A multimodal PDMS triboelectric nanogenerator sensor based on anodised aluminium oxide template preparation for object recognition

With triboelectric nanogenerators (TENGs) increasingly being used as touch sensors, their recognition accuracy is critical for the practical application of TENG devices. Here, we propose a method to prepare polydimethylsiloxane (PDMS) films with nano-array structures on the surface using an anodised aluminium oxide (AAO) template to assemble a TENG device with a high static contact angle. The method can improve the TENG output effect and enhance the accuracy of material recognition. The prepared nano-array structures (na-PDMS) are a series of cylinders with a radius of 150 nm and a height of 500 nm. The results show that the N-TENG device assembled from na-PDMS produced strain in the nano-array layer of na-PDMS at an applied stress of 0.89-19.50 kPa; when the applied stress of 19.50-134.52 kPa exceeded the tolerance range of the nano-array layer of na-PDMS, the non-nano-array structured layer of na-PDMS produced strain. Furthermore, under the test conditions of 14.95 kPa, the maximum power density of the N-TENG was 3.96 mW m-2 when the load resistance was 4 M omega. In addition, the feature signals generated when the N-TENG was in contact with the material/separated from the material were combined with the CNN deep learning technique. The following could be observed: the fabricated N-TENG device could recognise eight different materials with 97.7% certainty, which is better than the conventional S-TENG device. Overall, the method proposed in this paper provides a feasible way to improve the triboelectric properties of TENGs and sheds light on the surface structure design of triboelectric materials. With triboelectric nanogenerators (TENGs) increasingly being used as touch sensors, their recognition accuracy is critical for the practical application of TENG devices.