Plasmonic Bridge Sensor Enabled by Carbon Nanotubes and Au-Ag Nano-Rambutan for Multifunctional Detection of Biomechanics and Bio/Chemical Molecules

Wearable, noninvasive, and simultaneous sensing of subtle strains and eccrine molecules on human body is essential for future health monitoring and personalized medicine. However, there is a huge chasm between biomechanics and bio/chemical molecule detections. Here, a wearable plasmonic bridge sensor with multiple abilities to monitor subtle strains and molecules is developed. Hollow Au-Ag nano-rambutans and carbon nanotubes (CNTs) are adsorbed in the nonwoven fabrics (NWFs) conjointly, where the gap between the conducting network of CNTs is bridged by the Au-Ag nano-rambutans during the subtle strain sensing, and the detection sensitivity for stress is improved at least 1 order of magnitude compared to that with the only CNTs. In order to acquire the accurate human action recognition, a machine learning algorithm (support vector machines) based on output biomechanics data is designed. The average accuracy of our plasmonic bridge sensor reaches 89.0% for human action recognition. Moreover, due to the hollow structure and high nanoroughness, the single Au-Ag nano-rambutan particle has strong localized surface plasmon resonance effect and high surface-enhanced Raman scattering (SERS) activity. Based on their unique SERS spectra introduced by the hollow Au-Ag nano-rambutan adsorbed in the NWFs, noninvasive extraction and fingerprint recognition of bio/chemical molecules could be realized during the wearable sensing. In sum, the NWFs/CNTs/Au-Ag sensor bridges the barrier between the bodily strain detection and molecule recognition during the wearable sensing. Such integrated and multifunctional sensing strategy for universal biomechanics and bio/chemical molecules means to assess human health to be of importance.

Automated summary

Wearable plasmonic bridge sensor with multiple capabilities for monitoring subtle strains and molecules is developed. The sensor uses hollow Au-Ag nano-rambutans and carbon nanotubes (CNTs) adsorbed in nonwoven fabrics (NWFs) to bridge the gap between CNTs for strain sensing and improve stress detection sensitivity. A machine learning algorithm based on biomechanics data is used for accurate human action recognition.