Modeling elastic properties of 3D printed composites using real fibers. (15th October 2022)

• Record Type: Journal Article
• Authors: Polyzos, E.; Van Hemelrijck, D.; Pyl, L.
• Is Part Of: International journal of mechanical sciences. Volume 232(2022)
• Publication Date: 2022-10-15
• Subjects: 3D printed composites -- Fiber's contour -- Finite element modeling -- Stochastic modeling -- Effective elastic properties; Mechanical engineering -- Periodicals; Génie mécanique -- Périodiques; Mechanical engineering; Maschinenbau; Mechanik; Zeitschrift; Periodicals; 621.05
• DOI: 10.1016/j.ijmecsci.2022.107581
• Languages: English
• ISSNs: 0020-7403

Abstract

Abstract: The microstructure of 3D printed composites is inherently different than traditional composites due to the manufacturing process. The differences influence morphological characteristics such as the contour of the cross-section of the fiber and alter the macroscopic behavior of 3D printed parts. This article investigates the microstructural morphology of 3D printed nylon reinforced with continuous carbon fibers and the effect of microstructural irregularities on the macrostructural elastic response through stochastic homogenization modeling. The contour of the carbon fibers is extracted from scanning electron microscopy (SEM) micrographs available in the literature and used to generate realistic and ideal (ellipsoidal, circular) contours in a stochastic manner using a new methodology. Furthermore, a novel method is introduced to generate single- and multiple-fiber representative volume elements (RVEs) in finite element (FE) software for the approximation of the effective elastic properties. To minimize the computational effort associated with the full numerical modeling of multiple-fiber RVEs, a novel semi-analytical approach is demonstrated based on the numerical estimation of stiffness contribution tensors and the implementation of analytical effective field methods. The results of the numerical and semi-analytical models are compared with analytical models and exhibit a good agreement. Graphical abstract: Highlights: New methodology for generating stochastic fiber contours. Novel method for modeling single- and multiple-fiber Representative Volume Elements. Novel approach for semi-analytical effective field methods.