Research On Piezoresistivity Of Carbon Nanotube/Polymer Composites And A Piezoresistive Flexible Sensor Array

Flexible electronic devices have received growing attentions because theirexcellent flexibility allows them working with irregular or deformable objects. Flexibleforce sensors can be used for measuring of surface force distribution on a curvedstructure, which are indispensable in many fields, such as robotics, biomechanics andmedical measurements. Carbon nanotube (CNT) filled polymer composites are one kindof hopeful piezoresistive material as they combined excellent conductivity andsuper-aspect ratio of CNTs and good flexibility of polymers, which provide a new wayto improve the performance of flexible force sensors. In this dissertation, in order toenhance the piezoresistivity of CNT/polymer composites, the microscopic mechanismof the conductivity and piezoresistivity of CNT networks embedded in polymer matrixwas investigated, and an optimizing method was proposed and verified by experiments.A flexible thin-film pressure sensor array based on a CNT/polymer composite and itsscanning readout system are also fabricated and tested.A simulation model that takes account of the tunneling conduction between CNTswas established for the mechanism investigation of the CNT network. Tunnelingjunction resistances between CNTs were found as the dominant factor of the networkresistivity and the complicated CNT conducive network was hence simplified into apercolation system which is composed of junction resistors. A mechanism model of thenetwork resistivity in the microscopic aspects was introduced for the followingpiezoresistivity analyses and the key factors that dominate the network resistivity werefound as the conductivity CNT segments density and the effective tunneling junctionresistance.Simulation results showed that the piezoresistivity of a CNT network is originatedfrom the strain induced variations of the conducting CNT segment density and theequivalent tunneling junction resistance, which both can be quantitated by averagejunction gap variation (AJGV). The network piezoresistivity model was then developedbased on AJGV and a optimizing principle that includes maximizing the strain inducedAJGV and minimizing the conducting CNT segment density was proposed. Theorientation of CNTs to the normal plane of strain and a lower cross-linking density weresuggested to enhance the piezoresistivity of CNT/polymer composites. The dispersion process of CNTs in polystyrene and polydimethylsiloxane wascarried out and the composites were prepared for the piezoresistivity optimizing methodverification. The resistance variations with small tensile and compressive strains of thecomposites were measured and the influence of Poisson’s ratio on the piezoresistivitywas validated. The results showed that composites with lower CNT concentration,higher orientation degree, and lower cross-linking density have a higher piezoresistivity,and coincide well with the simulation predictions. The largest gauge factor obtained inour experiments is44, which is about one half of the largest value of silicon. Besides,the piezoresistivity was doubled by orientation of CNTs and was increased2times bydecreasing the cross-linking density.The CNT/polydimethylsiloxane composite was chosen as the sensitive material for itsexcellent flexibility, and then the electrical contact property with electrodes and thehysteresis of the piezoresistivity was measured. A flexible thin-film force sensor arraywith16×16elements was designed and fabricated. To measure the force distribution, ascanning readout circuit with crosstalk isolation and a LabVIEW program for dataprocessing were developed. Finally, the fundamental performance parameters of thesensor array are measured preliminarily and the sensor system can measure surfaceforce distribution in real-time.