Triboelectric interface using waste-derived jamun seed powder and EVA composites for self-powered handwriting recognition

Abstract

The use of waste-based biomaterials for energy harvesting is a new frontier in triboelectric nanogenerator (TENG) research and self-powered sensing systems. In this study, we demonstrate a triboelectric interface based on jamun seed powder embedded in an ethylene-vinyl acetate (EVA) polymer matrix for handwriting recognition. Jamun seed waste contains cellulose, lignin, polyphenols, and oxygenated functional groups, which contribute to the accumulation of surface charge during contact electrification. The optimal peak-to-peak output voltage and current of the EVA-JB composite at 15 wt% filler (EVA-JB15 abbreviated) loading was 260 V and 1500 nA, respectively. XRD- and FTIR-based structural evaluation showed that the filler was successfully incorporated into the EVA matrix without disturbing the basic semicrystalline polymer structure. The composite has an increased dielectric constant (∼4.55 at 10 kHz), suggesting higher charge storage capacity and interfacial polarisation than the pure EVA. The produced TENG exhibited a maximum power of 78.125 μW at the 200 MΩ external load resistance. A convolutional neural network (CNN) model was trained using TENG-derived handwriting signals from five individuals and obtained an overall classification accuracy of 99.18%. This study opens a possible path towards triboelectric interfaces for advanced self-powered sensing.