Abstract

Can we control quantum interferences and many-body interactions mechanically, i.e. by pulling on a nano-system? While many idealized theoretical proposals address this question, very few have been realized experimentally. To bridge this gap with single-wall carbon nanotubes (SWCNTs), we are developing simultaneously an experimental platform and an applied theoretical model. I have nanofabricated several high quality strain-tunable suspended SWCNT transistors. I first located and characterized SWCNTs (diameters ≤ 2 nm) grown by our former group member Andrew McRae, using scanning electron microscopy (SEM) and atomic force microscopy (AFM). I patterned nanoscale bowtie-shaped gold break junctions (≈ 300 nm wide) on top of SWCNTs, using electron beam lithography (EBL). Finally, I suspended these break junctions by removing the supporting SiO2 beneath them, using a buffered oxide etch (BOE). After opening nanogaps in gold break junctions via electromigration, it will allow straining of ultra-short SWCNT channels (≈ 20 nm) with our custom-built quantum transport strain engineering (QTSE) platform. Besides the fabrication, I have also extended and modified the previous applied theory from describing strain transport behaviors in graphene to those in SWCNTs. This theoretical model considers dominant uniaxial strain effects on the band structure and all relevant experimental parameters. In quasi-metallic SWCNTs, I predicted that the uniaxial strain can widely tune conductance, leading to outstanding quantum transistors. In metallic ones, I observed a valley filter behaviour where electrons are only allowed to flow through certain valleys of the band structure. In semiconducting ones, I predicted the strong tunability of electron-hole asymmetry via uniaxial strain, which would permit us to engineer two vastly different transport behaviors into a single device.