 o C (from DMA), HDT and VST increased with the mold temperature. The annealed bars of both PLLA-PEG-PLLA and scPLA-PEG-PLA showed good heatdistortion resistance because their storage moduli at 85 o C (from DMA), HDT and VST were the highest. It is worth mentioning that the degree of crystallinity of the injection-molded bars plays the key role in heat-distortion resistance.
It should be noticed that the injection-molded bars with the VST higher than 60 o C exhibited good heat-distortion resistance. While the HDT values of the PLLA-PEG-PLLA bars prepared at 90 o C mold temperature were lower than the PLLA3052D bars but they showed better heat-distortion resistance than PLLA3052D bars. This is due to the PLLA-PEG-PLLA being softer and more flexible than the PLLA [39,40]. The results indicated that the VST were more reliable than the HDT as heat-resistance parameters as shown in the literature [17]. In addition, the results of DMA and heat-distortion resistance in this work could be used to support the VST results for heat-resistant property of PLLA-based bioplastics.

Conclusions
In this work, the effects of mold temperature and cooling time on heat resistance of injection-molded bars were investigated for scPLA-PEG-PLA compared with PLLA3052D and PLLA-PEG-PLLA. DSC results indicated the crystallizability of samples was in order scPLA-PEG-PLA > PLLA-PEG-PLLA > PLLA3052D. From XRD, both the PLLA3052D and PLLA-PEG-PLLA bars prepared at 30 o C mold temperature were completely amorphous in character whereas the scPLA-PEG-PLA bars had both hc and sc-crytsallites. The XRD-Xc of PLLA-PEG-PLLA and scPLA-PEG-PLA bars increased with the mold temperature and cooling time but PLLA3052D bars did not. Increasing the mold temperature and cooling time improved the heat resistance of injection-molded bars for both PLLA-PEG-PLLA and scPLA-PEG-PLA except for PLLA3052D bars as observed from their increasing the lowest storagemoduli from the DMA study, HDT, VST and heat-distortion resistance.
It can be concluded that the 90 o C mold temperature seem to improve the XRD-Xc and heat resistance of both the injection-molded bars for PLLA-PEG-PLLA and scPLA-PEG-PLA except for PLLA3052D.
The results could provide guidance toward fabrication of high heat-resistant and flexible scPLA-PEG-PLA at industrial scale for use as bioplastic products.