Enhancing the accuracy of triboelectric sensor based on triboelectric material/electrode interface design strategy

Triboelectric nanogenerator (TENG) device is widely used in the field of ultra-biometrics because triboelectric signals with unique waveform features are generated when the different materials come into contact with TENG device surface. Nevertheless, the recognition accuracy is only improved from the perspective of designing sensor structure and optimizing working mode. To address these challenges, we developed a triboelectric material/electrode interface structure design strategy that can enhance the identification accuracy of TENG-based tactile sensor. Thus, a novel TENG device with elastic polymer-encapsulated metal material structure was fabricated. Cu-plated nonwoven is fabricated and used as electrodes, which are characterized by an unordered structure with a large number of pores formed between the fibers, greatly increasing the specific surface area. There is no orientation at the micron scale, avoiding distortion of the stress-signal feature relationship. Besides, the structure of elastic polydimethylsiloxane (PDMS)-encapsulated Cu-plated nonwoven fabric results in the stability of the output signal waveform under extreme environments, which assisted in improving the durability of the output voltage and signal waveform under extreme environments. More importantly, the separation and compression between the object and triboelectric material led to the flow of electrostatic electrons and the formation of unique output signals and self-powered power. There is a clear internal relationship between the triboelectric signal feature and the material characteristics. As expected, the accuracy of identifying different materials and palms after R-CNN model training reaches 98.3 % and 98.75 %, respectively. Finally, this work provides a reliable strategy for designing smart sensors.