Variation in characteristics of graphene nanoribbon field-effect transistors caused by edge disorder: Computational simulation of atomistic device

Abstract

We theoretically investigated the variations in the characteristics of graphene-nanoribbon-based field-effect transistors (GNR FETs) using the nonequilibrium Green's function method. In this study, the drain current (I d ) was calculated as a function of gate voltage (V g ) for GNR FETs with various edge disorder concentrations (P). From the obtained I d -V g curves, we estimated the device characteristics. We found that the variations of these device characteristics became larger with increasing P, as evidenced by a dramatic change in the shapes of their histograms. Furthermore, we clarified that these variations were caused by Anderson localization originating from the edge disorder.

Materials Science Challenges to Graphene Nanoribbon Electronics

“…Unless a scalable lithography technique with much improved resolution is developed, it is not clear if lithography will ever yield sub-10 nm wide GNRs with smooth edges, needed to harness superior electronic properties. There have been a number of promising efforts to pattern sub-10 nm features: notably, directed self-assembly (DSA) of block copolymers, extreme ultraviolet (EUV) lithography, and self-aligned quadruple patterning (SAQP). , While most experimental and simulation studies agree that edge roughness is detrimental to the performance of GNR FETs, ,, other works have concluded that a small amount of edge roughness is acceptable and will not significantly degrade the I on / I off or I on . This dichotomy is further evidenced by the demonstration of superior FET performance using GNRs grown heteroepitaxially in h-BN trenches that may possess mixed edge character .…”