Improving Mechanical Properties for Extrusion-Based Additive Manufacturing of Poly(Lactic Acid) by Annealing and Blending with Poly(3-Hydroxybutyrate)

Based on differential scanning calorimetry (DSC), X-ray diffraction (XRD) analysis, polarizing microscope (POM), and scanning electron microscopy (SEM) analysis, strategies to close the gap on applying conventional processing optimizations for the field of 3D printing and to specifically increase the mechanical performance of extrusion-based additive manufacturing of poly(lactic acid) (PLA) filaments by annealing and/or blending with poly(3-hydroxybutyrate) (PHB) were reported. For filament printing at 210 °C, the PLA crystallinity increased significantly upon annealing. Specifically, for 2 h of annealing at 100 °C, the fracture surface became sufficiently coarse such that the PLA notched impact strength increased significantly (15 kJ m−2). The Vicat softening temperature (VST) increased to 160 °C, starting from an annealing time of 0.5 h. Similar increases in VST were obtained by blending with PHB (20 wt.%) at a lower printing temperature of 190 °C due to crystallization control. For the blend, the strain at break increased due to the presence of a second phase, with annealing only relevant for enhancing the modulus.


Extra Information for Materials
MFI and TGA data for the materials in this study are shown in Table S1. The melt flow rate of PLA, PHB and PLA/PHB filaments ranged from 180 °C to 230 °C with load of 2.16 kg. The degradation behavior of the filaments was measured by thermogravimetrical analysis (TGA) under N2 flow. The samples were heated from room temperature to 550 °C at a rate of 10 °C min −1 . The MFI value for PHB and the PHB/PLA blends was much larger than that of PLA alone and the onset degradation temperature (T5% lost) of PHB was much lower than that of PLA. Notably, PHB is more sensitive to temperature than PLA. Based on previous studies, PHB/PLA can be printed using a lower temperature than PLA, and the material performs well when the melt flow index is close to 10 g (10 min) -1 [1]. PLA and PLA/PHB showed the best mechanical properties at 210 °C and 190 °C, respectively (Table S2). The impact strength was lower and was independent of temperature due to insufficient flow at low temperature or thermal degradation at high temperature. These results indicate that pure PLA and the PLA/PHB blend should be printed at 210 °C and 190 °C, respectively.  Both PLA and PLA/PHB were well printed and won't wrap after annealing. The non-annealed PLA was transparent, after annealing it turned opaque. PLA/PHB blend showed the same appearance before and after annealing.

DSC Results
PLA/PHB blend presents lower Tg, Tcc and smaller ∆Hcc than pure PLA, proving that PHB enhanced the crystallization ability of PLA. The annealed samples all reach maximum crystallinity with no recrystallization peak shown after annealing.
The 110/200 and 203 planes of PLA/PHB were shifted upward compared to PLA, suggesting that PHB increases the space for PLA between the crystal planes. However, the intensity of the 110/200 and 203 peaks for PLA/PHB was decreased slightly and broadened compared to that for PLA. This difference can be attributed to the PLA/PHB blends interfering with the spherulite structure of each other due to the different crystallization kinetics [10]. Tensile and impact results of PLA and PLA/PHB blend before and after annealing are shown in Table S4. Analysis of variance (ANOVA) was performed on the dataset for tensile tests using SPSS software to check data reliability (Table S5). Sig. values <0.05 indicated significant differences between the two groups.

Dimension Stability of Samples
Density test: A Precisa XR 2055M-DR was used for density determinations based on the Archimedes principle. As a reference, 99.8% ethanol with a density of 0.803 g cm -3 was used. Calibration was based on a standardized glass rod. The density was measured at room temperature (20 °C) according to ISO 1183.
Dimension changes: Length, width and thickness of the printed bars were measured manually before and after annealing using a caliper to determine the dimension stability after annealing. Negative values in Table S8 represent the percentage decrease in dimension, whereas positive values represent the percentage increase in the dimension after annealing compared to that before annealing.
Density results and dimension changes are shown in Tables S7 and S8, respectively. The percentage difference in density of annealed samples is within 1% of the non-annealed sample and hardly any voids were seen between the printed strands. The dimensions of non-annealed and annealed samples varied by less than 3%. Meanwhile, the thickness increased but the width and length of the samples decreased after annealing, in accordance with previous findings [11]. The increase in thickness may be due to the inner stress of layers during printing, wherein strands are forced onto earlier layers that then stretch back to a rounder shape after annealing. Given the small variations seen among density and dimensions, dimension changes that occur during printing should have virtually no effect on the mechanical properties of the samples.