摘要
Flexible self-powered sensors based on transition metal carbides/nitrides (MXenes) are considered for intelligent human-machine interfaces (HMIs), but their use is limited by ambient oxidation and mechanical fragility during repetitive deformation. In this work, a surface-coating strategy is used to prepare PVA/MXene composite films, in which poly(vinyl alcohol) (PVA) is bonded to MXene through interfacial hydrogen bonds. The PVA layer reduces MXene’s exposure to moisture and oxygen and limits crack propagation under cyclic mechanical stress. The resulting triboelectric nanogenerator delivers a peak output voltage of 50 V with an Ecoflex counter layer and a maximum power density of 6 μW/cm2. The device charges a 2.2 μF commercial capacitor to ≈2.7 V within 60 s. A self-powered sensor array combined with a machine-learning model is used to distinguish six hand gestures; a support vector machine classifier reaches 93.33% ± 6.09% accuracy on the collected signals (45 samples, stratified 5-fold cross-validation). These results support the development of durable flexible self-powered devices and HMIs.
危险品化学品经营许可证(带存储)