Machine learning-enabled textile-based graphene gas sensing with energy harvesting-assisted IoT application

Highlights: • With machine learning-assist and triboelectric-textile to power IoT, H₂ can be identified to wearable applications. • The gas sensor demonstrated 6 times higher sensing performance compared to other graphene gas sensors. • All-textiles flexible and foldable gas sensor and textile-triboelectric power source were produced. • The inkjet-printing provides the compatibility with various substrates, non-contact patterning and cost-effectiveness. Flexible gas sensing is attracting more attention with the development of machine learning and Internet of Things (IoT). Herein, we report flexible and foldable high-performance hydrogen (H₂) sensor on all textiles substrate-fabricated by inkjet–printing of reduced graphene oxide (rGO) and its application to wearable environmental sensing. The inkjet-printing process provides the advantages of the compatibility with various substrates, the capability of non-contact patterning and cost-effectiveness. The sensing mechanism is based on the catalytic effect of palladium (Pd) nanoparticles (NPs) on the wide bandgap rGO, which allows facile adsorption and desorption of the nonpolar H₂ molecules. The graphene textile gas sensor (GT-GS) demonstrates about six times higher sensing response than the graphene polyimide membrane gas sensor due to the large surface area of the textile substrate. An analysis of the temperature influence on the GT-GS shows better H₂ gas response at room temperature than at high temperature (e.g., 120 °C). In addition, with the machine learning-enabled technology and triboelectric-textile to power IoT (temperature and humidity for gas calibration), H₂ is well identified for wearable applications with a robust mechanical performance (e.g., flexibility and foldability).