Comparative Analysis of Process Parameter Variations in DGFinFET Device Using Statistical Methods

Abstract

This paper investigates on the design and optimization of the input process parameter variations in Double Gate FinFET (DG-FinFET) device through comparisons between two different statistical methods through Taguchi and 2-k factorial design. This research focuses on the effects on threshold voltage (VTH ), leakage current (IOFF ), drive current (ION ), and the subthreshold voltage (SS) towards various parameter variations. The fabrication of the device as well as its electrical characterization are both performed using TCAD simulator, specifically ATHENA and ATLAS modules. Optimization of the process parameters is implemented and merged with the aforementioned modules. The comparisons are also conducted for the Taguchi and 2-k factorial design, statistical methods after implementation is done for both. The optimum condition for the process parameters are obtained with Polysilicon Doping Dose at Level 3 (3.7E14 atom/cm3), Polysilicon Doping Tilt at Level 3 (-17°), Source/Drain Doping Tilt at Level 1 (73°) and Threshold Voltage Doping dose at Level 2 (1.95E13 atom/cm3). The S/N ratio of Threshold Voltage, Leakage Current, Drive Current and Subthreshold Voltage values are in the predicted range of the International Technology Roadmap for Semiconductors (ITRS) 2015 prediction. Based on comparisons made, optimization approach works best and most suitable with the Taguchi method due to the consideration of noise factor used in the orthogonal array, despite the fact that both Taguchi and 2-k factorial design process is able to produce optimum solutions that are within the desired values. © Published under licence by IOP Publishing Ltd.