Impact of Work Function Variations on Data Stability of FinFET based 6T SRAM Cell

Mitesh Limachia; Rajesh Thakker; Nikhil Kothari

Abstract

The performance of FinFET SRAM cell is severely affected due to metal gate work function variations (WFV) in FinFET device. In order to provide guidelines for SRAM cell optimization, it is essential to examine the data stability due to the contribution of each cell transistor from WFV point of view. In this work, we investigate the effect of VTH variations of each transistor of tri-gated FinFET 6T SRAM cell on stability parameters (WSNM and RSNM) in 20 nm technology at VDD 0.9 V to examine the overall data stability under the influence of each transistor’s WFV, which has a strong relation with number of fins. Simulation results reveal that VTH and subsequently, VTH variations of PD transistors and PG transistors have severe impact on σWSNM, which is minimized by using more number of fins in PD and PG transistors. This, in turn, helps to secure write operation stability and reduces number of write failures. The strength of correlation between the VTH of cell transistor and RSNM indicate that PDL and PUL variations correlate to SNML, and the PGL and PDL variations correlate to the SNMR. The trade-off is observed between VTH of PDL and RSNM.