This thesis presents the utilisation of giant piezoresistance of carbon nanotubes (CNTs) for high resolution pressure sensing. The nanoscale diameter of CNTs, used as sensing elements, increases the resolution of piezoresistive sensing by three orders of magnitude to that of silicon based sensors. The design of the sensor is based on sensing the strain in CNTs induced by the flow of gas and can be adapted to benefit cross-disciplinary fields like; flow and pressure sensing, microfluidics, Lab-on-chip and NEMS (nano-electromechanical systems).
CNTs were grown inside silicon micro-cavities so as to bridge the gap between two silicon substrates. The nickel catalyst coated silicon substrates act as electrodes connected to the two ends of CNTs. The CNTs grow on the nickel nanoparticles, thus self-anchoring on to the substrate. Diffusion of nickel in silicon provides low resistive NiSi contacts to CNTs. Growth of CNTs in this form have not been reported before and presents several merits including no chemical treatment or post-growth alignment of CNTs, thus keeping the process simple and robust.
CNT growth parameters; temperature, time and methane flow rate, were optimised in a custom designed chemical vapour deposition (CVD) rig, to control the CNT diameter. CNT diameter directly affects its piezoresistive coefficient, πL, and Young’s modulus, E, the factors that define piezoresistance in any material. Thus, optimised growth conditions allowed the direct tuning of piezoresistance of the sensor. Piezoresistance sensing was performed by inducing strain in CNTs with an applied differential pressure across the microcavity. Pressure loadings of as low as 0.1 atm (limited only by the gauge resolution) and a piezoresistance of as high as 16% at a pressure loading of 1 atm, were achieved. This piezoresistance is at least one order higher and the resolution is three orders higher than commercially available polysilicon and GaAs membrane based sensors.
Piezoresistance was modelled by applying Euler-Bernoulli beam theory, assimilating CNTs to rigid beams with special boundary conditions, accounting for self-anchoring to Ni islands. The resulting theory is found to be in good agreement with our experimental results and estimates the E, πL and the average radius of the CNTs. This modelling, to our knowledge, is an original attempt to modify Euler-Bernoulli beam theory with the assumed boundary conditions.
|Date of Award||19 Nov 2013|
|Supervisor||Alain Nogaret (Supervisor)|
- carbon nanotubes
- pressure sensing