Optimization of Mechanical Properties for Polyoxymethylene/Glass Fiber/Polytetrafluoroethylene Composites Using Response Surface Methodology

• Published in: Polymers Vol. 10; no. 3; p. 338
• Main Authors: Kunnan Singh, Jasbir Singh, Ching, Yern Chee, Abdullah, Luqman Chuah, Ching, Kuan Yong, Razali, Shaifulazuar, Gan, Seng Neon
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 20.03.2018; MDPI
• Subjects: Glass fibers; Mechanical properties; Naphthalene; optimization; Polymer matrix composites; polyoxymethylene; Polytetrafluoroethylene; Response surface methodology; RSM; Stiffness; Surface energy; surface etch; Tetrahydrofuran; Variance analysis; optimization; surface etch; RSM; polyoxymethylene; polytetrafluoroethylene
• ISSN: 2073-4360; 2073-4360
• DOI: 10.3390/polym10030338

This paper investigated the effects of polytetrafluoroethylene (PTFE) micro-particles on mechanical properties of polyoxymethylene (POM) composites. Since PTFE is immiscible with most polymers, the surface was etched using sodium naphthalene salt in tetrahydrofuran to increase its surface energy. The effects of two variables, namely PTFE content and PTFE etch time, on the mechanical properties of the composite were studied. Experiments were designed in accordance to response surface methodology (RSM) using central composite design (CCD). Samples were prepared with different compositions of PTFE (1.7, 4.0, 9.5, 15.0, or 17.3 wt %) at different PTFE etch times (2.9, 5.0, 10.0, 15.0, or 17.1 min). Four mechanical properties of the POM/GF/PTFE composites, that is, strength, stiffness, toughness, and hardness, were characterized as a function of two studied variables. The dependency of these mechanical properties on the PTFE etch conditions was analyzed using analysis of variance (ANOVA). Overall desirability, D global index, was computed based on the combination of these mechanical properties for POM/GF/PTFE composites. The D global index was found to be 87.5%, when PTFE content and PTFE etch time were 6.5% and 10 min, respectively. Good correlation between experimental and RSM models was obtained using normal probability plots.