Graphite/TiO2 Nanoparticle-Based Flexible Sensor for Volatile Organic Compound Detection with Machine Learning-Assisted Classification

Abstract

Flexible and low-cost chemical sensors have attracted significant attention for environmental monitoring, industrial safety, and healthcare applications. In this work, a Graphite/TiO₂ nanoparticle-based flexible sensor was fabricated on a cellulose paper substrate using a simple pencil drawing technique followed by the deposition of titanium dioxide (TiO₂) nanoparticles through a drop-casting method. The fabricated sensor was characterized using UV–Visible spectroscopy to investigate its optical properties. The sensing performance of the device was evaluated for the detection of volatile organic compounds (VOCs), including ethanol, chloroform, and ethyl acetate. The sensor exhibited distinct resistance changes upon exposure to VOC vapors, demonstrating its capability for chemical sensing. Among the tested analytes, ethanol showed the highest sensitivity of 117.1 with a response time of 10.3 s and a recovery time of 31.1 s, while chloroform and ethyl acetate exhibited sensitivities of 47.9 and 43.2, respectively. The enhanced sensing performance is attributed to the combined effect of the conductive graphite network and the high surface activity of TiO₂ nanoparticles. Furthermore, a machine learning-assisted classification framework based on the K-Nearest Neighbors (KNN) algorithm is proposed for intelligent VOC identification using sensor parameters such as sensitivity, response time, recovery time, and resistance variation. The developed Graphite/TiO₂ flexible sensor demonstrates excellent potential for low-cost, portable, and intelligent VOC monitoring applications.