Logo image
Beyond Fiber Alignment: Programmable Anisotropy in 3D-Printed Thermoset Composites
Dissertation   Open access

Beyond Fiber Alignment: Programmable Anisotropy in 3D-Printed Thermoset Composites

Kaan Nuhoglu
Doctor of Philosophy (PhD), University of Miami
2026-06

Abstract

Additive Manufacturing Carbon Fiber Reinforced Polymer Composites 3D Printing Direct Ink Writing Fiber Orientation Composite Materials

This research develops a direct ink writing (DIW) framework for controlling fiber orientation and anisotropy in 3D-printed short carbon fiber-reinforced thermoset composite structures. Using an epoxy-based short carbon fiber composite ink and a customized DIW 3D printing platform, three interconnected strategies were developed: electrical network activation through oscillatory infill to enhance electrical conductivity, Bi-Modal mechanical architecture through designed combinations of aligned and randomized regions, and functionally graded anisotropy through spatial control of oscillation amplitude. The results show that maximum fiber alignment is not universally optimal. For electrical functionality, oscillatory infill kinematically disrupts shear-induced laminar flow fields, statistically increasing the frequency of inter-fiber physical contacts to lower the effective electrical percolation threshold without the volumetric over-extrusion that compromises dimensional stability. For structural performance, Bi-Modal 3D printing with controlled fiber randomization and designed randomized regions outperforms fully aligned and conventional baselines under complex loading, with gains of up to approximately 20% in failure load and approximately 17% in energy absorption. For graded structures, programmed partial randomization and continuous anisotropy control improve energy absorption by nearly 10% while maintaining comparable flexural strength. Rather than treating fiber randomization as a defect, this research uses controlled randomization as a design state for improving conductivity, damage tolerance, and energy absorption. The major outcome of this work is Bi-Slice, a free analysis-to-manufacturing slicer software package that converts finite element stress fields into machine-executable G-code for stress-aware Bi-Modal composite 3D printing.

 

pdf
kxn390S2625.16 MBDownloadView
Open Access

Metrics

3 File views/ downloads
5 Record Views

Details

Logo image