Processing and mechanical characterization of short carbon fiber-reinforced epoxy composites for material extrusion additive manufacturing

• Published in: Composites. Part B, Engineering Engineering, 2021-10, Vol.223 (C), p.109122, Article 109122
• Main Authors: Hmeidat, Nadim S., Elkins, Daniel S., Peter, Hutchison R., Kumar, Vipin, Compton, Brett G.
• Format: Article
• Language: English
• Subjects: 3D printing; Additive manufacturing; Direct ink writing; Extrusion; Fiber length; Short-fiber composites; Thermosets
• ISSN: 1359-8368; 1879-1069
• DOI: 10.1016/j.compositesb.2021.109122

Fiber-reinforced polymer composites have been extensively utilized in recent years as feedstock materials for material extrusion additive manufacturing (AM) processes to improve strength, stiffness, and functionality of printed parts over unfilled printed polymers. However, the widespread adoption of AM of fiber-reinforced polymer composites requires a deeper understanding of the process-structure-property relationships in printed components, and such relationships are not well understood yet. Fiber length is critically important to the mechanical performance of short fiber composites, but very few studies to-date have focused on how the fiber length distribution (FLD) evolves during processing of composite feedstocks and how this evolution affects printing behavior and mechanical properties in 3D-printed composites. In this work, FLD is measured for carbon fiber reinforced epoxy composites over a wide range of ink compositions and shear mixing times, and the distributions are fit with a Weibull-type distribution function. The effects of FLD on the tradeoff between ink processability, ink rheology, printing behavior and mechanical properties are investigated. Furthermore, the effects of printing parameters (nozzle size and print speed) on mechanical anisotropy and fiber orientation distribution (FOD) in printed composites are explored. Mechanical properties of printed composites are characterized via 3 pt-flexural testing, and microstructure is investigated using optical and scanning electron microscopy (SEM), and x-ray computed tomography. Finally, the fitted Weibull parameters are fed into a composite model that incorporates FLD and FOD, and model predictions are found to be in excellent agreement with experimental observations.