Optimization of process parameter variation in double-gate FinFET model using various statistical methods
Double-gate FinFET is identified as a prospect in fulfilling the demands required in replacing the current conventional planar MOSFETs due to several advantages. Specifically in its scalability, reduced leakage current, high drive current, with steep subthreshold swing, subsequently improving the IO...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English English |
Published: |
2022
|
Subjects: | |
Online Access: | http://eprints.utem.edu.my/id/eprint/26899/1/Optimization%20of%20process%20parameter%20variation%20in%20double-gate%20FinFET%20model%20using%20various%20statistical%20methods.pdf http://eprints.utem.edu.my/id/eprint/26899/2/Optimization%20of%20process%20parameter%20variation%20in%20double-gate%20FinFET%20model%20using%20various%20statistical%20methods.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Double-gate FinFET is identified as a prospect in fulfilling the demands required in replacing the current conventional planar MOSFETs due to several advantages. Specifically in its scalability, reduced leakage current, high drive current, with steep subthreshold swing, subsequently improving the ION/IOFF ratio, thus reducing the power consumption of the device. The susceptibility towards the electrical performance of the device is exposed due to the process parameter variations from the device miniaturization. This research work is aimed towards optimizing the process parameter variation towards the device characteristics with several appropriate statistical methods used. Taguchi statistical method, the Taguchi-based Grey Relational Analysis (GRA), the 2k-factorial method, and the Response surface method-central composite design (RSM-CCD) have all been utilized to analyze the performance of the device. ATHENA module of Silvaco TCAD is utilized in this simulation-based fabrication The threshold voltage (VTH), drive current (ION), leakage current (IOFF) and subthreshold swing (SS) ramifications towards the adjustment of six process parameter that include polysilicon doping dose, polysilicon doping tilt, Source/Drain doping dose, Source/Drain doping tilt, VTH doping dose and VTH doping tilt is studied. The effect of the said process parameter variations were analysed with the utilization of L25 orthogonal array (OA), main effects, signal-to noise ratio (SNR) and analysis of variance (ANOVA) for the Taguchi statistical method, the Taguchi-based GRA and the RSM-CCD. Meanwhile the L32 OA is utilized in the 2k-factorial method where 1/8 fraction design of experiment is used with each requires 64, 32, 16 and eight experiment runs respectively. This has made the L32 the nearest available in the 2kfactorial method to the L25 used in other statistical methods. The performance of the device is analyzed through the ION values along with ION/IOFF ratio amongst the statistical methods used. The combination of Taguchi method and GRA is introduced to overcome the limitation of a standalone Taguchi method that can solve only a single response at a time into multi-response optimization for the 16 nm gate length of double-gate FinFET. The Taguchi-based GRA showcases the best improvements with 47.79% for the ION/IOFF ratio as opposed to 45.39%, 20.54% and 23.01% for the Taguchi method, 2k-factorial and RSMCCD respectively, with ION at 1656.27 ȝA/ȝm for the TagXchi based GRA. Meanwhile, the IOFF, SS and ION/IOFF ratio optimized at 34.498 pA/ȝm, 96.743 mV/decade, and 48.0113 M, respectively. The process parameters of 16 nm gate length double gate FinFETs were successfully optimized by using the L25 OA of Taguchi based GRA. Following that, a nominal VTH, a high ION and a low IOFF characteristics were all attained. These proved that the multi-response characteristics of the device can be optimized simultaneously through the implementation of the L25 OA of Taguchi based GRA. That said, the VTH, ION and IOFF value for both devices meet the International Technology Roadmap Semiconductor (ITRS) 2013 prediction for high performance and low power logic multi-gate technology. |
---|