ISSN print edition: 0366-6352 ISSN electronic edition: 1336-9075 Registr. No.: MK SR 9/7 Published monthly

Experimental and ANN-based mechanical analysis of chemically modified carbon fiber-reinforced polymer composite

Sana Kainat, Shahzad Maqsood Khan, Nafisa Gull, Saba Zia, Ayesha Sharif, Ahmad Faraz Khan, Sabaat Haroon, and Asma Saeed

Institute of Polymer and Textile Engineering, University of the Punjab, Lahore, Pakistan

E-mail: sanakainat2k17@gmail.com

Received: 22 April 2025  Accepted: 26 August 2025

Abstract:

Carbon fiber exhibits weak interfacial adhesion with the matrix due to its nonpolar nature, which subsequently limits the mechanical strength of the resulting composites. This research examines the chemical surface modification via oxidation, reduction, and silane functionalization to enhance the fiber-matrix adhesion in carbon fiber-reinforced polymer composites. Carbon fiber was treated using HNO₃ and H₂SO₄:HNO₃ mixture, followed by reduction and silane functionalization with tetraethyl orthosilicate (TEOS). The presence of hydroxyl (–OH) and silanol (–SiOH) groups on TEOS-grafted fiber was confirmed by Fourier transform infrared spectroscopy. This modified fiber, along with epoxy matrix, was used to fabricate the composite using hand lay-up along with compression molding. The resultant composite was investigated for tensile strength, Young’s modulus, and impact strength. The CFRC-4 composite (sample modified with HNO₃ and TEOS) showed highest tensile strength and stiffness, depicting improved interfacial adhesion. Moreover, based on experimental data, the artificial neural network (ANN) model was developed to predict the mechanical performance. At a specific hidden layer size, the optimal performance of stress and strain was precisely predicted by ANN. The experimental validation along with data-driven modeling was combined in this hybrid methodology, providing deeper knowledge of the relation between surface chemistry and composite performance. This approach enables the faster prediction of material characteristics for cutting-edge engineering applications and future research work.

Graphical abstract

Keywords: Chemical modification; Artificial neural network (ANN); Machine learning (ML); Carbon fiber-reinforced polymer composite; Silanol grafting

Full paper is available at www.springerlink.com.

DOI: 10.1007/s11696-025-04353-5

Chemical Papers 79 (12) 8815–8828 (2025)