Quantum Mechanical Simulations of the Impact of Surface Roughness on Nanowire TFET performance

In this work, the impact of the surface roughness (SR) on the variability in p-type InAs nanowire Tunnel FET (TFET) has been investigated. Using the Non-Equilibrium Green's Function (NEGF) module implemented in the University of Glasgow quantum transport simulation tool, called NESS, we have simulated a statistical ensemble of 200 TFETs with unique SR profiles. The SR in each device is defined by the characteristic values of the SR root mean square amplitude (RMS) and correlation length. Our results show that the larger the RMS, the stronger the variability. We find that the SR-induced variability is reduced in InAs-Si heterostructure TFETs when comparing with their homogenous InAs counterpart. The impacts of both metal grain granularity and random discrete dopants on InAs TFETs are also studied. Our finding suggests that SR is the weakest source of statistical variability.