Graphene as a Platform for Novel Nanoelectronic Devices

Abstract

Graphene's superlative electrical and mechanical properties, combined with its compatibility with existing planar silicon-based technology, make it an attractive platform for novel nanoelectronic devices. The development of two such devices is reported—a nonvolatile memory element exploiting the nanoscale graphene edge and a field-effect transistor using graphene for both the conducting channel and, in oxidized form, the gate dielectric. These experiments were enabled by custom software written to fully utilize both instrument-based and computer-based data acquisition hardware and provide a simple measurement automation system. Graphene break junctions were studied and found to exhibit switching behavior in response to an electric field. This switching allows the devices to act as nonvolatile memory elements which have demonstrated thousands of writing cycles and long retention times. A model for device operation is proposed based on the formation and breaking of carbon-atom chains that bridge the junctions. Information storage was demonstrated using the concept of rank coding, in which information is stored in the relative conductance of multiple graphene switches in a memory cell. The high mobility and two dimensional nature of graphene make it an attractive material for field-effect transistors. Another ultrathin layered material—graphene's insulating analogue, graphite oxide—was studied as an alternative to bulk gate dielectric materials such as Al2O3 or HfO2. Transistors were fabricated comprising single or bilayer graphene channels, graphite oxide gate insulators, and metal top-gates. Electron transport measurements reveal minimal leakage through the graphite oxide at room temperature. Its breakdown electric field was found to be comparable to SiO2, typically 1–3 × 108 V/m, while its dielectric constant is slightly higher, κ ≈ 4.3. As nanoelectronics experiments and their associated instrumentation continue to grow in complexity the need for powerful data acquisition software has only increased. This role has traditionally been filled by semiconductor parameter analyzers or desktop computers running LabVIEW. Mezurit 2 represents a hybrid approach, providing basic virtual instruments which can be controlled in concert through a comprehensive scripting interface. Each virtual instrument's model of operation is described and an architectural overview is provided.