Abstract

Poly(lactic acid) (PLA) is one of the most popular biodegradable thermoplastics in the market of 3D printing filaments used in the material extrusion (ME) technique. This is because it can be printed easily at low temperatures. However, its inherent brittleness limits its use in many applications. In this work, the toughness of PLA filament was improved by blending with various types of rubbers including natural rubber (NR), acrylic core–shell rubber (CSR), and thermoplastic polyurethane (TPU) in the amount of 15% by weight. PLA/TPU filament was found to have the smoothest surface with the best shape and dimension stability, while PLA/NR filament rendered the highest tensile toughness. In term of the effect of printing temperature, the highest printing temperature in this study (210°C) provided the highest smoothness with the best shape stability and dimension accuracy. Interestingly, the tensile toughness and elongation at break of 3D printed specimens were found to be higher than those of compression-molded specimens for all filament types. This could be explained by the ability of the 3D printing technique to produce specimens that aligned in the printing direction in a fiber-like pattern.

1. Introduction

Additive manufacturing, also known as 3D printing, has become a very interesting forming process as it possesses the ability to fabricate customized objects with complex designs at a small production cost [1, 2]. Furthermore, it could minimize material waste potentially making the production more sustainable [3, 4]. Material extrusion (ME) is the most widely used 3D printing technique due to its low machine and material costs. This technique creates products by extruding a molten thermoplastic filament layer by layer through a nozzle [5]. So far, 3D printing has been employed in the preproduction, prototyping, and postfabrication phases. In the automotive and aerospace industries, the use of 3D-printed components as part of the design process is commonplace. This is because new designs can be generated without needs of new molds or complex tooling.

Nowadays, many polymers has been investigated as 3D printing filaments, such as acrylonitrile butadiene styrene (ABS) [6, 7], polycarbonate (PC) [8], polyetherimide (PEI) [9], polyamide (PA) [10], polylactic acid (PLA) [11, 12], and high- and low-density polyethylene (HDPE and LDPE) [13]. However, the most popular commodity thermoplastic filaments used in ME 3D printers are PLA and ABS [14, 15]. The advantages of PLA filaments are lower printing temperature, minimal warpage, and good optical transparency. Furthermore, PLA is a cheaper thermoplastic filament with excellent biocompatibility and biodegradability [16, 17]. Nevertheless, its inherent brittleness limits the use of 3D-printed PLA specimens as demoprototypes rather than actual products in most cases [18]. Recently, PLA modified by blending with other polymers and additives has been investigated as filaments for 3D printers [1927]. Ou-Yang et al. [24] studied the blends of poly(butylene succinate) (PBS) and PLA as filaments for 3D printers and found that the higher PBS content led to higher elongation at break and impact strength, where the PBS/PLA ratios of 60/40 and 40/60 could provide good dimensional accuracy and gloss. Prasong et al. [25] found that the blends of 30 wt% poly(butylene adipate-co-terephthalate) (PBAT) in PLA could enhance the elongation at break of 3D-printed specimens by 26 times compared to PLA specimens. On the other hand, the research on using rubbers to improve the toughness of PLA filaments remains limited. So far, only two reports have been published: one used acrylic core–shell rubber (CSR) [26] and the other used natural rubber (NR) [27] as toughening agents for PLA filaments. In their work, the 3D-printed PLA/CSR specimens exhibited lower mechanical properties than those of 3D-printed PLA specimen [26]. For PLA/NR filaments, the elongation at break and impact strength of the PLA/NR 3D-printed specimens at NR content of up to 20 wt% were found to increase with increasing rubber content. However, the fluctuation in the diameters of PLA/NR filaments in this work resulted in undesirable porosities of the fabricated samples. Previously, our research group investigated the toughness improvement of PLA by blending with various rubbers and found that CSR and masticated NR could provide significantly higher impact strength and tensile toughness to compression–molded PLA samples. We also discovered that among all PLA/rubber blends, thermoplastic polyurethane (TPU) could provide the highest tensile strength [28, 29].

Therefore, the blends of PLA with CSR, NR, and TPU were investigated in this work for their potential as 3D printing filaments with enhanced toughness. The extrusion conditions used in producing filaments from these compounds were studied for their effect on the filament diameter and size consistency. Moreover, the effect of printing temperature was also investigated on various properties of 3D-printed specimens including the morphological, thermal, and mechanical properties and shape stability. The prepared 3D-printed specimens could be potentially used in the applications that require higher material toughness such as bumpers, cushioning materials, seals, helmets, or bicycle frames.

2. Experimental

2.1. Materials

Poly(lactic acid) or PLA grade 4043D was produced by NatureWorks LLC. Acrylic core–shell rubber (CSR) was manufactured by Dow Chemicals Company with the trade name Paraloid™ BPM-520. The thermoplastic polyurethane (TPU) grade WHT-1195IC was purchased from Able One Engineering Co., Ltd. (Thailand). Natural rubber (NR) grade STR5L was obtained from Natural Art and Technology Co., Ltd. (Thailand). The grade of commercial PLA filaments used in this study was the transparent grade produced by SUNLU.

2.2. Raw Material Preparation
2.2.1. Preparation of Natural Rubber Masterbatch

Due to the high molecular weight of NR, a masterbatch of NR/PLA was prepared at a 50 : 50 ratio to facilitate the dispersion of NR in the PLA matrix. The masterbatch was prepared in a Banbury-type internal mixer (YF-SBI-3 L, Yong Fong Machinery Co., Ltd., Thailand) at 150°C with a rotor speed of 60 rpm. The NR was first masticated for 3 min before adding PLA and other additives. Then, the compounding was continued until the mixing time reached 15 min.

2.2.2. Preparation of PLA/Rubber Blends

The rubber (NR/PLA masterbatch, CSR, or TPU) was blended with PLA in an internal mixer (Brabender® Plasticorder 350E 3Z) at 160°C at a rotor speed of 60 rpm for 12 min. The rubber content in the compound was fixed at 15 wt%. PLA, NR/PLA masterbatch, TPU, and CSR were dried to remove moisture before compounding. The compounds obtained were ground into small pellets with a plastic granulator (ZERMA Co., Ltd., Thailand).

2.2.3. Preparation of 3D Printing Filaments

In this research, filaments were fabricated by using a twin-screw plastic extruder (Labtech Engineering Co., Ltd., Thailand) with an L/D ratio of 40, a screw diameter of 16 mm, and a die diameter of 1.75 mm. The temperature profiles in the barrel were set at 160°C–190°C from the feeding hopper to the die head. To obtain the appropriate filament diameter in the range of 1.7–1.9 mm, the screw speeds were varied from 40 to 90 rpm, while the nip roll speed was fixed at 3 m/min.

2.2.4. Specimen Preparation

In this work, the PLA/rubber specimens were prepared using two methods: conventional compression molding and ME 3D printing.

(1) Preparation of Compression–Molded Specimens. In this study, the compression–molded specimens were prepared as a reference for comparison. The dried PLA and PLA/rubber blends were compressed into sheets at 190°C and a pressure of 1500 psi for 20 min. Then, the compressed sheets were then cut into dumbbell-shaped specimens according to the ASTM D368 type IV standard by laser cutting. The mechanical properties of at least five specimens were measured for each formulation.

(2) Preparation of 3D-Printed Specimens. The obtained filaments in Section 2.2.3 and commercial PLA filament were printed into dumbbell-shaped specimens by using a ME 3D printer (Finder 2.0, Zhejiang Flashforge 3D Technology Co., Ltd) with a fan cooling at the nozzle head. The specimen size conformed with the ASTM D368 type IV standard. The printed specimens were fabricated with a hexagon pattern and 100% infill. The specimens were printed on an unheated bed with a first layer height of 0.21 mm and the latter layer height of 0.14 mm. The printing temperatures varied at 165°C, 190°C, and 210°C. The printing and travel speeds were set as 50 mm/s and 60 mm/s, respectively.

2.2.5. Morphological Study

The surfaces of a filament and 3D-printed specimens were observed using a digital microscope (Dino-Lite Premier AM-3013 T) at 50x and 210x magnifications. The morphologies of the filament cross-sections and the 3D-printed specimens were investigated by the Hitachi (S-3400 N Type II) Scanning Electron Microscope (SEM) operated at 15 kV and magnifications of 100x and 3000x. For the preparation of SEM specimens, the filaments and the 3D-printed specimens were immersed in liquid nitrogen for 15 min and 6 hours, respectively, before being immediately fractured and coated with gold for 200 sec.

2.2.6. Melt Flow Index (MFI) Testing

Melt flow indexes (MFI) of the PLA and PLA/rubber blends were tested according to the ASTM D1238-13 (Procedure A), using a melt flow indexer (GOTECH Testing Machines Inc., Taiwan). The PLA and all blends were tested at 190°C with a 2.16 kg load. The units of measure are grams of material/10 minutes (g/10 min).

2.2.7. Thermal Analysis

Differential scanning calorimetry (DSC) analysis was performed using The TA Instruments Discovery DSC250 to determine the thermal properties of the samples. For the first and the second heating scans, the samples were first heated from 25°C to 250°C with a heating rate of 10°C/min and kept isothermal for 2 min. For the cooling step, the samples were cooled from 250°C to 25°C also at a rate of 10°C/min. The percentage of crystallinity of PLA in the blends was calculated using the melting enthalpy of 100% crystalline PLA (93.7 j/kg) [30].

2.2.8. Surface Roughness Measurement

The surface roughness of the 3D-printed specimens was measured using a Mitutoyo Surftest SJ-410 tester. The average surface index or Ra values were also calculated from the measured readings of conventional stylus-probe of surface roughness instrument, where the stylus speed was 0.5 mm/s.

2.2.9. Tensile Measurement

The tensile properties were determined according to ASTM D638 type IV standard using a universal tensile testing machine (Narin Instrument Co., Ltd., Thailand) with a load cell of 3 kN. The distance between the grips was 65 mm, and the crosshead speed was 5 mm/min. At least three test samples were tested for each sample, and the average values were presented. The tensile toughness in J/m3 was calculated as the area under the stress–strain curve following the equation:

where is the stress in MPa, while and are strain and final strain at break, respectively.

3. Results and Discussion

3.1. The Effect of Extrusion Conditions on Filament Morphologies

The 3D printing filaments in this study were produced using a twin-screw extruder. The appropriate die head temperature for all compounds was determined to be 190°C. The morphologies of the filaments produced from PLA/rubber blends at various screw speeds and the MFI of commercial PLA, and the PLA/rubber blends are summarized in Table 1 and Figure 1, respectively. The filament sizes of all blends became larger at higher screw speeds due to a stronger die swell effect. It is worth nothing that the PLA/NR blend required the highest screw speed of 90 rpm to achieve the filament diameter above 1.75 mm, potentially due to its higher MFI (lower viscosity), leading to a lower degree of die swell. However, at screw speeds of 70 rpm and above, the PLA/NR filaments were rougher, with higher fluctuation in the filament diameters. This could be explained by the shear-induced aggregation of NR particles, as observed in the SEM micrograph in Table 1(a). This effect could happen in incompatible blends under high shear [31]. On the other hand, SEM micrographs of PLA/TPU (Table 1(b)) and PLA/CSR (Table 1(c)) revealed the characteristics of compatible blends, where the rubber sizes were smaller than those of NR and were well-dispersed in PLA matrices at all screw speeds. Consequently, this led to smoother filament surfaces and lower diameter fluctuation. Moreover, these blends required a lower screw speed (70 rpm) than that for PLA/NR (90 rpm) to obtain filament diameters above 1.75 mm.

3.2. The Effect of Printing Temperature on the Specimen Properties

For the 3D printing step, the effects of printing temperature on the shape stability, surface morphology, and surface roughness of 3D-printed specimens were studied at 165°C, 190°C, and 210°C for all filaments. In this work, the shape stability was measured as a warpage angle and dimensional accuracy in the width direction of the 3D-printed specimen, where the warpage angle was measured as described in Figure 2. The warpage angle and dimensional discrepancy (shrinkage) of all specimens are summarized in Figure 3. As can be seen, the warpage angle and shrinkage of all samples reduced when increasing the printing temperature due to the lower difference between the temperatures of the top and bottom layers. When comparing between different PLA/rubber blends, PLA/NR provided the highest warpage angle and shrinkage indicating the poorest shape stability due to its highest percent crystallinity (%) as shown in Figure 4, while PLA/CSR and PLA/TPU specimens possessed the lowest warpage angles, which were even lower than that of commercial PLA consistent with their lower %.

The surface roughness of the specimens was measured as the average surface index or Ra values (Figure 5) and observed from the sharpness of streaks that appeared on the surface (Table 2). As can be seen, the highest printing temperature (210°C) provided the smoothest surface due to lower viscosity, which led to faster flow and better layer infusion. The surfaces were significantly rougher at the lowest printing temperature of 165°C, except for PLA/TPU, which possessed smooth surface at all printing temperatures. When comparing between different PLA/rubber blends, PLA/NR filaments provided the roughest surface when printing at 165°C, while their surface roughness was comparable at higher printing temperatures.

The effect of rubbers and printing temperature on mechanical properties of dumbbell-shaped specimens printed at 190°C and 210°C is shown in Figure 6. As can be seen, the additions of CSR and NR could increase the toughness of PLA, while the toughness of PLA blended with TPU could not change significantly. On the other hand, the toughness of specimens produced from PLA/rubber filaments increased slightly at the higher printing temperature. This is because the presence of low-Tg rubbers in PLA led to better adhesion between printed layers at higher temperature as indicated by the SEM micrographs in Table 3.

When comparing the effect of forming techniques, the stress-strain curves of specimens fabricated by 3D printing were compared to those fabricated by compression molding as shown in Figure 7. Their tensile strength, tensile toughness, and elongation at break are summarized in Figure 8. As can be seen, the shapes of stress-strain curves of samples fabricated by 3D printing and compression molding were similar for all formulations. Nevertheless, the 3D-printed specimens processed higher tensile strength, elongation at break, and tensile toughness. Interestingly, while most specimens produced by 3D printing could be stretched by almost twice as much as those produced by compression molding, the 3D-printed PLA/NR specimens could be stretched by almost 7 times more. This is probably due to the outstanding tensile strength of NR. However, it should be noted that further stretching of the 3D-printed PLA/NR specimen beyond twice the elongation of compression molded one resulted in fibrillation of the PLA/NR printed layers as shown in Figure 9(a). This could be explained by the ability of the 3D printing technique to produce specimens that aligned in the printing direction in a fiber-like pattern, while this phenomenon was not found in the compression-molded specimens (Figure 9(b)). Note that necking was not clearly observed in our heterogeneous systems as the stretching occurred predominantly in the dispersed rubber phase.

4. Conclusion

In this work, the PLA was blended with various types of rubber (TPU, CSR, and NR) and extruded into 3D printing filaments, and the results are concluded as follows: (i)It was found that PLA/NR required higher screw speed (90 rpm) than PLA/TPU and PLA/CSR (70 rpm) to obtain the appropriate filament diameter in the range of 1.7–1.9 mm(ii)The PLA/NR 3D-printed specimens possessed the highest tensile toughness among all. The recommended printing temperature is 210°C to obtain better shape stability(iii)PLA/CSR and PLA/TPU filaments could not provide higher toughness than the commercial PLA filament(iv)The PLA/CSR specimens possessed the best shape stability and the smoothest surface (at higher printing temperatures), while PLA/TPU could provide smooth surface even at the lowest printing temperature (165°C)(v)3D printing technique could provide higher tensile toughness, tensile strength, and elongation at break than compression molding(vi)These PLA/rubber filaments have a good potential for producing specimens with good shape stability, surface smoothness, and high tensile toughness, especially at the printing temperature of 210°C(vii)The performance of these PLA/rubber filaments could be further enhanced by improving on phase compatibility and thermal stability of the blend formulations

Data Availability

The data could be requested by emailing the corresponding author.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors gratefully acknowledge the Ph.D. Scholarship from the Faculty of Engineering, and partial funding from the Research Unit in Polymer Rheology and Processing, Thammasat University.