Graphene has the highest mobility of any material at room temperature; this property has attracted a great deal of interest for applications in high frequency electronics, specifically transistors and diodes. To date, there has been little success using graphene for these purposes because it lacks the bandgap necessary to create an efficient device. This work aims to approach this problem from a different angle; using device architecture that potentially does not need a bandgap. This could allow graphene's excellent electrical properties to be exploited fully. The first example of this is the ballistic rectifier; a device that exploits the long mean free path of two dimensional electron gasses so that carriers can be treated like "billiard balls". Here we demonstrate two different four-terminal ballistic rectifiers that redirect carriers from the two input leads to one of the two output leads; the effect of this is to rectify an AC signal into a DC signal. An extremely high voltage responsivity of 23,000 V/W and a very low noise equivalent power of 0.64 pW/Hz1/2 are achieved from a low-frequency AC signal at room temperature. This same device has been tested at 220GHz and showed no signs of a cut-off frequency.Another rectifier tested here is the self-switching diode, a device that uses two side gates attached to its own source to locally gate its own conducting channel. This architecture demonstrates a modest peak responsivity of 690 V/W, a result of graphene's missing bandgap. A side-gated transistor with a modest on/off ratio of ~2.33 is also fabricated in order to better understand the limited capabilities of the graphene self-switching diode.Part of the novelty of this work is the introduction of a modified stamp transfer technique that allowed more flexibility creating hetero-structures. A dry etching recipe for hetero-structures is introduced that does not damage soft masks allowing for a new type of ultra-clean 1D contact. This new contact demonstrates considerably better contact resistance and reliability than previous generations; important for any high frequency application.
|Date of Award||31 Dec 2016|
- The University of Manchester
|Supervisor||Ernest Hill (Supervisor) & Aimin Song (Supervisor)|