Electron-electron Interaction Driven Phenomena in Carbon Nanotube Devices

Graduation date: 2017CNTs also offer new opportunities to study new science and develop new technology enabled by their strong electron-electron (e-e) interactions. The lack of dielectric screening inherent in nanoscale structures like CNTs leads to strong e-e interactions, which produce unique physical phenomena. In this thesis, we study the effects of strong e-e interactions in CNTs through the experimental system of suspended CNT devices fabricated with two gate electrodes. The work presented here includes the development of a theoretical framework to understand of the behavior of these ‘split-gate’ devices. Using this framework, we are able observe signatures of the strong e-e interactions in CNTs.\ud The purpose of using split-gates (two gate electrodes) is to electrostatically dope the CNT into a pn junction. These suspended CNT pn junctions have been used by several groups to investigate the optoelectronic properties of CNTs. However, the device transport models proposed by previous authors have been unable to explain the disparities in experimental observations. In particular, different authors have seen different responses of the source-drain current vs source-drain bias (Isd-Vsd characteristic) of similar devices. To explore the reason for this variability, we investigate the Isd-Vsd\ud characteristic while varying the metal contact work function. The results allow us to develop a model that explains the variation in the literature in terms of variations in the metal work functions and/or CNT diameter. The device is modeled with a pn junction diode in the center of the CNT and Schottky diodes at the contacts. We are also able to use this model and temperature dependent measurements to extract the n-type and p-type Schottky barrier heights.\ud Carbon nanotubes (CNTs) are a promising material for high-performance electronics beyond silicon. Unlike silicon, the nature of the transport band gap in CNTs is not fully understood. The transport gap in CNTs is predicted to be strongly driven by e-e interactions and correlations, even at room temperature. The effects a dielectric material, like a SiO2 substrate, on the transport gap is important for implementation of this technology. Here, we use dielectric liquids to screen e-e interactions in individual suspended ultra-clean CNTs. Using multiple techniques, the transport gap is measured as dielectric screening is increased. Changing the dielectric environment from air to isopropanol, we observe a 25% reduction in the transport gap of semiconducting CNTs, and a 32% reduction in the band gap of narrow-gap CNTs. Additional measurements are reported in dielectric oils. Our results elucidate the nature of the transport gap in CNTs, and show that dielectric environment offers a mechanism for significant control over the transport band gap.\ud CNTs are candidates for next-generation photovoltaic technology, because they have the potential to break the Shockley-Queisser limit. Because of the strong e-e interactions, photogenerated carriers in CNTs can undergo carrier multiplication, where more than one electron-hole pair is created per absorbed photon. The photocurrent quantum yield\ud (PCQY), defined as the number of electron-hole pairs extracted from a device per absorbed photon should therefore be able to exceed unity. However, a previous measurement on a similar split-gate device only achieved PCQY of 1-5%. To address this discrepancy we study photocurrent generation in individual suspended carbon nanotube pn junctions using spectrally resolved scanning photocurrent microscopy. Spatial maps of the photocurrent allow us to determine the length of the p–n junction intrinsic region, as well as the role of the n-type Schottky barrier. We show that reverse-bias operation eliminates complications caused by the n-type Schottky barrier and increases the length of the intrinsic region. We develop a method of determining the PCQY that takes into account the beam waist, length of the intrinsic region, CNT diameter, resonant absorption cross section of CNTs, and intensity enhancement from reflection off the substrate. We find that the room temperature PCQY is approximately 30% when exciting the carbon nanotube at the S44 and S55 excitonic transitions. The PCQY value is an order of magnitude larger than previous estimates