Layer-level constitutive material model for representing non-linear stress-strain relationships in continuous carbon fiber-reinforced 3D-printed composites

3D-printed thermoplastic composites reinforced with continuous fibers exhibit nonlinear mechanical responses mostly due to the viscoelastic matrix and the presence of voids. We present a layer-level constitutive material modeling framework to address this behavior. Unidirectional and multidirectiona...

Full description

Saved in:
Bibliographic Details
Main Authors: Csenge Tóth, Blanka Ilinyi, László Kovács
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Polymer Testing
Online Access:http://www.sciencedirect.com/science/article/pii/S0142941825001084
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:3D-printed thermoplastic composites reinforced with continuous fibers exhibit nonlinear mechanical responses mostly due to the viscoelastic matrix and the presence of voids. We present a layer-level constitutive material modeling framework to address this behavior. Unidirectional and multidirectional composites were prepared using Composite Filament Co-extrusion technology, then, the engineering constants required to describe the transversely isotropic materials were determined with tensile tests. We chose carbon fiber reinforcement and polyamide matrix, and the fiber volume fraction was 9.5–13.6 %. We used the Bogetti relationship based on the explicit Ramberg-Osgood equation to describe the stress-strain curves above the yield criterion. We applied the Hill model to rescale the principal stress-strain curves to the shear curve with good accuracy. 4-point bending simulations were prepared to evaluate the application boundaries, which showed that the model is applicable for thinner plate-like structures (with a thickness limit of 3 mm). Lastly, microtomography was employed to analyze the microstructure, quantify void content, and identify 3D-printing defects. A significant void content was found in the multidirectional specimens (19 %), which may account for the scatter in the test results and modeling discrepancies. Overall, we demonstrate the suitability of the proposed modeling method for describing nonlinear behavior, and it contributes to more accurate finite element models for 3D-printed thermoplastic composites.
ISSN:1873-2348