Electrical and thermal transport in nanotube based thin film transistors

Satish Kumar, Purdue University

Abstract

Thin-film transistors (TFTs) based on networks of carbon nanotubes (CNTs) or silicon nanowires (NW) promise improved performance and novel applications in microelectronics and macroelectronics. Network transistors suggest the possibility of low voltage, highly reliable, high-speed (>GHz) flexible plastic electronics with potential applications in displays, e-paper, e-clothing, biological and chemical sensing, conformal radar, and others. Despite many promising experimental demonstrations, a number of puzzling technical difficulties have stymied the systematic development (and eventual commercial adoption) of the technology. The statistical, electrical, thermal and mechanical properties of such a network transistor are not presently understood; this requires detailed theoretical and numerical analyses in order to interpret experimental observations. The goal of the present work is develop a computational electro-thermal model to analyze the transport properties of nanocomposites composed of isotropic 2D ensembles of nanotubes in a substrate for use as the channel region of TFTs. The visually complex nanotube-network transistor is studied by representing it as a simple, two-dimensional, interpenetrating percolating network of metallic and semiconducting nanosticks. A model based on percolation theory, drift-diffusion and Fourier-conduction equations is developed to predict electrical/thermal characteristics for different parameters such as channel length, tube length, network density, tube-tube contact-conductance, tube-substrate contact-conductance and substrate-tube conductivity ratio (kS/kt). The developed computational model is used to compute the conductance properties of a CNT network. The numerical results are in good agreement with analytical results for short channel transistors with low tube densities, and with experimental measurements for longer channels at higher densities, providing broad validation of the model. Thermal transport in finite nanocomposites provides insight into the dominant transport mechanisms. For low values of kS/kt, percolating conduction in the network is seen to dominate over a wide range of tube-tube and tube-substrate contact parameters; as kS/kt increases, thermal transport through the substrate begins to dominate. An effective medium theory based analytical model is presented to compute the effective thermal conductivity of 2D composite. Theoretical results depart significantly from numerical predictions for higher volume fractions because of tube-tube interaction. Finally, the contact resistance between tubes and tube-tube coupling is characterized through classical molecular dynamics (MD) and wavelet methods.

Degree

Ph.D.

Advisors

Murthy, Purdue University.

Subject Area

Mechanical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS