Quantifying the Piezoresistive Mechanism in High-Performance Printed Graphene Strain Sensors

Abstract

: Printed strain sensors will be important in applications such as wearable devices, which monitor breathing and heart function. Such sensors need to combine high sensitivity and low resistance with other factors such as cyclability, low hysteresis, and minimal frequency/strain-rate dependence. Although nanocomposite sensors can display a high gauge factor (G), they often perform poorly in the other areas. Recently, evidence has been growing that printed, polymer-free networks of nanoparticles, such as graphene nanosheets, display very good all-round sensing performance, although the details of the sensing mechanism are poorly understood. Here, we perform a detailed characterization of the thickness dependence of piezoresistive sensors based on printed networks of graphene nanosheets. We find both conductivity and gauge factor to display percolative behavior at low network thickness but bulk-like behavior for networks above ∼100 nm thick. We use percolation theory to derive an equation for gauge factor as a function of network thickness, which well-describes the observed thickness dependence, including the divergence in gauge factor as the percolation threshold is approached. Our analysis shows that the dominant contributor to the sensor performance is not the effect of strain on internanosheet junctions but the strain-induced modification of the network structure. Finally, we find these networks display excellent cyclability, hysteresis, and frequency/strain-rate dependence as well as gauge factors as high as 350.