Synthesis and characterization of EPS reinforced modified castor oil-based epoxy biocomposite

Aydoğmuş E., Dağ M., Yalçın Z. G., ARSLANOĞLU H.

Journal of Building Engineering, vol.47, 2022 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 47
  • Publication Date: 2022
  • Doi Number: 10.1016/j.jobe.2021.103897
  • Journal Name: Journal of Building Engineering
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Compendex, INSPEC
  • Keywords: Density, Hardness, Modeling, Thermal conductivity, Thermal decomposition
  • Çanakkale Onsekiz Mart University Affiliated: Yes


© 2021In this research, both modified castor oil-based epoxy is synthesized and waste expanded polystyrene (EPS) is used as a filler in the newly improved biocomposite. The experimental work plan is optimized with response surface methodology (RSM) and the thermophysical properties of the biocomposites have been also evaluated with artificial neural networks (ANN). Chemical characterization of the synthesized modified castor oil (MCO) based biocomposite has been done by Fourier transform infrared spectroscopy (FTIR). Thermogravimetric analysis (TGA) and scanning electron microscopy (SEM) of the obtained biocomposite have been determined. According to the results, the activation energy of the biocomposite synthesized with modified castor oil is up to 21% higher than the pure epoxy composite. The use of MCO in the biocomposite is also reduced the epoxy components (petrochemicals) by up to 13 wt%. Besides, the recycling of waste EPS in biocomposite has been reduced the production cost up to 9% and the density of the synthesized biocomposite up to 15%. Also, EPS reinforcement reduces the thermal conductivity coefficient up to approximately 17%, while MCO reinforcement decreases the Shore D hardness and increases the workability of the biocomposite. Moreover, a new model equation with hyperbolic function has been improved to examine the thermal decomposition behavior of the biocomposite. Maximum correlation coefficient and minimum error values have been analyzed statistically with Flynn-Wall-Ozawa, Kissinger, and Coats-Redfern models.