Ultrastretchable and Self-Healing Polyacrylic Acid Hydrogel Strain Sensor with Pt Nanoparticles-Modified MXene for Machine Learning-Assisted Gesture Recognition

Flexible strain sensors have garnered significant interest in healthcare monitoring, intelligent robots, and human–computer interaction. Nevertheless, it is difficult for conventional hydrogel strain sensors to achieve high sensitivity, a wide stretching range, and good self-healing performance simultaneously. Herein, we fabricated a poly(acrylic acid) (PAA) nanocomposite hydrogel strain sensor by incorporating Pt nanoparticles (NPs)-modified MXene and tannic acid (TA) to the hydrogel network (denoted as PAA-TA@Pt@MXene). The structure, morphology, and mechanical and sensing properties have been investigated. The optimal processing for the hydrogel strain sensor was obtained, such as TA contents and Pt NPs loading amount. The optimized sensor exhibits high sensitivity with a gauge factor (GF) of 4.62, an ultrastretchable range up to 2862%, better self-healing efficiency of 84.9%, and robust durability of 1500 cycles, indicating superb mechanical properties. This can be ascribed to the synergistic effect of TA and Pt NPs decorated with MXene. TA evidently increases the hydrogen bond density and strength between PAA and MXene, conducive to stretchability and self-healing. In addition, Pt NPs improve the charge carrier between Pt and MXene due to the tunneling effect, which is beneficial to the GF value. Moreover, the combination of excellent sensing performance and good skin adhesion enables the sensors’ powerful human motion detection capability. The PAA-TA@Pt@MXene sensors achieve a 98.3% accuracy rate in various gesture recognitions after 10 epochs using a convolutional neural network. Our work provides an effective method for significantly enhancing the performance of self-healing hydrogel strain sensors with an ultrawide working range, thereby promoting their application in flexible electronics.