Highly Stable and Ultra-Fast Vibration-Responsive Flexible Iontronic Sensors for Accurate Acoustic Signal Recognition.

Yan Wang,Weiqiang Liao,Xikai Yang,Kexin Wang,Shengpeng Yuan,Dan Liu,Cheng Liu,Shiman Yang,Li Wang

DOI: https://doi.org/10.1039/d4nr03370a

IF: 6.7

2024-01-01

Nanoscale

Abstract:Wearable verbal language servers function as sophisticated and effective tools for fostering intelligent interactions between humans and machines. In the realm of collecting acoustic vibration signals, flexible iontronic pressure sensors have demonstrated their efficacy by incorporating microstructures into the functional layer, resulting in heightened pressure sensitivity. However, the substantial viscosity of the integrated iontronic materials or the lack of bonding at the heterogeneous interface emerges as a significant hindrance to capacitance recovery, leading to sluggish response speeds and mechanical instability. Here, we address the issue by introducing hydrogen bonding between naturally microstructured protein micro-fibers and hydrophilic ionic hydrogel into the dielectric layer. Due to the good resilience of protein micro-fibers and the enahnced interfacial bonding, this flexible vibration sensor demonstrates outstanding performance characteristics, featuring exceptional signal stability, a high-pressure resolution of 522 pF kPa-1, an ultra-fast response time of 0.6 ms, and a relaxation time of 0.6 ms, with a limit of detection (LOD) of 0.12 Pa, making it well-suited for acoustic vibration acquisition. By using a one-dimensional convolutional neural network (1D-CNN) deep learning to process and recognize collected acoustic signals, our sensor achieved an impressive accuracy of 98.2%. These wearable vibration sensors exemplify promising versatile applications in biometric authentication, personalized services, and human-computer interaction.