RESUMO
The variability induced by the work-function variation (WFV) in p-type ultra-scaled nanowire tunnel FET (TFET) has been studied by using the Non-Equilibrium Green's Function module implemented in University of Glasgow quantum transport simulator called NESS. To provide a thorough insight into the influence of WFV, we have simulated 250 atomistically different nanowire TFETs and the obtained results are compared to nanowire MOSFETs first. Our statistical simulations reveal that the threshold voltage (V th) variations of MOSFETs and TFETs are comparable, whereas the on-current (I on) and off-current (I off) variations of TFETs are smaller and higher, respectively in comparison to the MOSFET. Based on the results of the simulations, we have provided a physical insight into the variations of the I on and I off currents. Then, we compared the nanowire and Fin TFETs structures with different oxide thickness in terms of the WFV-induced variability. The results show that WFV has a strongest impact on the I off, and moderate effect on the I on and V th in nanowire TFET with smaller oxide thickness. Lastly, it is found that compared with the random discrete dopants, WFV is a relatively weaker variability source in ultra-scaled nanowire TFETs, especially from the point of view of I on variation.