Evaluating Measurement Bias in Flexural Modulus of Orthotropic UHMWPE Fiber/Nano-Epoxy Composites: A Finite Element Study
- 1Department of Materials Science and Nanotechnology Engineering, Faculty of Engineering and Natural Sciences, Yeditepe University, 26 Ağustos Campus, İnönü Mah., Kayışdağı Cad. 326A, Ataşehir, 34755 İstanbul, Türkiye
Abstract
Reactive graphitic nanofibers (r-GNFs) dispersed in epoxy increase the resin’s flexural modulus by approximately 20% at a 0.3 wt% optimum loading, serving as matrices for ultra-high-molecular-weight polyethylene (UHMWPE) fiber composites. This study reanalyzes published datasets, constructing a traceable orthotropic material library evaluated in three-point bending via finite element analysis. The isotropic matrix cards successfully reproduce source experimental values, showing a maximum error of 0.50% at a 16:1 span and 0.020% at 32:1. However, for the unidirectional composite, the standard 16:1 span is predicted to yield an apparent flexural modulus of only 77.6–80.7% of the true longitudinal modulus. Crucially, this loss is formulation-dependent: r-GNFs increase the longitudinal modulus considerably more than the axial shear modulus, increasing the anisotropy ratio. Consequently, a 16:1 span underreports the true modulus gain at the optimum loading by 13%—an error that drops significantly when using a 32:1 span. Furthermore, we document issues in the source dataset: a lack of common fiber volume fraction across groups invalidates direct raw modulus comparisons, and the empirical model’s optimum coordinates appear transposed. Finally, a rule-of-mixtures decomposition confirms the 0.3 wt% optimum, revealing a statistically significant residual excess stiffness exclusively at this loading.
Related
- Effect of Scrap Configuration on the Tensile and Flexural Properties of Grinding-Free Recycled CF/PA6 Composites
- Experimental and Numerical Investigation of Orthotropic CFR-PEEK and Ti-6Al-4V Osteosynthesis Plates Using a Multiscale Finite Element Modeling Approach
- Mesostructure, Toolpaths and the Finite-Element Prediction Gap in the Flexural Response of Planar and Non-Planar Fused Filament Fabrication
- Resolving the Strength–Modulus–Elasticity Tradeoff in Elastomers Using Dual Phase‐Separated Nanodomains
- Flexural Behavior of Structural Foam Concrete with Hollow Polymer Spherical Fillers Under Thermal and Mechanical Loading: Numerical and Experimental Analysis
BibTeX
@article{ogul2026evaluating,
title = {{Evaluating Measurement Bias in Flexural Modulus of Orthotropic UHMWPE Fiber/Nano-Epoxy Composites: A Finite Element Study}},
author = {Mehmet Vakkas Oğul and Halil İbrahim Yavuz and Umut Kaan Özbaş},
journal = {Polymers},
year = {2026},
month = oct,
doi = {10.3390/polym18212647},
url = {https://www.mdpi.com/2073-4360/18/21/2647},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Evaluating Measurement Bias in Flexural Modulus of Orthotropic UHMWPE Fiber/Nano-Epoxy Composites: A Finite Element Study
AU - Mehmet Vakkas Oğul
AU - Halil İbrahim Yavuz
AU - Umut Kaan Özbaş
PY - 2026
DA - 2026/10/26
JO - Polymers
DO - 10.3390/polym18212647
UR - https://www.mdpi.com/2073-4360/18/21/2647
AB - Reactive graphitic nanofibers (r-GNFs) dispersed in epoxy increase the resin’s flexural modulus by approximately 20% at a 0.3 wt% optimum loading, serving as matrices for ultra-high-molecular-weight polyethylene (UHMWPE) fiber composites. This study reanalyzes published datasets, constructing a traceable orthotropic material library evaluated in three-point bending via finite element analysis. The isotropic matrix cards successfully reproduce source experimental values, showing a maximum error of 0.50% at a 16:1 span and 0.020% at 32:1. However, for the unidirectional composite, the standard 16:1 span is predicted to yield an apparent flexural modulus of only 77.6–80.7% of the true longitudinal modulus. Crucially, this loss is formulation-dependent: r-GNFs increase the longitudinal modulus considerably more than the axial shear modulus, increasing the anisotropy ratio. Consequently, a 16:1 span underreports the true modulus gain at the optimum loading by 13%—an error that drops significantly when using a 32:1 span. Furthermore, we document issues in the source dataset: a lack of common fiber volume fraction across groups invalidates direct raw modulus comparisons, and the empirical model’s optimum coordinates appear transposed. Finally, a rule-of-mixtures decomposition confirms the 0.3 wt% optimum, revealing a statistically significant residual excess stiffness exclusively at this loading.
SN - 2073-4360
ER -