Characterization of nanocomposite filaments developed for additive manufacturing Cite as: AIP Conference Proceedings 1779, 100006 (2016); https://doi.org/10.1063/1.4965574 Published Online: 31 October 2016 Fugen Daver, Robert A. Shanks, and Milan Brandt ARTICLES YOU MAY BE INTERESTED IN Evaluation of thermal and electrical conductivity of carbon-based PLA nanocomposites for 3D printing AIP Conference Proceedings 1981, 020158 (2018); https://doi.org/10.1063/1.5046020 3D printing of polypropylene using the fused filament fabrication technique AIP Conference Proceedings 1896, 040014 (2017); https://doi.org/10.1063/1.5008040 Influence of 3D printing parameters on the properties of PLA/clay nanocomposites AIP Conference Proceedings 1981, 020064 (2018); https://doi.org/10.1063/1.5045926 AIP Conference Proceedings 1779, 100006 (2016); https://doi.org/10.1063/1.4965574 © 2016 Author(s). 1779, 100006 Characterization of Nanocomposite Filaments Developed for Additive Manufacturing Fugen Daver1, a), Robert A. Shanks2, Milan Brandt1 1 Centre for Additive Manufacturing School of Aerospace, Mechanical, and Manufacturing Engineering, RMIT University, PO Box 71, 3083 Victoria, Australia 2 School of Applied Sciences, RMIT University, GPO Box 2476, 3001, Victoria, Australia a) Corresponding author: fugen.daver@rmit.edu.au Abstract. The study aims to characterize innovative filaments in the form of polyolefin-rubber nanocomposites developed for additive manufacturing. Polyolefin-rubber filaments were consisted of linear low density polyethylene and de-vulcanised, activated rubber. A compatibilizer in the form of maleic anhydride grafted polyethylene was used to enhance adhesion between the two phases. Multi-walled carbon nanotubes were introduced for electrical conductivity. Various compositions of filament material were tested for mechanical properties, electrical conductivity and timedependent deformation behavior as demonstrated in creep-recovery experiments. A four element model of Maxwell and Voigt-Kelvin was employed to analyze the creep behavior of the nanocomposites. Results were discussed in terms of the effect of (i) carbon nanotubes, (ii) compatibilizers and (iii) composition of each nanocomposites. INTRODUCTION Additive manufacturing aims to produce parts directly from Computer Aided Design (CAD) model of the part without the need for building costly molds. Fused Deposition Molding (FDM) is one of the additive manufacturing techniques which allows production of free-form manufacturing of prototypes and end-use parts. In FDM, plastic filament of certain cross sectional area passes through an extruder and molten plastics is deposited on to a working platform layer by layer by means of an extruder head travelling across the platform according to the CAD model of the part. The platform is lowered as each layer is formed, subsequent layers bond via thermal fusion to form the complete part. Flexibility and convenience offered by FDM techniques has opened the doors for automotive, aerospace and biomedical applications. However, the number and type of materials available in filament form for FDM applications are currently rather limited. In this study, we have explored polyolefin-rubber nanocomposites as filament material in FDM applications and evaluated their time dependent deformation behavior and assessed their electrical conductivity. EXPERIMENTAL Materials Linear low-density polyethylene (LLDPE) was mixed with maleic anhydride grafted polyethylene (MA-g-PE) which is grafted with maleic anhydride 0.90 %·w/w. Functionalized rubber particles (FRP), which were produced according to a patented process, were obtained from Polymeric Powders Company [1,2,3]. Multi-walled carbon nanotubes (MWCNT) were incorporated into the polyolefin–rubber composites in the form of a masterbatch Plasticyl LDPE2001 which is based on low density polyethylene (LDPE) [4]. Proceedings of the Regional Conference Graz 2015 – Polymer Processing Society PPS AIP Conf. Proc. 1779, 100006-1–100006-5; doi: 10.1063/1.4965574 Published by AIP Publishing. 978-0-7354-1441-9/$30.00 100006-1 Sample preparation LLDPE was fed into the pre-heated non-intermeshing, counter rotating internal batch mixer at 180 °C and it was mixed at 50 rpm for 1 min to ensure complete melting of polymer prior to the introduction of the masterbatch Plasticyl LDPE2001, FRP and MA-g-PE, which aims to improve compatibility between the rubber and the polyethylene. All four ingredients were mixed for another 5 min at 180 °C before the mixture was removed for further processing. Table 1 shows the compositions of the LLDPE–FRP nanocomposites. All materials were compression molded for characterisation purposes. Whereas, the composites PB4 and C-PB4 were formed into filaments with an average diameter of 1.73 mm by means of a twin screw extruder which was equipped with a series of rollers to take-up and wind the filament to be used in the fused deposition modelling process. Details of the sample preparation are given elsewhere [5]. TABLE 1. Composition of the LLDPE–FRP nanocomposites. Sample Code Composition (%·w/w) FRP MA-g-PE 30 - PB1 LLDPE 70 MWNT - PB3 30 70 - - PB4 65 30 5 - PB5 25 70 5 - C-PB4 62 30 5 3 C-PB5 22 70 5 3 Fused deposition modeling (3D printing) of nanocomposite filaments Fused deposition modelling which is an additive manufacturing process was performed by means of Makerbot Replicator 2. Filament of polyolefin rubber nanocomposite was heated up to 260 oC and extruded through a nozzle that moves in the x-y plane. When the extruded material deposited, it solidified and generated a single layer. The fusion of subsequent layers resulted in the actual test specimens formed based on the 3D CAD model of (i) Type 1 standard specimen according to Tensile Testing of Polymers (ASTM D638) and (ii) Rectangular specimens of 3.2x12.7x125 mm dimension according to Flexural Modulus Testing of Polymers (ASTM D790). Characterization of polyolefin rubber nanocomposites Tensile testing was performed using an Instron Series 4467 Universal Test Instrument equipped with a 2 kN load cell and with Instron Bluehill Software (Instron Co., Canton MA, USA D638). The strain was measured with an Instron Long Travel Extensometer. The 3D printed specimens were compared with the compression moulded specimens. Compression molding was performed at 180 °C for 5 min under a force of 50 kN followed by watercooling of the mould to below 50 °C prior to pressure release and ejection from the mould. Data were collected by averaging over five measurements. Conductivity of the composites was measured by High Resistance Meter, Model HR2 (AlphaLab Inc., Utah, USA). The specimens of 10 mm length were then inserted into the test fixture where the spring-loaded plunger 100006-2 applied a consistent pressure to each specimen. Five specimens were tested for each composition and the results were averaged. A TA Instruments Q800 Dynamic Mechanical Analyser in tensile mode was used for creep - recovery (static force thermomechanometry, sf-TM). Sf-TM analysis was performed by subjecting the tests samples to an applied stress of 1.2 MPa for 30 minutes followed by a recovery period of 120 minutes. The applied stress was chosen within the linear viscoelastic region of the all samples. Tests were conducted at 25 oC. The four element model of Maxwell and Kelvin-Voigt was used to interpret the creep component [6]. The springs correspond to elastic section with moduli E1 and E, while the dashpots represent the viscosity K and K2. The overall deformation of the model is given in Equation (1). = + 1 + ( ) (1) RESULTS AND DISCUSSION The mechanical properties of all LLDPE–FRP composites showed significant decrease, except for elongation at break when manufactured by fused deposition moulding (3D printing) compared with the compression moulded composites. Porosity between the neighbouring extruded filaments is most likely to be the reason behind the observed decrease in mechanical properties of the composites prepared by fused depositon modelling [5]. The introduction of carbon nanotubes did not improve the mechanical properties of the compression moulded or the 3D printed composites. This is most likely due to incorporation of MWCNT into the composites via a low viscosity thermoplastic concentrate (Plasticyl LDPE2001) [5]. The electrical properties of the composites containing MWNT, both high rubber content, C-PB4, and low rubber content, C-PB5, were tested for conductivity. All composites used for electrical testing were prepared by compression molding. Composites with high rubber content (C-PB5) demonstrated much higher conductivity compared with the low rubber content composites (PB4) [5]. As the rubber content increased, conductive channels became more effective, due to the increased concentration of carbon nanotubes within the narrowing network of LLDPE. This approach, based on selective localization of MWNT in only one of the phases, offers an innovative method of increasing/adjusting the conductivity by employing polymeric composites with selective conductor location or composites with varying amounts of each phase [7]. There was not a significant difference between the conductivities of compression molded and 3D printed samples as discussed in our recent publication [5]. The creep and recovery curves of polyolefin-rubber composite (PB1) are presented in Figure 1 (a). The creep stress varied between 1.2 MPa and 6 MPa. The total strain just before the release of the stress was measured and it is plotted against the initial loading stress as shown in Figure 1 (b). The linear relationship between stress and strain enabled to establish the maximum creep stress as 3 MPa, so that the material is tested within its linear viscoelastic region during the following creep experiments. In all creep curves, an instantaneous increase in strain is occurs due to elastic response of the material. This is followed by a viscoelastic response, which involves time-dependent molecular rearrangement. Viscous flow is observed towards the end of the load application period. Removal of the load results in rapid drop in strain response, which is equal to the initial elastic response. The recovery period involves time-dependent molecular relaxation as the polymer attempts to regain original dimensions. Since the polymer experienced viscous flow, the full recovery is not reached resulting in permanent deformation [8,9]. 100006-3 (a) (b) FIGURE 1. Creep-recovery behavior of PB1 (a) at different initial stresses; (b) max strain vs initial stress The creep and recovery curves for the composites PB1, PB3, PB4 and PB5 are shown in Figure 2 (a). The addition of (MA-g-PE) proved to enhance the bonding between the phases and the creep deformation under the same load was less in blends with (MA-g-PE). The creep and recovery curves for the composites PB4, PB5, C-PB4 and C-PB5 are shown in Figure 2 (b). Incorporation of MWCNT into the composites do not seem to introduce a positive effect on the creep behaviour. This is most likely due to the fact that the carbon nanotubes were introduced via low viscosity masterbatch, Plasticyl LDPE2001. (a) FIGURE 2. The effect of (a) (MA-g-PE) (b) MWCNT on creep-recovery behavior (b) To enable to understand the long term deformation behavior of the nanocomposites under long periods of time at ambient temperatures, a series of creep experiments were conducted at different temperatures between 30 oC and 80 oC for the PB4 and PB5. Time temperature superposition principle was applied to estimate creep performance of the filament material at longer durations. 100006-4 ACKNOWLEDGMENTS The authors acknowledge the funding provided by the Victorian Government - Department of State Development Business and Innovation under the Technology Development Voucher Program. Special thanks go to M. Allan, Rheology and Processing Centre, RMIT University for his technical assistance in experimental work. The opportunity to undertake the project was provided by Polymeric Powders Company Pty Ltd who stipulated the polyolefin and rubber materials to be utilized, as well as provided direction on the incorporation of carbon nanotubes for electrical conductivity. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. M. Vainer, Method and Device for Polymer Pulverisation Extrusion. WIPO Patent WO 2014/107758 filed on Jan 9, 2013 and issued on July 17, 2014. M. Vainer, Utilisation of Gasses for Polymeric Materials Fragmentation and Activation and Related Device. WIPO Patent WO 2014/110617 filed on Jan 21, 2013 and issued on July 24, 2014. M. Vainer, Flexible Composite Material and Method of Producing Same. WIPO PCT Application PCT/AU2015/000104 filed on Feb 26, 2015 with priority date Feb 27, 2014. P. Pötschke, S. Pegel, M. Claes, D. Bounde, Macromol Rapid Commun. 29, 244–251 (2008). F. Daver, E. Baez, R.A. Shanks and M. Brandt, “Conductive polyolefin-rubber nanocomposites with carbon nanotubes” Comp Part A: Sci. and Manuf. 80, 13-20 (2016). K. P. Menard, Dynamic Mechanical Analysis - A Practical Introduction (CRC Press, Boca Raton, 2008), pp. 40–44. P.J. Brigandi, J.M. Cogen, R.A. Pearson, Polym. Eng. Sci. 54, 1–16 (2014). A. Genovese and R.A. Shanks, Macromol. Mater. Eng. 292, 184–196 (2007). S. Spoljaric, T.K. Goh, A. Blencowe, G.G. Qiao and R.A. Shanks, Makromol. Chem. Phys. 212, pp. 1778– 1790 (2011). 100006-5