Filament-based material extrusion (MatEx) additive manufacturing has garnered huge interest in both academic and industrial communities. Moreover, there is an increasing need to expand the material catalog for MatEx to produce end use parts for a wide variety of functional applications. Current approaches towards MatEx of semicrystalline thermoplastics are in their nascent stage with fiber reinforcements being one of the most common techniques. MatEx of commodity semicrystalline thermoplastics has been investigated but most of the current methods are extremely material and machine specific.
The goal of this dissertation is to enable MatEx of semicrystalline polymers with mechanical properties approaching that of injection molded parts. Tailored molecular architectures of blends that can control the crystallization kinetics from the melt state are investigated. By modifying the crystallization time window, the time during which chain diffusion can occur across the deposited layers is prolonged, which allows for a stronger bond between layers. Such differences in the crystallization process impact the z-axis adhesion and residual stress state, which directly affect mechanical properties and warpage in the printed parts. The impact of blend composition on polymer chain diffusion, crystallization profiles, and print properties resulting from the repeated non-uniform thermal history in filament based MatEx is studied. The melt flow behaviour is characterized using rheology and its effect on the interlayer adhesion of printed parts and print precision is explored. The extent of polymer chain re-entanglement post deposition on the printer bed is quantitatively determined using interrupted shear rheology protocols. Tensile bars are printed and mechanically characterized to analyze the tensile performance of the printed parts. Correlating the rheological findings with the mechanical performance of the printed parts provides valuable insights into the complex interlayer welding process during MatEx and is critical to improving existing machine designs and feedstocks in order to achieve printed parts with properties approaching their injection molded counterparts. The results will be essential in identifying optimal processing conditions to maximize material specific printed part performance as well as highlight the associated limitations to enable MatEx of next generation materials. / Doctor of Philosophy / Compared to traditional subtractive manufacturing techniques, additive manufacturing (AM) has the potential to transform modern manufacturing capabilities due to its unique advantages including design flexibility, mass customization, energy efficiency, and economic viability. The filament-based material extrusion (MatEx), also referred to as fused filament fabrication (FFF), employing thermoplastic polymers (and composites) has emerged as one of the most common AM modality for industrial adoption due to its operational simplicity. However, the widespread application of MatEx has been limited due to the lack of compatible materials, anisotropic mechanical properties, and lack of quality assurance. Most of the research on FFF has been performed on amorphous polymers with almost negligible levels of crystalline content such as polylactic acid (PLA) and acrylonitrile-butadiene-styrene (ABS). Semicrystalline polymers are an attractive choice for FFF feedstocks compared to the amorphous ones due to their improved thermal resistance, toughness, and deformability. However, processing semicrystalline polymers using FFF is challenging due to the volumetric shrinkage encountered during crystallization from the melt state. This results in the buildup of significant levels of residual stresses at temperatures lower than the crystallization temperature of the polymer resulting in warpage of the printed parts.
The research presented in this dissertation aims to address the aforementioned challenges by characterizing semicrystalline polymer feedstocks under conditions representative of the multiphysics encountered during a typical FFF process. Several strategies to limit shrinkage and warpage are discussed that involve tuning the thermal profile and crystallization kinetics during printing. The former is achieved by addition of thermally conductive carbon fiber reinforcements while the latter is realized by blending amorphous resins or low crystallinity polymers to the semicrystalline polymer matrix. The fibers results in a more homogenous temperature distribution during printing while the incorporation of the resins modify the rate of crystallization; both of which play a pivotal role in reducing the residual stress build-up and hence minimizing the warpage during printing. The printability of the materials is investigated based on the shear- and temperature dependent viscous response of the polymers. The printed parts with fiber reinforcements exhibit high levels of mechanical anisotropy compared to the blends with the resins, likely due to differences in polymer chain mobility at the interface. The tensile properties of the printed polymer blends are slightly inferior to those obtained using traditional manufacturing techniques; however, properties close to 90-95% of injection molded properties are recovered through a simple post-processing thermal annealing step. The obtained results will assist in optimizing the processing parameters and feedstock formulation in order to consistently produce printed parts with minimal defects and tailored mechanical properties for functional applications.
Identifer | oai:union.ndltd.org:VTETD/oai:vtechworks.lib.vt.edu:10919/111785 |
Date | 09 September 2022 |
Creators | Das, Arit |
Contributors | Chemical Engineering, Bortner, Michael J., Baird, Donald G., Williams, Christopher Bryant, Martin, Stephen Michael, McKnight, Steven Hugh |
Publisher | Virginia Tech |
Source Sets | Virginia Tech Theses and Dissertation |
Language | English |
Detected Language | English |
Type | Dissertation |
Format | ETD, application/pdf, application/octet-stream |
Rights | In Copyright, http://rightsstatements.org/vocab/InC/1.0/ |
Page generated in 0.0029 seconds