The rapid breakthroughs in the Internet of Things and artificial intelligence have driven the demand for portable, self-powered, and flexible sensor devices. Hydrogels with high conductivity, mechanical adjustability, environmental adaptability, and biocompatibility are an ingenious approach for developing flexible sensors using triboelectric nanogenerators (TENGs). The application of TENGs is limited by the scarcity of suitable biological materials and the demand for high-conductivity fillers (such as two-dimensional materials), which trade off in terms of transparency, output, and sensing. We have studied a unique, highly transparent, highly stretchable, and high-output biomimetic stachyose/PVA hydrogel-based triboelectric nanogenerator (S-TENG) to address this issue. Due to its abundant dynamic hydrogen bonds, the cost-effective biomimetic stachyose is added to polyvinyl alcohol (PVA) to enhance the cross-linking and crystalline domains of the hydrogel. These structural improvements make the mechanical strength of S-hydrogel 2-5 times that of two-dimensional, biological, and transparent material-based TENGs, and the electrical output is 3-8 times, while maintaining transparency. The S-hydrogel can be recovered through water-assisted dissolution and re-gelation, maintaining its voltage output. The improved S-TENG is a self-powered sensor suitable for various human movements, with extremely high sensitivity and a response time of 13 milliseconds. The XGBoost method achieved the highest classification accuracy of 95.29% among 11 machine learning models, demonstrating the potential of self-powered sensors in many applications. This research was published in Advanced Materials under the title "High-Performance Transparent, Deformable, and Recoverable Biomimetic Stevia-PVA Hydrogel Triboelectric Nanogenerator with Machine Learning-Assisted Motion Recognition".
References:
DOI: 10.1002/adma.73030