Modeling Carbon Nanotube Field Effect Transistors with Fixed and Suspended Nanotube Gates

Licentiate thesis, 2008

Carbon nanotubes (CNT) exhibit a range of properties that make them well suited for nanoelectronic and nanoelectromechanical devices. One-dimensional field effect transistors based on single-walled CNTs have demonstrated excellent electrical characteristics and are competitive with silicon-based solutions, and oscillators based on suspended CNTs have been shown to work at gigahertz frequencies with Q-factors exceeding 10$^3$. In this thesis, I present a theoretical model describing the operation of a carbon nanotube field-effect transistor (CNTFET) with a static or mechanically active CNT gate. Theoretical modeling gives a remarkably good agreement with experimental measurements on CNT-gated CNTFETs, fabricated and characterised in the Atomic Physics group at University of Gothenburg, and provides an explanation for the steep sub-threshold slope and the short gate delay of the devices. With the help of the model it is demonstrated that a substantial improvement in gate delay time can be achieved by reducing the thickness of the gate dielectric. Furthermore, I show that utilizing the mechanical degree of freedom of a suspended gate CNT may lead to a sub-threshold slope smaller than the thermal 60 mV/decade limit. I present two designs of suspended CNTgated CNTFET, with doubly clamped or cantileved CNT gate. In the first design, the sub-threshold slope reaches 32 mV/decade, and in the second as low as 15 mV/decade at room temperature. In the presented CNTFET designs, the instantaneous deflection of suspended CNT is mapped by transistor drain current. I show that the sensitivity of the CNTFETs towards the motion of suspended CNT surpasses that of the conventional nanoscale displacement detection methods.